HOW TO USE THIS CD-ROM

Congratulations on purchasing a self-study course from ASM International. The reading

material on this CD is packaged as a PDF for your use. This information is best viewed
by using Adobe Reader 9 or later version. Below are instructions on how to navigate the
material on this CD, print material, and access the online testing system.

If you should have any questions regarding this product, please contact ASM
International at onlinetraining@asminternational.org.

Navigating the CD:

This CD can be easily navigated using bookmarks.

1. Click the BOOKMARKS tab on the pane on the upper left of your screen. This
will bring up the navigation sidebar with hyperlinks to chapters, quizzes and tests.
2. Start the course by clicking on the TABLE OF CONTENTS link.
3. Simply click on any of the chapter or test/quiz links on the navigation pane to go
to the desired section.
4. Close or open the bookmark pane at any time.

Printing the CD:

If you wish to print the material, print the entire file, one module, or one quiz at a time.

1. Click FILE and choose PRINT from the drop-down menu. Either print all pages
or select any print range.

Online Testing:
In this file, you can print the quizzes provided at the end of each chapter. Taking tests
online is optional; however, in order to earn CEUs you must take the test using ASM’s
online testing system and pass the test with a minimum score of 70%.

Please click the ONLINE TESTING tab on the bookmark pane to read detailed
instructions about the ASM online testing system.

Please visit the ASM website for self-study topics and all of our education products:
 ASM Member ID: _______________

Taking Your Tests Online

1. After you have had a chance to review your material, begin reading the first lesson. When you are satisfied
that you understand the material thoroughly, take the open book unit exam (located at the end of each
lesson). We suggest you fill in your answers in the printed test and then submit your answers online. The
questions and selections in the online testing system are in the same order as the printed exam.

If you need assistance on specific test questions, you may telephone (440) 338-5151 or write the ASM
International Education Department, 9639 Kinsman Road, Materials Park, OH 44073-0002 USA. You
have one year from your registration date to complete the course.

Log into http://www.asminternational.org/onlinetesting You will be prompted for a Login Name and
Password. Your 6-digit ASM ID NUMBER is your Login Name and your LAST NAME is your Password.
The Password is case sensitive so it is necessary to capitalize the first letter of your last name in
order for the system to accept it. (Due to system limitations if your last name is fewer than 4 letters or
more than 10 letters you will use you ASM ID as both your login name and password.)

This site is available at any time, from any location.

2. From the Home page, click on the Enter button to access the course you are enrolled in. Please read the
test information page before continuing to your tests. The most important piece of information is that you
must select your answer and hit the Submit button for the system to accept your answer. After
selecting Click Here to Enter Assessment you will be in the first question. You must answer the question or
it will be marked wrong. You’ll be shown the feedback for your answer immediately. Click Next Question to
move on.

3. You can leave a test at any time to finish it later. The system will remember where you left off. Just be sure
that you’ve selected Submit after answering a question, or the system will count the answer as
wrong.

4. Tests are graded on a scale of 0-100%. The average of the unit tests is 2/3rds of your overall final grade,
and the final exam is 1/3 of your overall final grade. The overall final grade must be 70% or higher to
successfully complete the course and be awarded Continuing Education Units (CEUs).

5. Upon satisfactory completion of the course, you will be able to print a CERTIFICATE OF COMPLETION,
which notes the completed course, your name, and the date you finished the course. To print a certificate
from the online system, select Completed Enrollments, your course and then click the Certificate button.

About a week after you complete your course, an official certificate with Continuing Education Units (CEU’s)
earned will be automatically mailed to you.

6. ASM Education Department strongly recommends that you complete your testing within a period of one year
from receipt of the course materials.

If you have any questions about taking your tests online, email onlinetraining@asminternational.org.
 LESSON NUMBER AND TITLE

INTRODUCTION TO HEAT TREATING

1. General Aspects of Heat Treatment

2. Steel and Its Mechanical Properties
3. Microstructure and Mechanical Properties
4. Austenite and Its Transformation and Classification of Steels
5. Annealing and Normalizing Steel
6. Hardening of Steel
7. Tempering of Steel
8. Heat Treatment of Tool Steels
9. Constitution and Heat Treatment of Stainless Steels

(Lessons from Principles of Heat Treating – Course 41)

ASM International
9639 Kinsman Road
Materials Park, Ohio 44073-0002
Phone: 440.338.5151 or 800.226.5152
Fax: 440.338.4634
 The Materials
Information Society

General Aspects of
Heat Treatment
Alesson from PRINCIPLES OF HEAT TREATING

by
Paul G. Nelson, B.S.

Course 41
Lesson, Test 1
 Metric Conversion Factors
To convert from To Multiply by

in. mm 25.4
in. m 25.4 x 10-3
mil J..lI11 25.4
)Jin. J..lI11 25.4
in.2 m2 6.45 x 10"'"
in.3 m3 1.64 x 100s
ft. m 3.048 x 10-1
ft.2 m2 9.29 x 10-2
ft. 3 m3 2.831 x 10-2
oz. g 2.834 x 101
lb. kg 4.536 x 10-1
Btu J 1.054 x 1()3
Btullb. - OF Jlkg· K 4.18 x 103
Btulft. - hr • OF W/m-K 1.730
10 inJin.f'F 1()4S mm/111I11f'C 1.8
psi Pa 6.895 x 1()3
psi kPa 6.895
ksi kPa 6.895 x 1()3
ksi MPa 6.895
ksi - in. 11l MPa-mlll 1.099
ksi~in. MPa--.Jm 1.099
oz.f gf 28.4
lbf kgf 4.536 x 10-1
lbf N 4.448
lbf- ft. N - m (or J) 1.356
Ibflin. 2 kgf/cm2 14.223
Ibflin. 3 kgf/m3 2.768 x 10'
IbJft. 3 kglm3 16.019
Ib./in. 3 g/cm3 2.768 x 101
Ib./in. 3 kglm3 2.768 x 10'
gal (U.S. liquid) L 3.785
gal (U.S. liquid) m3 3.785 x 10-3
IbJgal gIL 119.826
ft/gal mlIL 748
OF °C (OF-32)/1.8
OF K (OF + 459.67)11.8
°C OF COC· 1.8) + 32
°C K °C + 273.15
K °C °C -273.15

Multiple and Submultiple Units

1018 •••••••••••••••••••••••••••• exa 10-1 deci
101' peta 10-2 centi
10 12 ~ •••••••••••••••••••••••••• tera 10-3 •••••••••••••••••••••••••• milli
1(JJ •••••••••••••••••••••••••••• giga . 1()-6 •••••••••••••••••••••••• micro
1()6 •••••••••••••••••••••••••• mega 1Q-9 nano
1()3 •••••••••••••••••••••••••••• kilo 10-12 •••••••••••••••••••••••••• pico
1(Jl •••••••••••••••••••••••••• hecto 10-1s ••••••••••••••••••••••• femto
101 deka 10-18 ano

Abbreviations
J ..........•..•....••.....]oule m meter Pa pascal
kgf ....•. kilogram force mm millimeter K kelvin
L liter N newton W watt
 PRINCIPLES OF HEAT TREATING
Lesson 1
General Aspects of Heat Treatment
by
Paul G. Nelson, ~.S.
Generally speaking, heat treatment
may be defined as the " controlled heat-
ing and/or cooling which is intended to
improve certain properties of metals.'9
There are numerous forms of heat
treatment, each having a different term;
some of these are quenching, temper-
ing, normalizing, martempering, full
annealing, and so on.
Some of these treatments will
harden and strengthen metals, whereas
others will soften and affect other me-
chanical and physical properties such
as impact strength, ductility, magnetic
susceptibility, toughness, machinability,
and numerous other properties. These
terms and many others will be dis-
cussed in this course.
HISTORICAL
As with many other fields, the art of
heat treating metals preceded the sci-
ence. For many centuries the art of
heat treatment was practiced by the an-
cient metalsmith. There is no scientific
evidence that the metal workers of the
bronze age hardened their bronze tools
by heat treatment. However, there is no
doubt that they used annealing to soften
various articles which were being cold
hammered from raw metal.
Iron was mown for many centuries
before the Christian era. In fact, iron
daggers 3000 years old have been found
This material may not be reproduced in whole or in part in any form whatsoever.
Fully protected by copyright.
@ 1969 by American Society for Metals
in Egyptian pyramids. The addition of
carbon to iron and its hardening effects
were later developments, but were
mown. before the Christian era.
According to an ancient Jewish leg-
end (depicted in Fig. 1: 1), when the
temple at Jerusalem was completed,
King Solomon gave a feast for the work-
ers employed in the construction. The
most important craftsman was to be
chosen and allowed to sit at the king's
right. When the throne was unveiled, it
was found that a metalsmith had usurped
the seat of honor, not yet awarded. The
people clamored angrily and the guards
rushed to cut him down. " Wait, let him
speak!~' commanded Solomon. "Thou
hast, 0 King, invited all craftsmen but
me. Yet how could these builders have
raised the temple without the tools
which I fashioned?~~ "True!" decreed
Solomon. "The seat is rightly his. All
honor to the metalsmith!~'
The metalsmith, who was later re-
ferred to as the blacksmith, was the
heat treater for many centuries and in
some plants still does the heat treating.
An art was developed for heating,
quenChing, and reheating (tempering)
carbon steel tools which produced steel
.and iron alloys of Widely varying hard-
ness and toughness for applications
which might require the hardness of a
razor, or the toughness and shock re-
sistance of a cold chisel.
2 PRINCIPLES OF HEAT TREATING
Fig. 1:1. The Metalsmith and King Solo-
mon.
This figure is from the original picture
painted in 1864 by C. SChussele for
Joseph Harrison, Jr., and in his collec-
tion in Philadelphia. From the engrav-
ing by John Sartain, Philadelphia, 1871.
REASONS FOR HEAT TREATING
Today, metals are still heat treated
for the same reasons as in ancient
days, namely, to improve their mechan-
ical properties. The common mechani-
cal prol;)erties are hardness, ductility,
strength, and toughness.
Hardness The hardness of a metal is
a very common term but
one not easily defined, except by de-
scribing the test and how metals be-
have. Therefore. the best definitions
state that hardness is a measure of the
resistance of a metal to scratching or
indentation. The tests will be described
in detail later.
Ductility Ductility, or ability to de-
form permanently prior to
fracture, is commonly measured by the
amount that a bar in tension will elon-
gate before fracturing. This elongation
is expressed as a percentage of the
original length of the bar.
*Normalizing and many other terms will be described in detail in later sections of this
course..
Strength Three types of strength are
commonly discussed. One is
the elastic strength, or strength before
the material deforms permanently. This
is referred to as the elastic limit. The
second is concerned with the strength
of metal before appreciable plastic de-
formation occurs. This is yield strength
and will be discussed in greater detail
in the next lesson. The third is the
maximum strength that a metal exhibits
during tensile deformation and is known
as the ultimate tensile strength, or
simply tensile strength.
Toughness Toughness is a less sci-
entific term, and is more
difficult to measure, but is a combina-
tion of strength and ductility. The basic
nature of these and other mechanical
properties will be further developed in
later lessons.
Variations in Properties
As stated before, metals are heat
treated to improve their mechanical
properties. In Table 1:1, comparative
strength properties of a plain carbon
steel (0.50% carbon) with various heat
treatments, but good ductility, are
shown.
It will be noted from these results
that maximum ductility and lowest
strength are obtained when this steel is
annealed. Normalizing,* another heat
treating process, increases strength
but lowers ductility somewhat. Hard-
ening this steel by water quenching, fol-
lowed by reheating or tempering at the
proper temperatures, greatly increases
strength as shown, but also lowers duc-
tility as compared to the annealed state.
For many applications the lowered duc-
tility is still adequate and the higher
strength steel permits reductions in
LESSON 1 3

Table 1:1. Effects of Various Heat Treatments on Other alloys such as those of mag-
Mechanical Properties of Ductile 1050 Steel
nesium, nickel, titanium, copper, mo-
Yield Ultimate lybdenum, etc. can also be strengthened
Condition Strength Tensile Elongation by heat treating the precil)itation-hard-
(psi) Strex:gth (%)
(pSI) ening type alloy compositions. The de-
Annealed 50,000 90,000 27
velopment of these alloys and their heat
(Heat to 1450° F treatment have made possible our mod-
and Furnace Cool) ern aircraft and other achievements.
Normalized 60,000 105,000 24 'Prior to the development of high
(Heat to 1650°F,
Air Cool)
strength metals, aircraft were con-
structed mainly of wood. Aluminum al-
Hardened and Tempered 95,000 125,000 24
(Heat to 1525 of, loys and their heat treatment made
Water Quench, modern metal aircraft feasible and
Reheat to 1000 0 F for
1 Hr., Air Cool)
practical, and were a prime factor in
the enormous develo-pment of the pres-
ent-day aircraft industry. However,
weight. In addition, the high-strength presenf-day applications of heat treat-
quenched-and-tempered steel frequently ment are just as important in our mod-
has greater toughness or resistance to ern automobiles, household appliances,
shock. automatic machinery, sewing machines,
typewriters, and missiles.
Alloys

Alloys are essentially mixtures of Heat Treating Effects

various individual metals, or sometimes
Im~rovements in
hardness, strength,
it may be a metal and non-metal. This
results in strengthening. The metal and other mechanical properties of
which is present in the greatest per- metals which are effected by heat treat-
centage is usually considered the ment are of prime interest to design-
"base" or " solvent/' ers. However, improvements in fabric-
ability which can also be obtained
Certain alloys can be further through proper heat treatment are of
strengthened by a heat treating process almost equal importance. These im-
known. as precipitation hardening or age provements are obtained primarily
hardening. This heat treatment which from heat treating processes which are
consists of a quench (usually in water) known as annealing.
from an elevated teml)erature, followed
by aging (holding the metal for a period
Table 1:2. Effects of Heat Treatment on
of time at some intermediate tempera- Mechanical Properties of Aluminum Alloy, 6061
ture) greatly increases strength. Table
1:2 shows the improvement in strength Yield Tensile
Condition Strength Strength Elongation
which can be obtained by this type of (psi) (psi) (%)
heat treatment on a commercial alu- Annealed 775°F 8,000 18,000 30
minum alloy. o
(Cool 50 /hr. to
500 ° F, then air
cool.)
These values also show maximum
Solution Treated 22,000 36 000 26
ductility and lowest strength with an- (Heat to 970°F
I

nealed material. Proper heat treating Water Quench)

can increase strength while still re- Solution Treated 41,000 46,000 16
taining sufficient ductility for most ap- as above and Aged
(8 Hrs. at 350°F)
plications.
4 PRINCIPLES OF HEAT TREATING
Annealing Annealing is a controlled
heating and cooling proc-
ess which softens metal. Hot-rolled
low-carbon sheet steel. which is an im-
portant industry item, is subsequently
cold-rolled in literally millions of tons
per year because of the superior finish.
dimensional accuracy, and latent im-
proved formability of cold-rolled sheet.
However, the improved formability is
not obtained unti~ the cold-rolled sheet
is annealed.
This annealing nrocess which con-
sists of heating the steel to a specified
temperature for a controlled time,
sometimes in a special atmosphere,
softens the cold-rolled steel and greatly
improves its ductility. Because of the
availability of today's special non-ox-
idizing atmospheres, surface quality is
improved.
Low carbon steel and practically all
other types Of steel may also be cold-
drawn into bar and wire forms. Stain-
less steels, aluminum alloys, copper
alloys, and almost all metals may be
processed by cold-forming methods
wherever superior finish and dimen-
sional accuracy are required. All of
these cold-formed 'products are even-
tually annealed to soften them, if the
highest formability is required.
Table 1:3 illustrates the typical ef-
fects of annealing on properties of cold-
rolled low-carbon steel sheet. It will be
Table 1:3. Typical Mechanical Properties of
Cold-Reduced and Annealed Low Carbon Steel
Yield Tensile
Elongation
Condition Point Strength
(%)
(psi) (psi)
Cold-Reduced 60% 85,000 85,000 2 als. By varying the heat treatment it is
Cold-Reduced 60% 29,000 43,000 42
and Annealed
*Microstructure is the appearance of polished and etched metal surfaces as viewed under an
optical microscope at magnifications ranging from about 50X to 2000X. These usually show
grains and various metallic and nonmetallic phases.
noted that cold work increases the yield
and tensile strengths of low carbon
steel but reduces the ductility. as meas-
ured by elongation. Although elongation
is not an absolute measure of formabil-
ity, it is a suitable guide.
Steel with only 2% elongation would
have extremely poor cold-forming
properties. Annealing of this steel at
temperatures of 1250-1300°F for times
of 24 to 48 hours markedly reduces the
strength~ but greatly improves the duc-
tility. Material with these properties
can be readily formed into complex
shapes, such as automobile fenders,
washing machine tubs, etc.
A great variety of other metals are
produced in cold-rolled, cold-drawn, or
cold-extruded forms. Cold-rolled cop-
per and aluminum alloys are as com-
monplace as cold-drawn copper and
aluminum wires. Cold-extruded alumi-
num alloy structural sections are also
extensively used. In fact, almost all
metals are available in some cold-
worked form. With practically all of
these materials, annealing is performed
either as an intermediate or final oper-
ation. ~Tith some metals, such as stain-
less steels, multiple annealing may be
required to cold-roll sheets or strip to
final thickness, particularly if very thin
sheet is the final product.
Machinability The machinability of
many metals is greatly
affected by heat treatment. (Machinabil-
ity refers to the ease with which metals
can be cut.) This is particularly true of
steels and cast irons, but to a certain
extent is also true of nonferrous met-
possible to produce different micro-
structures. *
LESSON 1 5

Fig. 1:2 shows several possible mi-
crostructures and hardnesses which can
be obtained with a typical low alloy
steel by varying the heat treatment.
Machinability ratings and hardness val-
ues of each structure are li.steel.
All machinability ratings are given
in percentage comparisons with Besse-
mer screw stock which is considered to
be about the best in machinability for
steel and other iron alloys. A machin-
ability rating of 65% would indicate that
the metal could be machined about 65%
as fast as Bessemer steel. It should be
noted that the quenched and tempered
structures have the poorest machinabil-
ity of the structures shown, the pearlitic
structures are intermediate, and the
spheroidized structures are best.
Dimensional In addition to the use of
Stability heat treatment for im-
provement of machina-
bility, heat treatment is also used as an
intermediate step in machining to im-
prove dimensional accuracy_ Because
quenching, machining, and previous fab-
rication can induce internal stresses in
complex parts, a heat treating operation
known as stress relief is used to re-
lieve them. stress relief may also be
used prior to final machining to insure
greater dimensional accuracy on parts
that are to be subsequently heat treated
either for high hardness or other spe-
cial mechanical properties. This heat
treating operation is also widely used
to relieve internal stresses in weld-
ments and castings and thus reduce the.
tendency to crack easily. The strength
of weldments and castings is often re-
duced because of internal stresses that
they .contain. Their elimination is
therefore important.
Corrosion stress relief is also
Widely used to prevent
failure by a corrosion phenomenon
known as stress corrosion. Many metal
parts which have been severely cold
worked may suddenly fail by this proc-
ess. This is particularly troublesome
with brass and austenitic stainless steel

a. Armealed Microstructure b. Spheroidized Microstruc- c. Quenched and Tempered to

(90% Pearlite 10% Ferrite)
Brinell Hardness 195
Machinability Rating 36%
ture
(Spheroidized Cementite
and Ferrite)
Brinell Hardness 170
Machinability Rating 56%
300 BHN
(Tempered martensite fine
spheroidite)
Brinell Hardness 300
Machinability Rating 25%

Fig. 1:2. Comparison of Various Microstructures at lOOOX to Hardness and Machinability of

an AISI 4140 Low Alloy Steel.
6 PRINCIPLES OF HEAT TREATING

in certain environments. Stress relief Table 1:4. Comparative H.oom Temperature

Mechanical Properties of Low Alloy and
by suitable heat treatment will prevent PIa in Carbon Steels
this type of failure. Yield Tensile Izod
Elongation
Condition- Strength Strength Impact
(psi) (psi) (9rJ (ft - Ib)
Heat treatment may also be used to 1040 Carbon Steel 67.000 95,000 26 68
prevent other corrosion failures. Some Oil Quenched 1575 F
0
0

of the austenitic stainless steels are
subject to serious chemical attack in
certain media because of carbide pre-
cipitation in grain boundary areas and
subsequent int~rgranular corrosion.
This is illustrated in Fig. 1:3 for an
18% Cr-8% Ni type alloy. Solution-
quenching.. a heat treatment which
places chromium carbides back in solu-
tion~ results in a restoration of good
corrosion resistance.
Alloy Selection
Frequently in choosing a heat treat-
ment. a compromise must be made be-
tween certain properties. As-quenched
steels usually are very hard and strong
but lack ductility and toughness. For-
tunately by proper selection from var-
ious alloys and heat treatments, it is
possible to obtain a wide range of prop-
erties. Table 1:4 shows the comparative
strength and ductility properties of a
plai~ carbon steel and an alloy steel af-
ter different heat treatments.
Tempered at 1000 F
4140 Low Alloy Steel 122,500 132,700 26 85
0
Oil Quenched 1550 F
Tempered at 1200 0 F
·Both samples were 1 in. round.
3. Increasing solid solubility of one
metallic or nonmetallic ele-
ment in another with in-
creases in temperature.
All changes in mechanical proper-
ties of metals which can be induced by
heat treatment result from some change
in microstructure or internal crystal-
line structure. All metals and their al-
loys are crystalline, i.e., they usually
consist of an aggregate of a large num-
ber of tiny crystals (grains). Under
special conditions large single crystals
of metals can be produced. In each sep-
arate crystal, its atoms are regularly
arranged in a distinctive geometric pat-
tern with fixed angles and distances be-
tween atoms. Typical crystalline ar-
r-angements or lattices will be dis-
cussed in later lessons.

It will be noted from these results
that in this case the alloy steel is
superior to the plain carbon steel in
strength and toughness (measured by
the Izod test). Ductility values are equal.
REQUISITES FOR HEAT TREATMENT
Mechanical properties of metals and
their alloys can be changed by heat
treatment prOVided certain conditions
are met. Principal conditions which can
permit pronol.mced changes in mechani-
cal properties by heat treatment are:
Cold Work
Cold working metals and their alloys
changes their mechanical properties.
Cold work distorts the regular crystal-
line structure with resulting increase in
strength and reduction in ductility. This
is illustrated in Table 1: 5 below.
Table 1:5. Effects of Cold Work on Some
Mechanical Properties of Commercially
Pure Aluminum
Cold Work Yield Tensile Elongation
(% Reduction of Area) Strength Strength (~ )
(psi) (psi) 0

1. Extensive prior cold work. None 5,000 13,000 40

20 13.000 15,000 18
2. Changes in crystalline structure 40 14.000 17,000 15
of a metal with changes in 60 17,000 20,000 12
temperature. 80 21,000 24,000 10
LESSON 1 7

M
I
C
r
o
s
t
r
u
c
t
u
r -
e
I

r-----------------------
t t
93
Chromium Depletion
Adjacent To Precipitated
Chromium Carbides
Chromium Carbide
Cr4C
20 Initial Uniform Chromium Content
/

J i ! ! !i !i !' i! i! i! ! ! ! ! ! i i! !i ·il!i !I I ~ i: .
(,)
I5

Fig. 1:3. UPPER: Diagram illustrating microstructure of 18% Cr-8% Ni stainless steel;
white area is austenite of 18-8 composition, white particles in grain boundaries are
chromium carbides, shaded areas are chromium-depleted areas.

LOWER: Differences in chromium content caused by carbide precipitation; cross-

hatched area (low chromium) can be severely corroded.

Stress Relief When cold - worked This type of heat treatment, which re-
metals are heated to a sults primarily in lower internal stress
low, but sufficiently high temperature, in the metal, is commonly termed
some degree of rearrangement of the "stress relief."
grain structure occurs which results in
at least a partial restoration of original Recrystallization With appropriate
properties i.e., lower strength and in- amounts of cold
creased ductility. However, no visual work, reheating to a temperature which
microstructural change can be seen. is higher than the stress relief temper-
8 PRINCIPLES OF HEAT TREATING

ature, results in the formation of en-
tirely new crystals or grains. The low-
est temperature which produces this
change is known as the recrystallization
temperature.. It is interesting to note
that the recrystallization temperatures
of the pure metals bear a relationship
to their melting points. This relation-
ship, shown in Fig. 1:4, turns out to be
close to the follOWing formula:
Initially the grain size immediately
after recrystallization is fine, but the
grain size increases if the metal is held
for longer periods of time, at or above
the recrystallization temperature. By
varying the amount of cold work, an-
nealing temperature, and time at tem-
perature, it is 'P9ssible to produce a
wide range of mechanical properties
with many common metals.

Re =- 0.4 Mp Crystal Some metals crystallize

where Re is the recrystallization tem-
perature on the absolute scale (degrees
Rankine or Kelvin)* and Mp is the melt-
ing point on the same temperature
scale.. This can serve as a basis for es-
timating the recrystallization tempera-
ture of a new metal.
Structure from the molten state in
one crystalline or lattice
arrangement, but transform spontane-
ously at lower temperatures to some
other lattice. These changes are usually
reverSible, i.e., on being reheated again
to higher temperatures the lattice will
revert to its original form . Such trans-

Fig. 1:4. The Relationship Between Melting 'Points and Recrystallization Temperatures.
As melting temperatures increase, the recrystallization temperatures increase in a
near-linear fashion.
0
*To get degrees Rankine, simply add 460 to 0 Fahrenheit; to get degrees KelVin, simply add
0
273 to Centigrade (also called CelsiUS).
0
LESSON 1
formations from one crystalline form to
another are known as allotropic
changes. Iron, for example, has two al-
lotropic forms, namely ~ the body-cen-
tered cubic (bee) and face-centered cu-
bic (fcc). Titanium also nas two lat-
tices, body-centered cubic and hexag-
onal close-packed. Plutonium, a fis-
sionable element which has been used
in the newer atomic bombs, has five al-
lotr90ic form_s. On the other hand, many
common metallic elements such as
nickel, copper and zinc have only one
crystalline lattice, and undergo no
changes.
Alloy An alloy is a metal composed
of two or more elements, one
of which must be a metal. In alloys. the
extent of solid solubility of any element
in a given metal will be affected if the
metal undergoes an allotropic change.
For instance, 2% carbon is the maxi-
mum solubility of carbon in fcc iron,
whereas only 0.025% carbon is the max-
imum solubility in bcc iron.
Solid solubility of one metal in an-
other may also vary consider~bly with
changes in temperature. The phenome-
non of solid solution will be discussed
in detail in subsequent lessons, but the
following analogy may help in under-
standing crystals. In an alloy composed
of two metals in which there is essen-
tially no solid solution, each metal
would have a distinctive crystalline ap-
pearance, analogous to a mixture of red
and green colored salts. The iron-lead
system is typical of such behavior.
In an alloy of two or more metals in
which there is complete solid solution,
only one type of crystal will be ob-
served, since each will have mixed
completely with the other. The copper-
nickel system typifies complete solid
solubility.
The melting temperatures, changes
in extent of solid solubility, and pos-
9
sible occurrence of compounds with
changes in temperature and composition
are usually portrayed as phase dia-
grams such as Figs. 1:5, 1:6, and 1:7.
Fig. 1:5 shows an equilibrium diagram
for copner-nickel alloys, whereas Fig.
1: 6 is the most important part of the
phase diagram for iron-carbon alloys.
Fig. 1: 7 is a portion of the phase dia-
gram of the aluminum-copper alloy
system.
The diagram for the copper-nickel
alloys (both metals are face-centered
cubic) shows that neither metal under-
goes any crystal change. All possible
allOYS, regardless of temperature, are
face-centered cubic (if below the melt-
ing point). Also, there is 100% solubility
for each metal in the other at all tem-
peratures. Alpha (a) solid solution
(Greek letters are generally used to in-
dicate solid phases) exists at all tem-
peratures and for all compositions of
solid allays.
Fig. 1:6 shows that the solid solu-
bility of carbon is extremely small in
body-centered cubic iron (also mown as
ferrite), but it is much more soluble in
face-centered cubic iron (also known as
austenite).
M.P OF
PURE NICKE
2600
LL.
~ 2400F";~,,:~::-:oO:·.~:"';~~:-::~~~~
LLJ
a::
::;)
!;i 2200
a::
LLJ
Cl.
~
MP OF PURE
COPPER
~ 200°t::::;::»~::~:::~:;:>~<::::<:~<:i:;;i:~:;:>
o 20 40 60 80
100 80 60 40 20
COMPOSITION
Fig.. 1:5. The Nickel-Copper Phase EqUilib-
rium Diagram
Complete solid solution is possible be-
cause both atoms are approximately the
same size and their crystal structures
are both face-centered cubic.
 10 PRINCIPLES OF HEAT TREATING

2000 iron did not have the bee-fcc crystal

change . which makes possible tremen-
1800 r
Austenite
dous increases in strength, civilization

,{: ! I I:!~i~ ~! 1
would not be at its present stage of ad-
l..l...
vancement. Skyscrapers would be a
0
I
maximum of about one-quarter their
Q)
~ 1400 . present size, machinery would wear out
-0
~
::3

1200
a Ferrite
Q.)
a.
E
Q)
r- 1000
800 a+Cm
600
300 a>
d
a I
a 05
Per Cent Carbon
Fig. 1:6. Most Important 'Portion of the
Iron-Cementite Phase Diagram.
Iron dissolves about 2% carbon when it
is heated into the austenite region, but
much less than 0.1 %when cooled below
the A 1 temperature.
The above diagrams, when well un-
derstood, give an indication of possible
heat treating cycles, or suggest tem-
peratures from which quenching may
have important effects. These matters
will be developed in detail in later sec-
tions.
For the present, let us only conclude
that the simple phase diagrams of the
copper-nickel type, where there are no
solid-phase variations in solubility or
crystal structure changes, suggest that
no hardening heat treatments are pos-
sible.
However, when crystal structure
changes occur, as portrayed by the
austenite-ferrite regions in Fig. 1:6,
there are possibilities of controlled
heat treatments with profound effects.
This is indeed the case for steel (alloys
of iron and 0.1% to about 2% carbon). If
in much less time~ automobiles would
be impractical or much too expensive,
clocks would be less accurate, there
would be no battleships or machine
guns, and who can say-we might have
been better off without some of these!
Nevertheless, we have it, and we should
learn to make the best use of it.
Phase diagrams such as the one
1.0 1.6 shown in Fig. 1: 7, which portrays a re-
duction in solubility as a function of
temperature (here copper is shown to
val)" in solubility from 5.65% at l018°F
to 0.1% at 100°F), suggest the pos-
sibility of precipitation hardening. This
is a hardening process in which proper
heat treatment develops a very fine
dispersion of precipitated particles
throughout the metal, so that hardening
and strengthening result. This harden-
ing technique, surprisingly enough, is
the newest of those discussed so far. It
was not known in preceding centuries
and was first described in 1911 by a
German metallurgistl' Dr. Alfred Wilm,
who found unexplainable variations in
some aluminum-copper alloys that he
was studying. It wasn't until about 1919
that the age-hardening phenomenon was
understood in principle and discussed
intelligently.
Briefly, what happens is this. Cop-
per, when added to aluminum, has lim-
ited solid solubility 'When this amount
of copper is exceeded, the copper forms
an intermetallic compound CuAl 2 (theta,
I) with aluminum. Raising the tempera-
ture, however, may increase the amount
of copper that can be dissolved. This is
illustrated by the curved line A in Fig.
1:7.
LESSON 1 11

\\'hen an alloy of approximately 4.5% formation of compound within the alloy)

copper (dotted line in Fig. 1:7) is heated
to lOOQoF and held for a sufficient time,
all the copper goes into solution. If this
alloy is then cooled slowly, the excess
copper will again be precipitated from
solid solution in the form of the inter-
metallic compoWld, CuAl 2 , the theta
phase.
greatly increases strength.
Reactions of this type are known as
aging reactions because the net change
in mechanical properties is dependent
upon "aging" time at room, or slightly
higher temperatures which are com-
mercially used.

TYPES OF HEAT TREAT:MENT

Cooling such an alloy rapidly to
room temperature, however, will retain
all the co-pper in solid solution. Such
solid solution is supersaturated, or un-
stable (commonly called metastable,
meaning partly stable). Given sufficient
time at room temperature, it slowly
transforms. This transformation (With
Since heat treatment has been de-
fined as an operation or combination of
operations involving the controlled
heating and cooling of a metal in the
solid· state for the purpose of obtaining
desirable conditions or properties, let
us review the terms in common use.

1200i1;~t::::~: 1220°F -

~:·:·:·:·:·:·.I ~+L
:.:.:.:.:':.:':.:.:'.. 0
53.5%
::::::'::::::::::::::1::.. 1018 F
1000 ~:::.1\ ::. 565 '0
L( :.:.:. 0J
52.5%'
-
!
.

~Ilt~::::-· -
800~~;:~A
::::::
I
I ~:' CuAlt -
::. I CuAI2/' ::."+ _
t:::: I
I (e phase)
600 ~ 14 .5 % Cu -
,I
-
400 -
j -
l
200 -
-
:J
O..~:f_--...II",--_ ....I_ _... I _ _~,_.....rl...-._...I.I_ _....L.I_ _.L-I_ _IL....-_"""_ _..J....-_~

o 10 20 30 40 50 60
Copper- %
Fig. 1:7. Portion of the Aluminum-Copper Phase Diagram.
An alloy of 4.5% eu is single phase and homogeneous at 10OQ°F, but is likely to be two-
phased (kappa plus particles of CUA!2) at room temperature..
12 PRINCIPLES OF HEAT TREATING

There are many types of heat treat- Annealing Procedures

ing processes; some are intended to
apoly only for steels, others should be AnnealiIlg has frequently been de-
applied to certain kinds of alloys. Prin- fined as a heat treating process which
cipal classifications and their subdivi- is used primarily to soften. However,
sions are listed below: common usage has extended the term to
other related heat treating processes
whose primary purposes are such as
1. ANNEALING (Softening) PRO-
homogenization of composition, refine-
CEDURES ment of grain size~ elimination of gas,
a. Homogenization Anneal solution treatment for age hardening, or
b. Full .A.rmeal solution treatment for restoration of
c. Spheroidizing Anneal (Critical corrosion resistance.
Range Anneal. or Sub-Criti-
cal Anneal) Homogenization Annealing This is a
d. Isothermal Anneal process in
e. Normalizing (for ferrous al- which an alloy is heated almost to the
loys only) melting temperature for long periods of
f. Stress-Relief Anneal time. The purpose of the process is to
g. Recrystallization Anneal homogenize, that is, to permit more
h. Solution Anneal (Quench-",\!1- uniform diffusion of various alloying
neal) elements which tend to segregate during
i. Temper (draw) solidification. It is widely employed in
2. HARDENING PROC EDURES operations like the soaking pit homoge-
a. Quench Hardening nization of steel ingots prior to bloom-
b. Time-Quench and Temper ing mill rolling.
c. Marquenching (Martempering)
d. Austempering Because this heat treating is done at
e. Precipitation Hardening (Age at such high temperatures, grain growth
Hardening) is very likely to occur. However, in the
f. Small Particle Hardening case of ingots~ subsequent hot work
g. Martensite Transformation such as forging or hot rolling refines or
Plus Age Hardening reduces the grain size, thus counteract-
h. Order-Disorder Reactions ing the harmful grain growth effects of
i. Ausforming (Deformation plus the homogenization heat treatment.
Martensite Transformation) Homogenization is frequently used on
all types of castings. Because castings
3. SURFACE HARDENING
are not subsequently hot-worked, grain
a. Induction Hardening
refining heat treatments such as nor-
b. Flame Hardening
malizing are used to reduce the grain
c. Carburizing
size of ferrous alloys. The extreme im-
d. Nitriding
portance of this heat treatment is read-
e. Cyaniding
ily realized, when it is known that more
f. Carbonitriding
than 100 million tons of ingots are
g. Other surface diffusion treat-
rolled by the nation's steel mills each
ments (Siliconizing, Chrom-
year. Practically all of this steel is
izing, etc.)
homogenized in soaking pits prior to
4. MISCELLANEOUS HEAT TREAT- rolling.
MENTS
a. Patenting Full .A.IUlealing Steel is heated above
b. Stiffening Temper its upper critical (A 3)
c. Malleablizing temperature and then cooled very
LESSON 1
slowly in the furnace to achieve full an-
nealing. With plain carbon steels con-
taining up to 0.60% carbon, this treat-
ment results in a microstructure which
is readily machinable. Full annealing,
which results in a coarse pearlite
structure, is also widely used on alloy
steels up to 0.40% c.
Spheroidization Higher carbon alloy
- and higher carbon
steels are frequently given a spheroi-
dizing anneal because of the superior
machining properties of spheroidized
structures. Full annealing is widely
employed for softening carbon and alloy
constructional steel bars, plates, struc-
tural sections, tubing and forgings.
Higher carbon alloy steels (over
0.40% carbon) as well as various alloy
and plain carbon tool steels are usually
spheroidized for best machinability.
In spheroidizing, the steel is heated
to just below the lower critical temper-
ature (A 1) and either held there for an
extended period of time or cycled just
above and below this temperature. This
is sometimes referred to as a sub-
critical anneal or critical range anneal.
The structure which results is a mix-
ture of ferrite (low carbon C1 iron) and
iron carbide in spheroidal particles.
The spheroidized carbide structure is
more easily machined than the pearlitic
structure which result from full anneal-
ing. As stated before, this type of heat
treatment is Widely used in annealing
high carbon alloy and tool steels.
Isothermal Annealing of steels and
cast irons is
achieved by heating above the upper
critical temperature, then cooling mod-
erately rapidly (air-cooling for ex-
ample) to an intermediate temperature
of about 1200° F. It is then held at this
temperature for an extended period of
time until transformation is complete
13
after which it may be cooled moderately
again (air cooling).
The final structures which are ob-
tained by isothermal annealing "are ap-
proximately the same as those which
are produced in full annealing, but they
are obtained in much less time- Iso-
thermal annealing in many cases will
produce in 4 to 8 hours results which
require as much as 30 hours in full an-
nealing. Because of the time savings,
isothermal annealing is being more .
Widely applied for annealing of tool and
alloy steels.
Normalizing In a normalizing heat
treatment, a steel or
cast iron is heated above the upper
critical temperature and air cooled to
room temperature. The primary pur-
pose of normalizing is to reduce the
grain size. However, it may also have
the effect of softening some low carbon
and low alloy steels. but with higher
carbon and higher alloy steels normal-
izing may actually become a hardening
treatment. In this latter instance, tem-
pering of the normalized metal is com-
mon practice.
NormaliZing is Widely used to re-
duce grain size of bars, forgings, and
castings. Large alloy steel forgings are
commonly hardened by normaliZing plus
tempering. Low carbon hot-rolled
sheets are frequently normalized when
maximum formability is required Nor-
maliZing should not be confused with
stress relief annealing even though
some stress relief is obtained during
normaliz ing .
Stress Relief Anneal To reduce in-
ternal stresses,
a metal or alloy is heated and held at a
temperature where the yield strength of
the material is greatly reduced. inter-
nal stresses are then relaxed to this
lower value of yield strength.
14
With iron and steel alloys, stress
relief is done at a temperature below
the lower critical temperature. The us-
ual temperature range is about 1000 to
1250°F. Fairly complete stress relief
of carbon and low alloy steels is ob-
tained at temperatures slightly above
0
1200 F . For best stress relief effects,
cooling rates for iron and steel are
usually relatively slow, such as air or
furnace cooling_.
Nonferrous alloys such as copper
alloys, nickel alloys, and aluminum
alloys may also be stress relieved .
Temperatures may be lower or higher
than for iron alloys, depending on the
alloy under consideration. Cooling rates
are similar. In certain instances with
nonferrous alloys it may be difficult to
properly stress relieve them without
also lowering their strength.. Similarly,
stress relief of hardened steels must
be done- below the tempering tempera-
ture if loss of hardness and strength is
to be avoided.
Large numbers of weldments, par-
ticularly pressure vessels, are stress
relieved to eliminate or reduce welding
stresses. HeaVily cold worked parts are
also frequently stress relieved.. Such
stress relief increases the usable
strength of such materials by reducing
adverse internal stresses. stress relief
is also used to eliminate stress corro-
sion cracking of such alloys as brass
which are extremely susceptible to this
type of failure in the heavily cold
worked condition.
It has been reported that some
stress relief can also be achieved by
ultrasonic treatment.
Recrystallization Cold worked met-
Annealing als such as low
carbon steel strip,
aluminum alloys, copper alloys, nickel
alloys, and many others are annealed
by being heated above the recrystal-
PRINCIPLES OF HEAT TREATING
lization femperature. As the term im-
plies, the deformed grain structure is
replaced by new recrystallized strain-
free grains. Cooling may be slow such
as is obtained by box annealing of low
carbon steel.. or moderately rapid as
with air or water cooling of copper!,
aluminum, or some stainless steels. Ex-
tremely large quantities of materials
are heat treated in this way.. Millions of
tons of tin plate and millions of tons of
cold rolled automobile steel sheets are
processed in this way each year. Prac- .
tically all nonferrous alloys (such as
copper, aluminum, and nickel) in sheet,
strip, and tubing forms are also proc-
essed using this technique.
Solution Annealing is sometimes re-
ferred to as
"Quench Armealing." It involves heat-
ing to a sufficiently high temperature
for a sufficiently long time, and is fol-
lowed by rapid cooling (usually quench-
ing). This may be applied to certain
ferrous or nonferrous alloys. Some
stainless steels are heated to a rela-
0
tively high temperature, 1900 F or
higher, and cooled rapidly. This treat-
ment places free carbides in solid solu-
tion, improves formability, and restores
corrosion resistance to certain acid
media.
A preliminary step to hardening of
appropriate nonferrous alloys is their
"solution treatment." This involves
heating the alloy to sufficiently high
temperature to place the alloying ele-
ments into solid solution; this is illus-
trated in Fig. 1:8 as temperature T 1.
After the quench, a metastable super-
saturated solid solution results which
can be age hardened. These metastable
solid solutions vary considerably in
their degree of stability. Copper-alu-
minum alloys (when supersaturated) are
quite unstable and will age harden with
time, at room temperature. On the other
hand copper-beryllium alloys (when
supersaturated) are moderately stable
LESSON 1 15

and will not transform or age harden

until heated to temperatures of 500° F
or more. Solution treatment and con-
trolled precipitation is widely used on
nonferrous and austenitic superalloys a
because this is the most practical tech- TI ·····································9
nique which permits significant harden- I
I
I
ing and strengthening of this large I

I
grou? of alloys. I
t
I
I
I
a+ Precipitate
Tempering is a heat treating process I
I
originally intended only I
+
I
for toughening and restoring some de- I
I

gree of ductility to fully-hardened steel. P.H. Alloy

It was formerly called "drawing," but Composition
T2 · ~
this term is now considered to be ob- I
I
solete. For nonferrous metals, I'tem- I
I
per" refers to a degree of hardness I
I
I
achieved either by heat treating or ~
mechanical working; thus we have
terms as "annealed temper" for fully % Alloy
soft material. or "hard temper" and Fig. 1:8. Phase Diagram Showing Change of
,. spring temper' 9 for the very hard Solubility with Temperature.
condition.
If an alloy having a composition illus-
trated by the dash line is held at T 1 it
In ferrous metallurgy, tempering will dissolve any precipitate present
refers to a reheating process which and become single phased. When
toughens and usually softens a quench- quenched to room temperature, it will
hardened or normalized steel. The tern- probably remain single phase for a
pering temperature is below the trans- while. In time, especially if the temper-
formation temperature (usually well be- ature were raised to T 2, precipitation
low) and the subsequent cooling rate is will tend to occur.
of little or no consequence. Therefore,
oil quenching or air cooling is most loy constructional steels are usually
common. Since the as-quenched steel tempered at temperatures between
microstructure is known as martensite, 1000°F and 1200° F.
the tempered structure is referred to
simply as tempered martensite. Hardening Processes

As the tempering temperature is in- The descriptions of these heat treat-

creased hardness decreases, but duc-
tility and toughness are increased. For
tool applications which require maxi-
mum hardness, tempering temperatures
are usually held to a maximum of
400 ° F. However, some alloy steels such
as air-hardening or high speed steels
are tempered at temperatures between
900°F to 1050°F with practically no
loss in hardness. For maximum tough-
ness with moderately good strength, al-
ments may not be fully understandable
to the "uninitiated," but they serve to
define the variety of terms normally
encountered. Furthermore, they will be
clearly understood when you review this
lesson after haVing studied steel heat
treatment in later lessons.
Quench Hardening of various iron and
steel alloys is ac-
complished by heating them above their
16
upper critical temperatures (to the aus-
tenitic condition),. and cooling them by
quenching in brine~ oil, or air.
Cooling rates which are required to
obtain full hardness are dependent upon
the alloy composition of a particular
steel or cast iron. Alloying elements
such as manganese, chromium, nickel,
and molybdenum decrease transforma-
tion rates thu~ permitting slower cool-
ing rates. such as oil or air cooling,
while still.resulting in maximum hard-
ness. This subject is more fully dis-
cussed in a later lesson.
Because oil or air-quenched parts
have less drastic temperature gradi-
ents. internal stresses will also be less.
Minimizing internal stresses reduces
distortion and cracking of hardened
parts. .PAfter full hardening, most steels
and cast irons are relatively brittle,
because of the nature of their micro-
structure and the residual internal
stresses which are present. To reduce
this brittleness, hardened steels and
cast irons are usually tempered.
Time Quench and Temper treatments
are modi-
fications of conventional quench and
temper treatments. Plain carbon or low
alloy steels which normally would re-
quire brine quenching for full hardness,
are initially brine quenched followed by
a quick transfer to a warm oil bath.
Because cooling rates on these steels
need only be very rapid until cooled be-
low about 800°F full hardness is ob-
tained without the danger of excessive
distortion or cracking encolh,tered with
a full brine quench.
After reaching room temperature,
parts are immediately tempered. Al-
though this type of heat treatment is
rather infrequently used as compared
to conv~n~ional quench and temper
treatments, it does fit'ld useful applica-
PRINCIPLES OF HEAT TREATING
tions in heat treating some carbon steel
tools.
Marquenching, also known as mar-
tempering, is a heat
treating process in which steel or
cast iron is quenched from the aus-
tenitizing temperature into molten salt
(or other media) which is usually held
just above the M s temperature. In some
cases it might be just below the M s
temperature. Transformation to the
hardened condition does not occur Wltil
the temperature drops below the M s
temperature. The part is held in the
bath for a sufficient time to deYelop
uniformity of temperature between
varying section thicknesses, after which
it is air cooled. During the remaining
cooling cycle transformation of austen-
ite (the high temperature structure) to
the hard martensite structure occurs.
After air cooling, the material is
tempered to the desired hardness. Mar-
quenching is widely employed for hard-
ening steels where a minimum am01.mt
of distortion is required. It is most use-
ful for hardening materials of relatively
thin section or of more highly alloyed
composition because the quenching ef-
fectiveness of molten salts is somewhat
inferior to that of oil quenching baths.
Austempering is accomplished by
heating cast irons or
steels above the upper critical temper-
ature and quenching them into a molten
salt bath, which is held at temperatures
well above those used for marquench-
ing. When the parts are held at the
proper temperature for varying periods
of time dependent upon composition,
(15 min. to 2 hrs.) transformation oc-
curs to a constituent mown as bainite
which is strong, tough, and ductile. The
parts may then be air-cooled or oil-
quenched.
Tempering of this structure is usu-
ally not necessary. Austempering is a
LESSON 1
useful process for producing greater
toughness with equivalent hardness than
can be obtained by the conventional
hardening and tempering process. Aus-
tempering is being used in production
heat treatment of steels S1:1ch as A.ISI
1050 which requires only a short time
for transformation.
It is a more costly process than the
conventional quench and temper when
used for heat treating certain alloy
steels which require considerable time
for transformation.
Precipitation This is a hardening
Hardening process which is also
known as ' I age harden-
ing/' In precipitation hardening, an al-
loy of appropriate composition is first
solution treated as was described pre-
viously in Fig. 1:8 at temperature T 1.
It is then "aged" at room temperature
or by holding for specified periods of
time at a slightly elevated temperature.
Temperature T 2 in Fig. 1:8 indicates
the temperature and phase relationship
which results in precipitation. This al-
lows numerous tiny precipitates to form
which harden the alloy.
The most common heat treating
process available for hardening nonfer-
rous alloys such as aluminum, copper,
and nickel alloys is precipitation hard-
ening. The high strength aluminum al-
loys which have made modern .aircraft
possible are all precipitation hardening
alloys. Precipitation hardening stain-
less steels and superalloys have also
been Widely used for jet engine and
missile construction.
Small Particle Since the introduction
Hardening of SAP (sintered alu-
minum product) from
Europe about ten years ago, the prac-
ticability of strengthening pure metals
and alloys by introducing fine relatively
"inert~' phases has been investigated.
17
This is sometimes referred to as " dis-
persion hardening."
SAP is a powder metallurgy product
in which flaked aluminum particles,
lightly coated with an oxide layer are
compacted and sintered to shape. Ex-
cellent strengths almost up to the melt-
ing point of pure aluminum have been
achieved. Improvements have also been
obtained with copper, nickel, and other
metals.
Other than introducing inert or ox-
ide particles by the power metal tech-
nique, attempts are being made to cast
liquid metal with a fine suspension of
solid particles; however, their segre-
gation, sinking, or floating remains a
major problem.
Some success has also been achieved
by controlled corrosion or internal oxi-
dation of silver alloys containing a few
per cent aluminum. This results in
Al 2 0 g precipitation in the affected
zones, resulting in hardening. Increas-
ing the AI content in silver to higher
levels results in a continuous Al 2 0 3
surface film which yields tarnish-re-
sistant sterling silver.
Ordered Reaction This is a process
Hardening wherein an alloy
must first be solu-
tion treated. Alloys of the proper com-
position are then reheated to a tempera-
ture below the solution treatment tem-
perature for extended periods of time
(10 to 20 hours) and air cooled. This
treatment converts the random arrange-
ment of solvent and solute atoms of the
normal solid solution to a regular or
ordered arrangement in the crystal.
This is illustrated in Fig. 1:9.
The strength of these ordered solid
solutions is greatly increased as com-
pared to random solid solutions. High
temperature strength is also greatly
improved. Alloys of this type (S-816)
 18
Disordered Ordered
Fig. 1:9. Schematic Illustration of Disord-
ered and Ordered Solid Solution.
have been used extensively for high
temperature applications in jet engines.
Transformation- occurs in several
Precipitation special stainless
Hardening steels. Heat treat-
ments usually con-
sist of a solution treating anneal which
places the steel in the austenitic condi-
tion. A subsequent reheating to 1400° F
followed by cooling to a temperature of
60° F results in transformation to mar-
tensite as is obtained with conventional
hardening. An alternate method of re-
heating to 1700° F followed by cooling to
-100°F produces a similar transforma-
tion.
Reheating this martensitic structure
to temperatures of 950°F-1050°F tem-
pers the martensite but also causes
precipitation hardening, which result in
a further increase in hardness. These
alloys because of their high strength
and relatively easy heat treatment (no
severe quenching is required) are find-
ing wide use in high speed aircraft and
missile parts.
Ausforming Deformation of metasta-
ble austenite just prior
to martensite formation is a relatively
new technique which results in ultra-
high strength. Below the pearlite nose
of a TTT curve is an area where Wl-
stable austenite persists for a rela-
tively long time (it is metastable). De-
pending on the alloy composition, this
temperature range may be between 1000
to 750°F or even lower.
PRINCIPLES OF HEAT TREATING
To give ample time for the defor-
mation process, a steel of relatively
high hardenability should be used. The
process results in some work harden-
ing, preferred orientation, and most
important, a refined martensite struc-
ture.
The thermal cycle involves 1) aus-
tenization (about 1650°F), 2) quench to
ausforming temperature (about 800°F),
3) deformation at 800°F, 4) oil quench
to room temperature, and 5) temper to
achieve desired final hardness or duc-
tility .
Surface Hardening
The two major classes of surface
hardening techniques are 1) those in-
volving no change in surface chemistry;
only a differen~e in surface heat treat-
ment, and 2) a change in surface chem-
istry. Medium or high carbon steels are
used for the first technique, and low
carbon steels (less than 0.30% C) are
generally used for the second technique.
In the first category we are princi-
pally concerned with induction harden-
ing and flame hardening, although salt
baths are occasionally used for surface
heating. In the second category are:
carburizing, nitriding, carbonitriding,
cyaniding, and special surface impreg-
nation treatments.
Carburization In carburizing, parts
are heated in either a
solid, liquid or gaseous carburizing
medium for varying periods of time de-
pendent on depth of penetration which is
required. Carbon is absorbed in the
surface of the steel. Temperatures are
usually about 1650-1700°F. After car-
burizing, the parts are usually hardened
by conventional water-quench and tem-
per processes.
Cyaniding In cyaniding, parts are
heated in a molten cyanide
bath (about 1600° F). Both nitrogen and
LESSON 1
carbon are absorbed. Quenching in oil
or water is usually done directly from
the bath. Final case depths are usually
about 0.003 to 0.008 in. Hardness of
Rockwell C-65 are common.
Nitriding is a process which employs
dissociated ammonia gas at
relatively low temperatures (about
0
1000 F). The work is held in this gas in
a sealed furnace jor extended periods of
time (24 to 72 hours). No quenching is
required.
Very hard shallow penetration of the
surface (0.010 to 0.012 in.) is obtained
with very little distortion. 'When the
specially selected alloys are used, ex-
tremely hard wear-resistant surfaces
can be obtained.
Carbonitriding is similar to carbu-
rizing in a gaseous
atmosphere except that ammonia is also
added to the atmosphere. The final case
surfaces which are obtained are similar
to those obtained by cyaniding, and may
be fully hardened in an oil quench.
AJl of the above surface hardening
processes find wide use in industry for
hardening such varied products as
gears, bolts, cylinder liners, etc.
Surface Impregnation As carbon or
nitrogen can be
diffused into steel, so can other ele-
ments such as Cr, Si, AI, etc. Not only
is hardness changed, but highly im-
proved corrosion and wear resistance
can be achieved. Appropriate gases or
compounds surround the work piece (a
chlorine gas is commonly used to help
transport metal atoms or ions from the
compound to the metal surface). Terms
such as siliconizing, chromizing, alu-
minizing, as well as certain proprietory
names are applied to these processes.
*Sometimes called" ductile cast iron" or SG iron (for spheroidal or spherulitic graphite).
19
In addition to the many heat treating
processes which are described above
there are many specialized heat treat-
ing processes:
Patenting, for example, is used to
pretreatwire prior to cold
drawing. In the process, wire is heated
well above the upper critical tempera-
ture and then quenched in molten lead
held at a temperature below the lower
critical temperature. Cold drawn wire
which has been patented prior to draw-
ing has proven superior to other wire
for suspension bridge cables.
stiffening Temper is a heat treatment
wherein severely
cold worked steel parts are heated to a
0
temperature of 400 to 800 F held for
one or more hours and air cooled. This
treatment increases the yield strengths
of such metals as cold drawn austenitic
steels by reducing harmful residual
stresses. In some alloys it also is
probably accompanied by some precipi-
tation or age hardening. The increase in
yield or elastic properties permits
higher stressing of the parts in service
with accompanying weight savings.
Malleablizing is a process for con-
verting brittle white
iron castings to ductile malleable iron
castings. Because of the superior cast-
ability of cast iron, it is preferred over
steel for many parts. However, all cast
irons, except nodular cast iron* are
relatively brittle in the as-cast form.
Brittle white iron castings are made
quite ductile by packing in an inert
material such as ferrous silicate scale.
c
The castings are then heated to 1600 F
held for 40 to 60 hours, and cooled
slowly to room temperature. Malleabl-
izing converts the numerous iron car-
bides of white cast iron to iron and free
20
spheroidal-shaped graphitic carbon. The
resulting microstructure has strength
and toughness comparable to some
lower strength steels. Thousands of
tons of malleable iron castings are
processed each year for a wide variety
of products for use in' such varied in-
dustries as automotive, agricultural im-
plements, appliances, buildings and con-
struction, etc.
The above list of heat treatments
are some' of the more commonly used
PRINCIPLES OF HEAT TREATING
processes. In addition to these, there
are many specialized processes with
others being constantly developed. Heat
treatments are indeed a widely applied
and constantly growing group of im-
portant processes.
It has been necessary to refer to
terms in this lesson which will be
thoroughly discussed in subsequent
lessons. The student should not be
overly concerned if some concepts or
phases are not clear.
 Telnperature COil vers ions
Albert Sauvellr type of table. I."ook up reading in middle column; collirnn; if in degrees Fahrcnheit, read Centigrade equivalent in
if in degrees Ccntigrade, read Fahrenheit equivalent in right-hand left-hand column. Values as printed in Ullethiehem Alloy Steels".

C
-459.4 to 0

-=;;;; --- ~ -0-

F C lij 0 to 100
C
~ --;;;-
F
100 to 1000
C----------F -·0·----- --- ----;- - 0- - ; - - 0
ill -;; --;;;;; m -;; ---;;;;
1000 to 2000
I I P
elF
2000 to 3000

0 I F
- -2-;-
-268 -450 -17.2 1
32
33.8 10.6
122.0 - ;
51 123.8 43
-;;;
110 230 266 510 950 543 1010
1832 --;;- --;;;;;
1850 821 1510
-;:;;;-
2750
~ -;;;;- ~ ~ -;;;; ~2
-262 -440 -16.7 2
I
35.6
1099 I 2010 3650 1377 I 2510 4550
11.1 521125.6 49 120 248 271 520 968 549 1020 1868 827 1520 2768 1104 2020 3668 1382 I 2520 4568
-257 -430 -16.1 3 37.4 11.7 53 127.4 54 130 266 277 530 986 554 1030 1886 832 1530 2786 1110 2030 3686 1388 I 2530 ! 4586
-251 -420 -15.6 4 39.2 284
12.2 54 1129.2 60 140 282 540 1004 560 1040 1904 838 1540 2804 1116 2040 3704 1393' 2540 ~ 4604
-246
--240
-410
-400
-15.0
-14.4 I
5
6
41.0
42.8
12.8
13.3
55 131.0
56 132.8
66
71
150
160
302
320
288 550 1022 566 105~ 1922 843 1550 2822 1121 2050 3722 1399 I 2550 1 4622
293 560 1040 571 1060 1940 849 1560 2840 1127 2060 3740 1404 2560 I 4640
-234 -390 -13.9 7 44.6 13.9 57 134.6 77 170 338 299
-229 --380 570 1058 577 1070 1958 854 1570 2858 1132 2070 3758 1410 2570 1 4658
-13.3 8 46.4 14.4 58 136.4 82 180 356 304
·-223 -370 580 1076 582 1080 1976 860 1580 2876 1138 2080 3776 1416 2580 4676
-12.8 9 48.2 15.0 59 138.2 88 190 374 310 590 1094 588 1090 1994 866 1590 2894 1143 1 2090 3794 1421 2590' 4694
--218 -360 -12.2 10 50.0 15.6 60 140.0 93 200 392 316 600 1112 593 1100 2012 871 1600 2912 1149 1 2100 3812 1427 I 2600 1 4712
-212 -350 -11.7 11 51.8 16.1 61 141.8 99 210 410 321 610 1130 599 1110 2030 877 1610 2930 1154 I 2110 I 3830 1432; 2610 4730
--207
-201
-340
--330
-11.1 12 53.6 16.7 62 143.6 100 212 413.6 327 620 1148 604 1120 2048 882 1620 2948 \160; 2120 I 3848 1438! 2620 I 4748
-196 -320
-10.6 13
14
55.4
57.2
17.2 63 145.4 104 220 428 332 630 1166 610 1130 2066 888 1 1630 2966 1106! 2130 I 3866 1443 I 2630 : 4766
-10.0 17.8 64 147.2 110 230 446 338 640 1184 616 1140 2084 893 1640 2984 1171 ~ 2140 3884 1449' 2640 I 4784
-190 -310 - 9.4 15 59.0 18.3 65 149.0 116 240 464
. I I
343 650 1202 621 1150 2102 899 I 1650 3002 1177 2150 3902 1454 2650 4802
-184 -300 - 8.9 16 60.8 18.9 66 150.8 121 250 482 349
-179 -290 660 1220 627 1160 2120 904 1660 3020 1182 2160 3920 1460 2660. 4820
- 8.3 17 62.6 19.4 67 152.6 127 260 500 354 670 1238 632 1170 2138 910 1670 3038 1188 I 2170 3938 1466 2670 I 4838
-173
-169
-280
-273 -459.4
-
-
7.8
7.2
18
19
64.4
66.2
20.0
20.6
68 154.4
69 156.2
132 270
138 280
518
536
360
366
680 1256 638 1180 2156 916 1680 3056 1193 I 2180 3956 1471; 2680 4856
690 1274 643 1190 2174 921 1690 3074 1199 2190 3974 1477' 2690 4874
-168 -270 -454 - 6.7 20 68.0 21.1 70 158.0 143 290 554
-260
371 700 1292 649 1200 2192 927 1700 3092 1204 2200 3992 1482 I 2700 4892
-162 -436 -- 6.1 21 69.8 21.7 71 159.8 149 300 572 377 710 1310 654 1210 2210 932 1710 3110 1210 2210 4010 1488: 2710 4910
-157 -'250 --418 - 5.6 22 71.6 22.2 72 161.6 154 310 590 382 720 1328 660 1220 2228 938 1720 3128 1216 2220 4028 1493' 2720 4928
-151 -240 -400 - 5.0 23 73.4 22.8 73 163.4 160 320 608 388 730 1346 666 1230 2246 943 1730 3146 1221 2230 4046 1499 2730 4946
-146 -230 --382 - 4.4 24 75.2 23.3 74 165.2 166 330 626 393 740 1364 671 1240 2264 949 1740 3164 1227 2240 4064 1504 2740 I 4964
-140 -220 -364 - 3.9 25 77.0 23.9 75 167.0 171 340 644 399 750 1382 677 1250 2282 954 1750 3182 1232 2250 4082 1510 i 2750 4982
-134 -210 -346 - 3.3 26 78.8 24.4 76 168.8 177 350 662 404 760 1400 682 1260 2300 960 1760 3200 1238 2260 4100 1516 I 2760 5000
-129 -200 -328 - 2.8 27 80.6 25.0 77 170.6 182 360 680 410 770 1418 688 1270 2318 966 - 1770 3218 1243 2270 4118 1521 i 2770 5018
-123
-118
-190
-180
-310
--292
-
-
2.2
1.7
28
29
82.4
84.2
25.6
26.1
78
79
172.4
174.2
188
193
370
380
698
716
416
421
780 1436 693 1280 2336 971 1780 3236 1249 2280 4136 1527 i 2780 5036
790 1454 699 1290 2354 977 1790 3254 1254 2290 4154 1532 I 2790 5054
-112 -170 -274· - 1.1 30 86.0 26.7 80 176.0 199 390 734 427 I 2800
-107
-101
-160
-150
-256
-238
- 0.6
0.0
31
32
87.8
89.6
27.2
27.8
81 177.8
82 179.6
204
210
400
410
752
770
432
438
800
810
820
1472
1490
1508
704
710
716
1300
1310
1320
2372
2390
2408
982
988
993
1800
1810
1820
3272
3290
3308
1260
1266
1271
2300
2310
2320
4172
4190
4208
1538
1543
1549
I 2810
2820
5072
5090
5108
- 96 -140 -220 0.6 33 91.4 28.3 83 181.4 216 420 788 443 830 1526 '121 1330 2426 999 1830 3326 1277 2330 4226 1554 2830 5126
- 90 -130 -202 1.1 34 93.2 28.9 84 183.2 221 430 806 449 840 1544 727 1340 2444 1004 1840 3344 1282 2340 4244 1560 2840 5144
- 84 -120 -184 1.7 35 95.0 29.4 85 185.0 227 440 824 454 850 1562 732 1350 2462 1010 1850 3362 1288 2350 4262 1566 2850 5162
-
-
79
73
-110
-100
-166 2.2 36
37
I 96.8 30.0 86 186.8 232 450 I 842 460 860 1580 738 1360 2480 1016 1860 3380 1293 2360 4280 1571 2860 5180
-148 2.8 98.6 30.6 87 188;6 238 860 466 870 1598 743 1370 2498 1021 1870 3398 1299 2370 4298 1577 i 2870 5198
- 68 - 90 -130 3.3 38 100.4 31.1 88 190.4 243 460
470 1 878 471 880 1616 749 1380 2516 1027 1880 3416 1304 2380 4316 1582 2880 5216
- 62 - 80 -112 3.9 39 102.2 31.7 89 192.2 249 480 896 477 890 1634 754 1390 2534 1032 1890 3434 1310 2390 4334 1588 2890 5234
- 57 - 70 - 94 4.4 40 104.0 32.2 90 194.0 254 490 I 914 482 900 1 1652 760 1400 2552 1038 1900 i 3452 1316 : 2400 4352 1593 2900 5252
- 51 - 60 - 76 5.0 41 105.8 32.8 91 195.8 488 910 1670 766 1410 2570 1043 1910 3470 1321 2410 4370 1599 2910 5270
- 46 - 50 -- 58 5.6 42 107.6 33.3 92 197.6 493 920 1688 771 1420 2588 1049 1920 3488 1327 1 2420 4388 1604 2920 5288
- 40 - 40 - 40 6.1 43 109.4 33.9 93 199.4 499 930 1706 777 1430 2606 1054 1930 3506 1332 I 2430 4406 1610 2930 5306
- 34 - 30 - 22 6.7 44 111.2 34.4 94 201.2 504 940 1724 782 1440 ~ 2624 1060 1940: 3524 1338 I 2440 4424 1616 2940 5324
.....:. 29
- 23
-
-
20
10
-- 4
14
7.2
7.8 I
45
46
113.0
114.8
35.0 95 203.0
35.6 96 204.8
510 1I 950 I 1742
516
788 1450' 2642 1066 1950' 3542 1343 2450 4442 I
1621 2950 5342
960 /1760 793 1460 2660 1071 1960 3560 1349 2460 4460 1627 2960 5360
- 17.8 o 32 8.3 I 47 116.6 36.1 97 206.6 521 1 970 . 1778 799 1470 2678 1077 1970 3578 1354 2470 4478 1632 I 2970 5378
8.9 48 118.4 36.7 98 208.4 527 ; 980
49 1796 I 804 1480 2696 1082 1980 3596 1360 2480 4496 1638 ~ 2980 5396
904 120.2 37.2 I 99, 210.2 532 990 i 1814 810 1490 2714 1088 1990 3614 1366 2490 4514 1643 2990 5414
I
II

37.8 ; 100 . 212.0 538 ~ 1000 1832 1093 2000 3632 1649 3000 5432

IT I i I
1 I "-------II I I !---J
I'
 American Society for Metals
Metals Park, Ohio 44073

COURSE 41 TEST 1
E IGENERAL ASPECTS OF HEAT TREATMENT IPAGE 1 OF 2
PLEASE PRINT YOUR NAME AND AODRESS BELOW

N"ME _

Regis. No.
STREET _

CITY STATE ZIP _

Date

, not write
this space MULTIPLE CHOICE
Place the appropriate letter (A, B, C, 'or D) in the box.
(Only one answer is correct and each guestion has a value
of 10 points.)

1. A 1050 steel can be spheroidized by:

A.
B.
Oil quenching from the austenitic condition.
Normalizing.
D
C. Austempering.
D. Holding for a long time just below the Al
temperature.
2. The primary purpose of tempering a steel after quench

A.
hardening is to:
Harden it.
D
B. Strengthen it.
C. Induce recrystallization.
D. Toughen it.

3. Of the following, the microstructure that will give

the best machining response in a medium-carbon low-
alloy chromium-molybdenum steel is: D
A. Quenched and tempered martensite with a hardness
of 300 Brinell.
B. 90% pearlite plus 10% ferrite produced by
annealing.
C. A combination of spheroidized cementite and
ferrite.
D. A fine dispersion of precipitated particles.
4. A heat treatment that is often used on alloy castings
prior to further processing is:
A. Homogenization. c. Age hardening.
D
B. Spheroidization. D. Tempering.
I'
F
I
I
I
I
I
I
I
I
I
I

,
I
 t COURSE 41 TEST 1

I GENEFAL ASPECTS OF
HEAT TREATMENT IPAGE 2 OF 2
Do not write
in this space 5. A basic premise of all heat treating operations is that
A. Mechanical properties are changed by changes in
internal crystalline structure or microstructure. I
B. Atoms in a crystal are randomly arranged rather tha
in a geometrical pattern.
C. All metals are completely soluble in each other.
D. Melting point and recrystallization temperature are
inversely related to each other.
-
To provide ductility to white iron castings you would:
6•

A.
B.
Carburize.
Malleablize.
c.
D.
Austemper.
Martemper.
D
7. The kappa (K) phase in the aluminum-copper phase
diagram
A.
F~gure 1:7:
Is an intermetallic compound of copper and aluminum
D
B. Is 100% liquid at 1018 F.
c. Can contain up to 52.5% copper.
D. Is a single phase solid solution of copper and
aluminum.
8. If a 1040 carbon steel part were to be surface hardened
without changing its chemical composition, you would: [:1-
A. Carburize. c. Ni tr ide.
B. Induction harden. D. Siliconize.
9. If a low alloy steel part were susceptible to distortio
and cracking during the standard oil quench through
hardening treatment, this could be minimized (assuming
that mechanical properties will be adequate) by:
D
A. Increasing the cooling rate.
B. Austempering.
C. Malleablizing.
D. Chromizing.
10. A hot rolled steel is subsequently cold rolled before
final forming into finished parts. The most important
heat treating process required to assure good form-
ability is:
D
A. Ausforming. c. Age hardening.
B. Quench hardening. D. Annealing.
 ASM Intlmatloftll 1t is a Society whose
mission is to gather, process and
disseminate technical information.
ASM fosters the understanding and
application of engineered materials and
their research., design., reliable manu-
The Mlterlals facture, use and economic and social
Informadon Society benefits. This is accomplished via a
unique global information-sharing
network of interaction among members
in forums and meetings, education
programs, and through publications
and electronic media.

Printed on DocuTech February 1994

 The Materials
Information Society

Steel and Its

Mechanical Properties
by
George F. Melloy, B.S.

Course 10
Lesson, Test 1
Course 11
Lesson, Test 10
Course 41
Lesson, Test 2
 Metric Conversion Factors
To convert from To Multiply by

in. mm 25.4
in. m 25.4 x 10-)
mil ~ 25.4
J1in. ~ 25.4
in.2 m2 6.45 x 10--
in. 3 m3 1.64 x 10-'
ft. m 3.048 x 10-1
ft.2 m2 9.29 x 10-2
ft.3 m3 2.831 x 10-2
oz. g 2.834 x 101
lb. kg 4.536 x 10-1
Btu J 1.054 x 1()3
Btullb. - of Jlkg - K 4.18 x l(Y
Btulft - hr • of W/m-K 1.730
10 inJin.JOF 1()-6 mmll11JI1?C 1.8
psi Pa 6.895 x 1()3
psi kPa 6.895
ksi kPa 6.895 x 1Q3
ksi MPa 6.895
Icsi - in. 1l2 MPa-ml12 1.099
ksi...Jin. MPa-~m 1.099
oz.f gf 28.4
lbf kgf 4.536 x 10-1
lbf N 4.448
lbf· ft. N· m (or J) 1.356
Ibflin. 2 kgf/cm2 14.223
IbfJin. 3 kgf/m3 2.768 x lOC
IbJft. 3 kglm3 16.019
Ib.Jin. 3 g/cm3 2.768 x l()l
Ib.Jin. 3 kglm3 2.768 x lOC
gal (U.S. liquid) L 3.785
gal (U.S. liquid) m3 3.785 x 10-3
IbJgaI gIL 119.826
ft./gal mlJL 748
OF °C (OF - 32)/1.8
OF K eF + 459.67)/1.8
°C OF (OC· 1.8) + 32
°C K °C+ 273.15
K °C °C -273.15

Multiple and Submultiple Units

1018 •••••••••••••••••••••••••••• exa 1(}-1 ••••••••••••••••••••••••••• deci
1015 ••••••••••••••••••••••••••• peta 10-2 centi
1012 ••••••••••••••••••••••••••• tera 1(}-3 •••••••••••••••••••••••••• nrilli
1(JJ •••••••••••••••••••••••••••• giga 1()-6 •••••••••••••••••••••••• micro
1()6 mega 10-9 nano
1()3 kilo 10-12 pico
1()2 hecto' 10-1S ••••••••••••••••••••••• femto
101 ••••••••••••••••••••••••••• deka 1(}-18 ••••••••••••••••••••••••••• atto

Abbreviations
J .•...•••.•••...•.•..•••••Joule m meter Pa pascal
kgf kilogram force mm millimeter K kelvin
L liter N newton W watt

by

George F. Melloy, B.S.

e 1s-n American Society tor Metals

This material may not be reproduced in whole or in part in any term whatsOeVer. FUlly protected by copyright
 Steel and Its Mechanical Properties
It may truly be said that we are living
in an age of steel. If there is anydoubt
about the importance of steel in our every-
day life, review briefly the typical dayt s
activity of any average American, and
note how often steel is encoWltered, and
in what important uses. We awake in the
morning to the sOWld of an alarm clock
made of many steel parts, get out of our
bed with its springs of steel, shave with a
steel razor blade, cook breakfast on a
steel stove, using steel cooking utensils,
and go to work in an automobile made
largely of steel. On the way to work we
may pass over a steel bridge and notice a
steel train riding on steel rails. When we
arrive at work we may sit on a steel chair
at a steel desk or perhaps perform our
daily tasks with a piece of steel machinery
or with steel tools. After completing our
day's work, we may seek recreation using
golf clubs made of steel, go hunting and
fishing with equipment made of steel, or
relax to the strains of a piano concerto
whose music results from the vibration of
steel wires.
RevieWing our day's activity, itbecomes
apparent that steel plays not only a large
part in our life but also a very diversified
one. Consider the bed spring which can
be compressed into very small space and
which, upon removal of the load compress-
ing it, will "spring" back to its original
shape and size. Certainly the behavior of
the bed spring is vastly different from that
of our automobile fender which can be bent
far out of shape but, alas, will not "spring"
back to its original form when the load is
removed. Our razor blade behaves in still
another fashion; while it takes a keen edge
and cuts our whiskers nicely, it will break
in a sudden, abrupt fashion if bent exces-
sively.
The diversity of steel is never more
strikingly illustrated than when we use a
file to round off a rough edge, or a steel
tool bit to shape another piece of steel in
a metal lathe or similar metal working
equipment. Here we have two pieces of the
same metal so different in properties that
one, without damage to itself, will actually
cut the other.
Steel, then, is a very versatile material
and, like the chameleon which changes color
to match its background, can be made to
exhibit a wide range of physical and me-
chanical characteristics, as in the case of
the bed spring, automobile fender, and razor
blade.
If we are to understand steel we must
know several things:
(a) VVhat steel is
(b) What properties make it useful
and im portant
(c) How the properties are measured
and evaluated
(d) How the properties are attained
or "tailor made" to suit our needs
In this lesson we will be concerned Nith
items (a), (b), and (c), and will discuss (d)
in subsequent lessons.
WHAT STEEL IS
The basic ingredient of steel is the ele-
ment iron. An elelnent is a pure substance
which is not", by ordinary chemical means,
di visible into substances differing from
itself. Each element possesses a set of
chemical, physical, and nlechanical proper-
ties peculiar to itself, and which distinguish
it from the other elements.
In nature, elements are rarely found in
the pure state but generally contain at
least small amoWlts of other elements. In
fact, it is only with a great degree of diffi-
culty that these small amounts of impurities
can be removed effectively and a really
 2
100% pure elem~ntproduced.* The pres-
ence of a very small amount of a second
clement may alter the properties of the
original, or base, element considerably.
In fact, the addition of a second element
~ay actually change tbe properties of the
base element to such an extent that the
properties possessed by the combined ele-
ments are quite different, not only from
those of the base element but also from
those of the added element. In many cases,
the alteration of properties is beneficial,
so that other elements may deliberately be
added to secure a certain desired change
in properties.
Iron is one ~)f a large group of elements
classed as nletal.:;. When combined With
other elements, it is called an alloy.
An alloy is a substance having metallic
properties, and is composed of two or
more chemical elelTlents of which at least
one is a metal. Metals and alloys have
certain properties which distinguish them
from nonmetals. The properties most
easily recognized and evaluated are:
(a) Malleability and ductility. These
two properties are closely related
and simply mean the ability to be
hammered into thin sheets (malle-
able) and drawn into thin wire
(ductile). They represent the abil-
ity to be formed.
(b) Good heat conductivity
(c) Good electrical conductivity
(d) "Metallic" luster or ability to reflect
light.
(e) Strength, or the ability to resist de-
formation. A lump of modeling .
clay is easily formed with finger
pressure and is said to be weak.
In comparison, a piece of steel is
formed only with great difficulty
and is called strong.
When carbon is added to iron, in amounts
ranging up to 2.0%, it forms an alloy known
as steel.
* The purest metallic element produced by man to date has been silver of 99.9999% purity.
and if purer metals are prod~ced, the ability to verify the presence of such minute amounts of
impurity presents a serious practical problem.
A steel containing only carbon and iron
is called a carbon steel and is capable
of having properties quite different from
those of pure iron. During the steel mak-
ing operation a wide variety of elements
may be added to the steel deliberately.
These elements are added because they im-
part some particular additional benefit in
properties, which makes the steel more
useful. Such steels, to which alloying ele-
ments have been added deliberately, are
known by various general terms, such as
alloy steels, tool steels, free-machining
steels, etc. The general designation is de-
termined by the alloying element or ele-
ments added, the amount added, the use
the steel will be put to, or the manner
in which the steel is made. The various
groups of steels will be discussed more
fully later in the course.
It should be emphasized that the various
alloying elements, though desirable in some
instances, may under other circumstances
be very Wldesirable. Certain elements
exist in the steel as impurities in the raw
materials and cannot be economically elimi-
nated in commercial steelmaking processes..
When their presence is not beneficial, such
elements are usually reduced to a suitable
minimum during the steelmaking operation.
Thus, while not eliminated entirely, they
are present only in such small amounts
that they do not adversely affect the proper-
ties of the steel, and in these reduced
amounts their presence may then be toler-
ated. For instance, many steel specifica-
tions permit up to 0.040% phosphorus and
0.050% sulfur to be present. If more than
these amounts of phosphorus and sulfur
are present, serious problems in hot form-
ing operations are encountered. However,
special purpose steels may contain higher
amounts of sulfur (up to about 0.20%), since
this element produces ease of machining.
In like fashion, amounts of phosphorus in
excess of 0.040% may be specified to im-
prove the strength and atmospheric cor-
rosion resistance of the steel.
When

In any case, the properties of steel
vary widely from those of iron. The de-
gree of this variation depends to a certain
extent upon the amount of carbon and other
elements present. Heat treatment, how-
ever, is the main factor in controlling the
amoWlt of variation possible.
THE MECHANICA L PROPERTIES OF
STEEL AND THEm EVALUATION
The word steel has become synonymous
with strength, and the term "strong as
steel" is common in everyday language.
Steel is usually associated with strength
because in many of its uses it is neces-
sary that steel, in one form or another,
support a load. The bed spring supports
the weight of your body, the car the weight
of the passengers and their equipment.
The bridge must support the weight of
many cars and pedestrians, and so on.
Strength is a somewhat ambiguous term
and can. mean several things. An onion is
said to be strong, and we also speak of a
strong light; but the aroma of the onion
cannot support a load, nor can the beam of
light. Therefore, when we speak of
strength in relation to steel, it is neces-
sary to be specific in our definition. By
strength 'we mean the mechanicalproperties
of steel. The mechanical properties are
those concerned with the deformation (di-
mensional change) which occurs when an
external force (load) is applied. Simply
stated, STRENGTH is the ability of the
steel to support a load. Of the, various
properties possessed by steel, the me-
chanical properties are by far the most im-
portant and give steel its great usefulness
and diversity of application.
From our consideration of the bed
spring, the automobile fender, and the
razor blade, we saw that the steel could
be made to deform to a large extent, as in
the case of the bed spring, but the defor-
mation was not permanent, and when the
load was removed the spring would auto-
matically return to its original size and
shape. The car fender would deform readily,
but when the load was removed, would not
"spring" back to its original form. In
both cases, a large amount of deformation
3
was possible without fracture. Conversely,
relatively little deformation would break
the razor blade. In all three cases we
were concerned with the dimensional change
which took place when a load was a.pplied
or, more simply ~tated, we were interested
in one or more mechanical properties.
It is, therefore, apparent that when
steel is deformed its behavior during and
after deformation can be markedly different.
At this stage it is desirable to point ,out
what the various mechanical properties of
steel are, how we measure them, and what
they can mean to us in a practical sense.
Elasticity One important property of steel
is that, Wlder certain conditions
of loading, it is elastic in nature and can
be stretched like a rubber band and like the
rubber band will return to its original size
and shape after the load is removed. The
elastic nature of a rubber band is easily
demonstrated and recognized, since very
light loads will cause the rubber to stretch
to an appreciable extent, readily observed
by the unaided eye.
The .elastic nature of steel, while just
as real as that of the rubber band, is
more difficult to demonstrate. In order to
stretch the steel, very high loads are re-
qUired, far beyond those required for rub-
ber. Furthermore, the extent of the elastic
movement is far less in steel than it is in'
rubber and cannot usually be seen by the
unaided eye. Therefore, measuring instru-
ments must be used in order to measure the
stretching and recovery of the steel. The
matter of demonstrating elasticity in steel
is merely a matter of having the proper
testing equipment. Although the degree of
elasticity in steel appears to be quite small,
it should not be considered an unimportant
property. The elastic behavior of steel is
of extreme importance to those concerned
with the design of all load-bearing steel
structures or parts.
If a bar of steel L inches long and A
square inches in cross sectional area is
fixed at one end so that it cannot move,
and a load P is applied to the free end,
the bar will actually stretch or elongate
some small amount. This is illustrated in
4

e
~

(0) (b) (c)

Fig. 1:1. Elastic Deformation of Steel Bar; (a) before loading,
(b) loaded and (c) after removal of load. The defor-
mation, e, shown greatly exaggerated, is entirely elastic.

Fig. 1:1. If the load P is then removed, (A) of four square inches, the stress would
the bar will return to its original length be determined as follows:
in exactly the same fashion as a rubber Stress = piA = 200 pOWlds/4 square
band appears to do under light loading con-
ditions. This type of nonpermanent de-
inches = 50 pounds per square inch
formation is called elastic deformation, This would be similar to a 50-pound load
and elasticity is the property which hanging on a bar one square inch in cross
permits a material to deform under a load sectional area.
and to return to its original size and shape
The increase in length e divided by the
after the load is removed. With elastic
original length L gives us the increase in
deformation a bar can be loaded and un-
length per unit of original length, or the
loaded any number of times and, upon final
strain. As long as they can be measured
removal of the load, the bar will always
accurately, the units of measurement may
return to its original length.
be of any type, miles per mile, centimeter
per centimeter, etc. Strain is generally
- expressed as inches per inch, or per cent.
The load P divided by the cross sec-
As an example, consider a bar 12 inches
tional area gives us the load per unit of
long which has been loaded heavily enough to
cross sectional area and is called the
make it stretch O. 024 inch. The strain
stress. Stress is usually expressed in
would be:
this country as pOWlds per square i.nch.
For example, if a 200-pound load (P) is Strain = elL = Q.024 inch/12 inch =
hung on a bar having a cross sectional area 0.002 inch per inch, or merely 0.2%

: ,________t_ f . 1 Permanent
(0)
Fig. 1:2. Plastic Deformation of Steel Bar
(a) before loading
(b) loaded
(c) after removal of load
The deformation, e, is partly elastic and partly plastic; only
the elastic portion is recovered after removal of load.
If more weight is added to the load P,
the bar will elongate some small amount
greater than e, and for each successive
increase in load there will be a correspond-
ing proportional increase in the amount the
bar will elongate. Thus, for each increase
in stress there will be a corresponding
proportional increase in strain, and when
the stress is removed the strain will also
disappear as the bar returns to its original
length.
Plasticity The elastic condition does not
continue to exist without limit,
and, at some point of loading, we find that
when the load is removed, the bar does not
return to its original length but is actually
sonlewhat longer than it was originally. The
bar is then said to be permanently deformed
or plastically deformed. The maximum
stress that the bar can take without assum-
5
. _J_~~~~1: _
4
(c)
ing permanent deformation is known as the
elastic limit.
In metals, the elastic limit practically
coincides with the proportional limit,
which is detinec1 as me greatest stress a
material is capable of sustaining without
deviation from the law of proportionality of
stress to strain (Hooke'S Law). PLAS-
TICITY is the ability of a material to as-
sume large permanent deformation without
breaking. Fig. 1:2 illustrates plastic de-
formation in a bar stressed beyond the
elastic limit and into the so-called "plastic
range".
The Tensile Test
The properties of elasticity and plasticity
form the basis of one of the most valuable
tests (the tensile test) used in the evalua-
tion,of several of the mechanical properties
6

of steel. In the tensile test a standard
specimen, generally of the type 'shown in
Fig. 1:3, is machined from the steel to
be tested.
The important considerations in the ten-
sile specimen are that a parallel section
2-1/4 inches long and 1/2 inch in diameter
be accurately machined, and that within
this section a length of 2 inches be marked
off. This 2-inch length is called the gage
length and is used as the reference for
mea&'uring the dimensional changes which
occur. A generous fillet should join the
parallel section and the thicker end sec-
tions. The end sections are used as a
means of .holding the tensile bar in the
testing machine and may be smooth or

Minimum Radius Recommended

:" ,But Not Less Than Permitted +"
Parallel Section

C!) I Til ± 0.01 11

C!)

-;;;:;;
11
.
2"±0.005 Gage Length For
Elongation After Fracture
Fig. 1:3. Dimensions for Standard Tensile Specimen.

Fig. 1:4. Performing the Tensile Test.

The specimen A is mounted in the frame at left; the load
applied by the operator is shown on the dial B; a stress-
!!train diagram is automatically recorded at C.
(Courtesy Bethlehem Steel Co.)
 7

threaded, depending upon the type of chucks secured. These data give us the useful
or holders used in the testing machine. information we can obtain from the tensile
test.

The specimen is mounted by the ends

in a testing machine, as shown in Fig. 1:4,
and a continuously increasing load applied,
so that the specimen is stretched and
finally torn apart. By plotting the rela-
tionship of stress and strain which exists
during the test, a stress-strain diagram,
such as that shown in Figure 1:5, may be
It should be noted in the stress-strain
diagram that up to the point A, the elastic
limit, any increase in stress results in a
similar, proportional increase in strain.
This portion of the stress-strain diagram
represents that phase of the tensile test
where perfectly elastic conditions exist.

c c

" Soft, Ductile Metal

Certain Soft Steels Also Characteristic
Of Metals At High Tem

c
.~

en
en
Q)
i.-
Proportional Limit
+- {
A Approximate Elastic Limit
C/)
B - Upper Yie ld Point
1
8 - Lower Yield Point
C- Ultimate Tensile Strength

Strain ---~.

Fig. 1:5. Stress-Strain Diagram Plotted from Tensile Test Results.

Upper curves show other possible variations.
 8
In this region of stress, if the load were
removed, the elongated bar would return
to its original length with no permanent
defornlation occurring. This would be true
regardless of the number of times the load
was applied and removed.
In the region of stress above the elastic
linlit, the bar will not return to its original
length when the load is removed but will
permanently remain somewhat longer be-
cause of the fact that some plastic deforma-
tion has occurred (A). On some of the older
nlachines, which measure the load by means
of a weight, or "poise," on a graduated
beam, the yield point load is determined by
the ·'drop of the beam" method. This drop
of the beam occurs when the specimen sud-
denly stretches cr gives to make an increaSe
in strain with little or no increase in
stress.
Certain alloys exhibit what are called
"upper" and "lower" yield points. In high
carbon steels and most nonferrous metals,
the phenomenon of a well-defined yield point
is not observed.
After the upper yield point (B) has been
exceeded, elongation continues for a short
distance without addition of load. This
value of stress is called the lower yield
point (B ,) and is an average value, since
the load varies considerably during yielding
as a result of the testing condition.
The condition of increased strain (stretch)
without similar increased stress is soon
halted, and the situation again develops
where a further increase in stress is neces-
sary to get a further increase in bar length.
This is caused by the phenomenon of work
hardening.
Work hardening may be easily demon-
strated by repeatedly flexing a piece of soft
copper wire or a paper clip. With repeated
flexing, the wire becomes more and more
difficult to bend; that is, the work done in
flexing and plastically deforming the wire
causes it to become harder and stronger.
The same thing occurs when the steel
tenSile specimen is plastically deformed.
The metal in the plastically deformed area
becomes stronger from the work hard-
ening. This means that higher stress is
necessary if the area of work hardened
metal is to be further deformed. This con-
dition continues to exist until point (C)
is reached, which represents the ultimate
tensile strength. The ultimate tensile
strength is often referred to more simply
as the tenSile strength and is defined as the
greatest tensile stress (as based on original
cross sectional area) a bar can withstand
without rupture.
The phenon1enon of work hardening may
be understood by considering a log drive on
a river. The logs float down the river
easily if they are all lined up in the direc-
tion of the flow of water. If, however,
some of the logs are turned perpendicular
to the water flow, they may be forced into
the bottom or banks of the stream and catch
other logs which also become disoriented and
thus cause a log "jam".
In metals, plastic flow occurs by means
of groups of metallic atoms "flowing" over
each other. This flow occurs quite readily
if all of the atoms are properly oriented
with respect to each other. However, when
the atoms become disoriented an "atom jam"
occurs, requiring more force to cause the
metallic movement than was necessarywhen
the atoms were flOWing smoothly. Additional
movement causes the "atom jam" to become
tighter, requiring still more force to cause
further movement. This condition continues
until the force required for "metal flow" ex-
ceeds the strength of the metal itself, which
then starts to tear apart.
After the tensile strength of the steel
has been exceeded, the metal begins to de-
form quite rapidly. As the metal plastically
deforms, the length increases and the cross
section thins out. (The same conditions of
change in dimension are readily observed by
stretching "bubble" gum.) In steel, the plas-
tic deformation is often localized, and as a
result the reduction of the cross section is
visible to the unaided eye. With the reduc-
tion of cross section, the actual stress per
unit area mounts rapidly, causing the unusual
condition of strain increasing without a seem-
ing increase in stress. The actual increase

Fig. 1:6. Comparative Appearance of Tensile Specimen Before and After
Testing.
Note gage marks (at arrows) for elongation measurements
punched in specimen.
in stress is not noticed, since in the con-
ventional stress-strain curve the stress is
calculated on the basis of the original cross
sectional diameter of the specimen and not
the actual cross section of the moment.
When the cross section of the specimen be-
gins to reduce rapidly, it is said to "neck
down". Fig. 1:6 compares a tensile speci-
men after pulling with another specimen
having the same original standarddimen-
sians. The elongation due to plastic flow
and the "necking down n are readily observed.
On rare occasions, more than one "necked-
down" zone can be observed.
The stress in the conventional stress-
strain diagram is based on the original
cross sectional area, because the continu-
ous accurate measurement of the localized
reduction of cross sectional area would be
very difficult. This would make the tensile
test very slow, laborious, and difficult, de-
stroying its usefulness as a production test
procedure. Furthermore, the ability to
identify the yield point and tensile strength
affords several reference points of extreme
value to the design engineer, who must for
9
(Courtesy Bethlehem Steel Co. )
practical r.easons base his calculations on
original cross sectional areas.
Ductility After pulling the tensile speci-
men to the point of rupture, the
broken ends are removed from the testing
machine and carefully fitted together. By
means of a ca·liper and scale, the amount
that the specimen has permanently or plas-
tically elongated can be determined, as shown
in Fig. 1: 7. This measurement is taken
over the gage length (usually 2. 0 inches)
which is stamped on the bar before it is
pulled. After pulling, the gage length is
somewhat longer than the original 2 inches
on account of plastic deformation which has
taken place. The increase in length is ex-
pressed as elongation. Elongation is
the increase in the length of the gage length
expressed as a percentage of the original
gage length.
Percentage of elongation =
(Final length of gage length) - (Initial
length of gage length) X 100
Initial length of gage length
 10
Fig. 1:7. Method of Measuring Increase in Length Resulting from Plastic
Deformation During the Tensile Test.
Reduction of Area By means of a "point"
micrometer, the very
smallest diameter at the breaking point in
the necked-down portion of the tensile
specimen can be measured as in Fig. 1:8.
From this diameter we can determine the
final smallest cross sectional area and
thus obtain the reduction of area. Reduc-
tion of area is the difference between
the area of the smallest cross section
measured and the area of the original
cross section, expressed as a percentage
of the area of the original cross section.
Percent reduction of area =
(Original area) - (Area at breaking point) X 100
Original Area
(Courtesy B~thlehem Steel Co. )
Yield Strength The stress-strain diagrams
of some steels do not ex-
hibit any discernible yield point. In such
cases, the value used in place of the yield
point is the yield strength. Yield strength
is the stress at which a material exhibits a
specified limiting permanent elongation
(strain). An offset of 0.2% is generally
specified on the strain scale, and then a
line parallel to the linear part of the dia-
gram (below the elastic limit) is drawn. The
intersection of this parallel line and the
stress-strain diagram gives us the value for
the yield strength, as shown in Fig. 1:9.
Briefly, we have reviewed the method of
making a tensile test, and various values
measured by means of this test. How these
 11

Fig. 1:8. Measuring after Tensile Testing.

Diameter of specimen is measured in "necked" region to
determine reduction of area.
(Courtesy Bethlehem Steel Co.)

,
,
I

50000 ,
I

:..- Line Drawn Parallel to

.5 : Proportional Portion of Curve
g- 40000 I
I
at 0.2 % Offset
~ I
I
I
I
I
I
,
I
I
I
I
I
I
I
I

,,
I

0002 QOO4 0.006 0008 0010 0.012 0014

Stroin 7 in/in

Fig. 1:9. Determining Yield Strength by the Offset Method.

 12
I ,~
I
I
,
/' Yield Point
0.2 % Offset
--I
Fig. 1:10. Variations in Stress-Strain Diagram According to Carbon Content.
values may be used to determine the actual
service behavior of steel should now be
considered.
The stress-strain diagram obtained
from the tensile test may vary somewhat
according to the steel tested, as shown in
Fig. 1: 10. Al~oying elements, other than
carbon, have similar effects but will not
be considered here.
From this set of curves we can readily
see that the higher the carbon content, the
stronger the steel, as indicated by the
tensile strength. The tensile test may be
used to evaluate the relative strength of
different steels, or accurately place a
practical stress limitation on a given steel.
It may also be seen from Fig. 1: 10 that
the yield point (or the yield strength) varies
considerably. If we reconsider the bed
spring, the .automobile fender, and the
razor blade, which we originally used to
illustrate the fact that steel can vary con-
siderably in its properties, we will see the
usefulness of yield point or yield strength.
Applications If we wanted a good bed
spring, we would need a
steel of high tensile strength so it would
not rupture or break when heavily loaded.
We would also want the springs to retain
their original shape and not sag or distort
badly. This would require a high yield
strength, so that when loaded no great
Medium-Carbon Steel
Low-Corboh Steel
Strain - - -......
amount of plastic deformation could take
place in the springs. If absolutely no sag-
ging or plastic deformation were permissi-
ble, a high elastic limit would be specified,
so that any movement of the springs would
be entirely elastic in nature. These require-
ments would dictate the use of either a high-
carbon or medium-carbon steel. Certainly
the low-carbon steel would not be suitable'
because its low yield point would allow plas-
tic deformation, and after some use the
springs would permanently flatten and be
useless.
At first glance, we would say that the
automobile fender would be ideal if made of
high-carbon steel, for then it would resist
deformation and if deformed would spring
back to its original shape (high elastic limit
and yield strength). This, of course, has
some merit, but how would we form the
fender? The fender, as we know, is made
by placing a flat sheet of metal in a die
and forming it in a press. If the sheet
were high-strength, high-elastic limit steel,
a great amount of pressure would be needed
to form it, requiring extremely heavy and
costly equipment. Furthermore, the sheet,
after being pressed to shape, would spring
back toward its original flat condition when
the pressure was removed. Therefore, for
this purpose a low-carbon steel is much
more suitable, for its low strength makes
it easy to form, and its low yield point al-
lows it to plastically deform easily; and, as
indicated, it is capable of a large amount of
 13

plastic deformation before rupture (length The tensile test can then tell us many
of curve in the strain direction). things about how a steel will behave under
service conditions and what we can expect
Generally speaking, the higher the ten- from it. The properties of elastic limit,
sile strength the lower the elongation and yield point, tensile strength, elongation,and
reduction of area, and the lower the tensile reduction of area enter into the selection of
strength the greater the elongation and re- the proper type of steel for a service appli-
duction of area. The elongation and reduc- cation, as do certain nonmetallurgical as-
tion of area are the ductile properties and pects such as cost, availability, etc.
serve as indicators of the plasticity of steel.
If a steel is described as plastic, this
SELF CHECK QUIZ
means that it is capable of a great deal of
deformation without rupture, as indicated
1. A bar of steel 1/4 in. thick by 1 in. wide
by the large amount of strain possible with
was subjected to a tensile test. From the
the low-carbon steel. Therefore, the low-
following test data, plot the stress-strain
carbon steel with its low strength, low
diagram and compute the following me-
yield point, and high formability is an ideal
chanical properties (assume a 2 in. gage
material for making shapes such as an auto- length):
mobile fender.
a. proportional ·limit
The properties of elongation and reduc- b. upper yield point
tion of area are also useful because they c. lower yield point
frequently forewarn of failure in a part. d. ultimate tensile strength
If these values are high, considerable de- e. fracture stress
formation will usually occur before rupture f. % elongation
and give us adequate warning that failure of g. % reduction of area
the piece is going to occur. * Such behavior
is ductile behavior. If, however, these Load - pounds Strain - inch (Measured
values are low, the failure may occur with over 2 in. gage length)
little or no warning deformation, as in the
case of the razor blade. These facts have 2500 0.00067
been utilized by those employing lifting 5000 0.00133
chains. A chain made from steel jlaving 7500 0.002
high elongation can plastically deform a 8000 0.00233
very considerable amount before failure re- 8500 0.004
sults from overloading. Overloading here 8000 0.006
is usually due to a series of lifts over a 8500 0.00733
period of time. If a gage length is laid 10500 0.010
out on the chain and measured periodically, 12500 0.014
the extent of the cumulative overloading can 13500 0.018
be determined and the chain discarded be- 14000 0.025
fore failure occurs. The gage length is, 13900 0.031
of course, quite long and should cover num- 12000 ~racture) 0.360
erous links. Between these two extremes 3/16 in. x 7/8 in. at fracture
of behavior lies the medium-carbon steel,
capable of developing a reasonably high Example: To calculate the location of the
strength and reasonable ductile properties. first point to be plotted on the stress-
Such a material is often described as being strain diagram, the first load given in
tough.

,. Under certain conditions, "brittle failure" may occur in even the relatively soft steel, such as
used in ship plate. Under these conditions, measurement of the ductile properties is of little or
no use in predicting failure. Other tests described later a~e used to predict "brittle failure."
 14
the data must be converted to stress by Brinell Many hardness testing machines
using the formula: of the penetration type are mar-
keted and used today. One of the more com-
strain = 1~ 0.0~067 = 0.000335 in/in mon penetration-type hardness testing de-
vices is the Brinell hardness machine shown
Load 2500 lb 2500 lb in Fig. 1:11. In this test, an extremely hard
Stress =Area = 1/4 m
. x 1 .'
m = •250 . = 10000
sq m psi steel ball, 10 mm. in diameter, is pressed
into the flat surface of the steel, using a
This point can now be plotted on the 3000 kilogram load (500 kg. for soft metals)
stress- strain chart provided and each applied by operating a hydraulic pump.
subsequent relationship of stress to When the load is removed, it is seen that
strain can be calculated and plotted. By the hardened steel penetrator has dented the
drawing a line through the plotted points, steel and left a small round depression. The
a stress-strain curve will be established diameter of this depression is measured by
and the value of various mechanical prop- means of a portable hand microscope, which
erties can be computed from the curve. has a scale graduated in tenths of a milli-
meter engraved in its field of view. (See
Fig. 1:12.)

From this diameter the surface area of

the depression is calculated, and, by elividing
the load by the surface area of the depres-
sion, a Brinell hardness number (frequently
abbreviated BHN) is secured. For the sake
of convenience, the Brinell hardness num-
bers have been calculated for a wide range
of Brinell diameters, so that the actual
calculation need not be made at all. A
hardness value is obtained merely by re-
ferring to a conversion table.

Fig. 1:11. Brinell Hardness Testing Machine.

Test specimen is placed at A; the
load, usually 3,000 kilograms, is
sho'WIl at B.
(Courtesy Steel City Testing Machines, Inc.) Fig. 1:12. View Through Brinell Micro-
scope ShOWing How Diameter of Im-
Hardness Testing pression is Measured.

A secondmechanical test of considerable At least two readings should be

usefulness is the hardness test. When we taken since the iDlpressions are
not always perfectly round. Turn
speak of hardness, we generally mean pene- the microscope 90 degrees be-
tration hardness of the ability of the steel tween readings.
to resist penetration.
 15

Several. modifications of the Brinell test
are used to measure the hardness of rela-
tively hard or relatively soft materials. In
these modified tests several types of pene-
trator balls and variations of load are em-
ployed; the general procedure is, however,
the same. In all cases, the larger the de-
pression made by the penetrator, the lower
the Brinell hardness number and the softer
the piece, as one might expect.
more suitable hardness measurement tech-
niques are usually employed.
The Brinell test is,.however, quick, ac-
curate, and reproducible, and for these
reasons has WOD much favor. FUrthermore,
the equipment is fairly rugged, making it
suitable for use in the shop where the use of
more delicate instruments would be im-
practical.

The Brinell test is usually considered a

destructive test because of the relatively
large depression left by the penetrator.
The surface to be tested must be flat and
smooth, thus requiring grinding in many
cases. This also adds to the destructive
nature of the test.

A further limitation of the Brinell hard-

ness test lies in the fact that the hardness of
specimens less than 1/4 inch thick cannot
be determined accurately when the standard
10 mm. hardened steel ball and 3000 kg.
load (or 500 kg. in the case of softer metals)
is used·. The inaccuracy of hardness meas-
urements on thin sections is brought about by
the fact that when the penetrator is forced
into the steel, the load follows the penetra-
tor. The penetrator will, therefore, keep
moving until sufficient resistance is en-
countered to oppose the load. In soft ma-
terials, the ball may penetrate to an ap-
preciable depth. In thick sections, all re-
sistance to penetration of the ball can be
supplied by the material being tested; in Fig. 1:13. Rockwell Hardness Testing Machine.
thin sections, however, there is insuffi- Specimen is placed at A; the load
cient material to allow the amount of pen- (variable) is shown at B; standard
etration equivalent to the hardness of the test blocks at C are used to assure
piece being tested. Movement of the ball is, proper operation of machine, which
therefore, stopped by the hardened steel is tripped by levers D or E (not all
anvil upon which the test piece is placed, models have lever D).
rather than by the resistance of the test (Courtesy American Chain & Cable Co., Inc.)
piece. In thin sections, therefore, a wrong
hardness value may be obtained, because the Rockwell A second and widelyused method
hardness measured is partially that of the of determining hardness is by
test piece and partially that of the supporting means of the Rockwell test. A typical Rock-
anvil. well hardness tester is shown in Fig. 1:13.
The Rockwell test differs from the Brinell
While it is true that the hardness of thin test in several ways. A Rockwell hardness
sections may be accurately evaluated by determination is made in two stages. The
using smaller diameter balls and lighter first stage consists of applying a primary
loads, such nonstandard procedure is not 10 kilogram preload to the penetrator. This
very popular. When it is desired to deter- preload seats the penetrator into the mater-
mine the hardness of thin sections, other ial being tested and overcomes any minor
 16

surface imperfections present. The second

stage consists of applying the major· load to
the penetrator, which causes it to sink to
some additional depth. The major load is
then removed, and the difference in depth of
penetration occurring with the primary load
and the major load is used as the evaluation
of hardness. The difference 'in penetration
is measured automatically on a dial which
indicates the hardness of the metal. In the
Rockwell test, the load is applied through a
lever system by means of weights, and the
hardness values are read directly from a
calibrated dial. Several types of penetrators
are used in conjunction with different load
conditions, so that a wide range of hard-
ness values can be measured ·accurately.
The various combinations of penetrators
and loads together with their designations
are shown below. The ball penetrator is
made of highly hardened steel. The brale
consists of sphero-conical diamond point
SUitably mounted in a holder. The indicat-
ing dial has two sets· of values, in red and
Fig. 1:14. Vickers Hardness Testing Machine.
black, for the different scales.
(Courtesy Pittsburgh Instrument
Rockwell and Machine Company)
SCALE: TYPE MAJOR DIAL
OR OF LOAD IN HARDNESS
PREFIX P£NETRATOR leGS. NUMBERS
surface area of the impression is measured,
-a.!• the hardness value is determined by dividing
B Y.." BALL 100 Bed
aD ~
.,u
-U2
C
DiamoDd 150 Bleck the load (in kilograms) by the area (in square
millimeters) of the surface of the impres-
sion. The diamond pyramid. hardness (DPH)
A .~.
is determined by the formula -
D ..~.
60
100
Blac.:k
Black

..,
lo:-
a"U
~C'12
E
F
G
2/." Ball
'I,,' BaIl
'116
1P
Ball
100
60
150
Bed
Red
Bed
DPH = 1.8544L
d2
H If." Ball 60 Bed.
Ie If." Ball ISO Bed
L = Load in .kilograms
The Rockwell test leaves less of an in- d = Length of diagonal in millimeters
dentation than the Brinell test and can suc-
cessfully test thinner pieces. The surface In the Vickers test, a load varying from
to be tested must be quite smooth (consider- 1 to 120 kilograms is employed. The load
ably more so than in the Brinell test) and is applied by means of a weight- operated
level if the test is to be accurate. cam, and is both applied and removed auto-
matically by release mechanisms activated
Vickers Very accurate hardness values by the operator.
can be determined by means of
the Vickers hardness tester shown in Fig. The Vickers instrument is capable of ac-
1:14. In the Vickers test a predetermined curately determining the hardness of very
load is imposed upon the specimen through hard materials beyond the normal range of
a diamond penetrator in the form of a the standard Brinell test. Furthermore, it
square- based pyramid. The penetration of is possible, by controlling the load, to
the diamond penetrator leaves a square make only a minute impression and still
cross sectional indentation in the specimen secure accurate readings. This makes it
being tested, as in Fig. 1:15. After the feasible to take hardness readings on small

Applied I Load
Diamond ~
Penetrator" =-
:". '. ~', .: :.. . "'1 :':'..:: ~. :.': '4':. :' .:.
Test--.·,
\:
.'., .'1' . ' I . ' ··· .. ·4·
.. ·'1·::1
~ : ..
Specimen ." ~.
(0)
Fig. 1:15. S<:hematic Diagram Showing
(a) Vickers indentation
(b) measuring technique (diagonal length may be converted to
hardness number by means of conversion chart)
parts as well as on the various individual
areas of microscopic size within a given
large piece of steel. This latter fact
makes the Vickers instrument very useful
for investigational. purposes. Because of
the fine nature of the indentation made, the
surface to be tested must be flat and well
polished, or else difficul ty will be encount-
ered in reading the distance between the
corners of the impression.
Tukon When it is desired to determine
the hardness of small parts such
as fine wire, shallow case-hardened areas,
individual crystals or constituents of micro-
scopic size, or surface conditions, the
Tukon hardness tester is frequently em-
ployed. The device used may be either the
square- based diamond penetrator described
under the Vickers hardness test or a spe-
cial Knoop penetrator. The Knoop pene-
trator is a diamond ground to pyramidal
form but haVing a rhomb-shaped base, so
that it produces a diamond-shaped indenta-
tion with a long diagonal approximately
seven times the length of the short diagonal.
The Tukon test is fully automatic after
the specimen has been brought into contact
with the penetrator. The load applied may
be varied from 1 to 50,000 grams, depend-
ing on whether the hardness is being de-
termined on a micro or macro scale, light
17
(b)
loads being used for hardness testing on the
microscopic scale, and the heavier loads
for hardness testing on a macroscopic
scale. The length of the indentation (along
the long diagonal.) may vary from 5 microns
to 1,000 microns (one micron being equal
to 0.001 millimeter). Measurement of the
length of the indentation is done by means
of a microscope which is furnished with the
instrument.
When the square- based diamond pene-
trator is used, hardness is calculated as
DPH, as previously described under the
Vickers test. When the Knoop penetrator
is used, the Knoop Hardness Number (KHN)
is determined from the load and length of
the diagonal of indentation, as below:
L
KHN = l2 Cp
L = Load applied (kilograms)
1 = Length of long diagonal of indentation
Cp = Constant for each individual penetrator
Shore Scleroscope Whenever it is desired
to measure the hard-
ness of a specimen of large size or irregu-
lar shape (that cannot be placed in a Rock-
well, Brinell, etc., hardness tester), one
generally resorts to a rebound type of test-
ing device, such as the scleroscope. While
 18
Fig. 1:16. ScleroscopeHardness Testing Instru-
ment.
A diamond- t.ipped "tup" falls in the
glass tube; the height of a rebound
is observed and converted to hard-
ness. The instrument shown here
is attached to a stand but can be
removed for manual. holding when
testing large pieces.
(Courtesy Shore Instrument & Mfg. Co., Inc.)
the scleroseope has lost much favor in re-
cent years, it still enjoys a considerable
field of usefulness if properly employed.
The scleroscope test consists of placing the
open end of a hollow calibrated glass tube
on the object to be tested. A small dia-
mond- tipped hammer inside the tube is
drawn to the top of the tube where it is re-
leased and allowed to fall to the surface to
be tested. The hammer rebounds from the
surface, the height of the bounce being
noted and used as the measure of hardness.
The higher the hammer rebounds, the harder
the piece being tested. A Shore Scleroscope
is being used in Fig. 1:16.
The success of the rebound type of hard-
ness test· depends strongly ¥pon the skill of
the operator, since the instrument must be
held plumb (absolutely vertical) so that the
true rebound height of the hammer is not
reduced through frictional losses caused by
the hammer riding against the side of the
tube. A further difficulty lies in the fact
that the height of the rebound must be meas-
ured by the qUick eye of the operator. This
difficulty is eliminated by a second Shore
instrument deSigned as the "D" model,
which automatically measures the height of
the rebound. It has, however, not as much
popularity as the previously mentioned "e"
model.
When using the scleroscope, care must
be exercised in comparing readings taken on
one type of material with readings taken on
another type. For instance, a piece of rub-
ber would give as high or higher readings
than a piece of steel, and yet, by any other
standard, the rubber could not be considered
as hard as steel.
The scleroscope test has the great ad-
vantage of being usable on practically any
size or shape of specimen, and it does not
noticeably mar the surface being tested.
The hardness tests briefly described
here are the ones most commonly used in
the United States, although many other help-
ful and accurate types are used to a more
limite'd extent. Most of the values from the
hardness tests can be converted with reason-
able accuracy from one type to another. Fig.
1:17 shows the comparative equivalent ~d­
ness values for the standard Brinell test and
the Rockwell test using the "C" scale. The
hardness numbers from other testing sys-
tems are equally easy to convert; the relia-
bility of the conversions varies with the dif-
ferent tests and they should not be considered
as accurate in all cases but rather as ap-
proximations.
Hardness testing is useful in the evalu-
ation of a given steel in several ways. Prob-
ably one of the most useful aspects of hard-
ness is the fact that anempirical relationship
exists between hardness and tensilestrength.
This relationship is quite proportional, as
shown in Fig. 1:18, and while not always 100%
correct is quite useful in many instances.
Since the tensile strength is inversely re-
lated to the ductile properties of elongation
and reduction of area, we may also say that
hardness is often an indication of elongation
 19

60
(1)
c ""
..-.'-
" "
()
~'
en 50
:u /'
/
/'

~ 40 /
.:.::: V
() /

C;
o
30
/v
UJ
Q)
c:
/
~
c 20
,/
::c ,,
'/

10 :'1
100 200 300 400 500 600
Brine II Hardness Number
Fig. 1:17. Relationship Between Brinell and Rockwell Hardness Values for
Steel.

and reduction of area, keeping in mind that
the relationship is not always a precise one
but more a "rule of thumb."
These relationships are useful in that
hardness testing equipment, as compared to
tensile testing equipment, is more readily
available, as well as relatively portable.
The hardness test does not require a ma-
chined specimen, and the actual testing is
quicker and less expensive than the tensile
test. In many cases, it is possible to take
a hardness test on a part and not destroy it,
whereas to machL~e a tensile specimen from
the same part would require the destruction
of all or a portion of the part to be tested
and, in all probability, render it useless.
The 'hardness test can therefore give us
a quick and economical evaluation of the
tensile properties of a steel and, in the case
of a finished part, is the only possible means
of evaluating the tensile properties, short of
destroying or damaging the piece.

c
(l)

()
GOl
Cf)

~o
(l)
50~
~
~
()
0 40
a:::I
UJ
W
Q,)
c: 30
-0
~
0
:r::
20~---'--------'----_--I-_~----'1..-.-----I-_-"'-----I

100 150 200 250 300 350

Tensile Strength - 1000 PSI
Fig. 1:18. Approximate Relationship Between Hardness and Tensile Strength
for Steel.
 20
From the preceding it might appear that
by the use of hardness testing, the tensile
test could be dispensed with. Such is not the
case, since the relationship of hardness and
tensile properties is not a precise one for
all steels. Furthermore, basic considera-
tions in the hardness test allow for a certain
amount of testing error and this error would,
of course, be carried along or increased on
any conversion. It is also true that the hard-
ness test cannot give us a reliable estimate
of the elastic limit and yield strength. The
hardness test is, therefore, used as a sup-
plementary test, and for a really accurate
and reliable evaluation of the tensile prop-
erties the tensile test must be used.
The hardness test can be used as a means
of evaluating the general. condition of a steel
whether it is hard or soft. This evaluation
is useful in connection with many steel form-
ing operations. In an operation such as cold
bending links of a chain, it is imperative
that, among other factors, the hardness and
strength be low for ease of forming. In such
a case, the hardness tester is used as a con-
trol tool. In machining operations, hardness
is of vital importance and is used as an in-
direct measure ·of machinability. As a gen-
eral rule, a hard steel is more difficult to
machine than a soft steel, and when the hard-
ness of steel exceeds a hardness number of
400 Brinell, it becomes very difficult to ma-
chine. A rough evaluation of the relation-
ship of hardness and machinability may be
made as follows, assuming carbide cutting
tools will be used.
Brinell Ease of
Hardness Number Machinability
250 easy
300 not difficult
350 moderately difficult
400 difficult
Over 400 very difficult
When considering the above relationship,
it must be borne in mind that machinability
is a very complex property and is strongly
influenced by many factors other than hard-
ness. The relationship shown is subject to
numerous exceptions, depending upon the
type of steel, its heat treatment history, the
conditions of machining, etc. A notable and
common exception is very soft steel of the
nickel- bearing type, which is very difficult
to m~chine on account of its "gummy" na-
ture. In similar fashion, hardness is used
as a rough measure of a steel's resistance
to wear, a hard steel being considered more
resistant to wear than a soft steel, although
again it should be emphasized that in a sub-
ject as complex as this, many factors are
involved and numerous exceptions exist to
the generality.
EFFECT OF TEMPERATURE ON THE
MECHANICAL PROPERTIES
If, instead of making a tensile test or
performing a hardness test at room temper-
ature, we heat the specimen to some ele-
vated temperature and then perform the test,
we find that the steel is softer and weaker at
the elevated temperature than it was at room
temperature. This change in the mechanical
properties as temperature increases is il-
lustrated in Fig. 1:19.
It is also seen from Fig. 1:19 that as the
tensile strength is reduced, the ductile prop-
erties of elongation and reduction of area
are increased. This means the steel is not
only easier to form (less strength) but also
becomes able to withstand greater plastic .
deformation without rupturing. This is why
the major forming operations of rolling and
forging are performed at elevated tempera-
tures where mechanical strength is low and
plasticity high.
Creep Temperature affects steel in an-
other important manner. If a
load slightly less than the elastic limit is
applied at room temperature, the steel will
elongate and, when the load is removed, will
return to its original length. This elastic
behavior will continue regardless of the
number of loadings and regardless of how
long the load remains in a single loading. If,
however, an identical piece of steel is heated
to some slightly elevated temperature and a
load is applied, it is found that after a length
of time the specimen will assume a perma-
nent deformation even though the stress ap-
plied is below the elastic limit determined at
the testing temperature, using the relatively
rapid tensile test. This phenomenon is known
as creep, and while not fully explainable
with our present range of knowledge, it is
measurable and very important in steel ap-
plications where elevated temperatures are

N
c 120,....-~~-r--.,...-..,..--.---.--.....-~-.,.-......--.--~--r----....-.....
CD
~
<X:
-...
o
90
60
30
O-----------...."...-----"-.l..-..r..--....-----I.----I~""--..&--..L-....
300
Fig. 1:19. Effect of Increasing Temperature Upon the Strength and Ductility
of a Plain O. 15% Carbon Steel.
involved and deformationis not permissible.
In some instances, a limiting creep exten-
sion of 0.1% in 10,000 hours maybe allowed.
As noted, time is an important factor in
consideration of creep. For example, the
flanges of steam lines are held together with
bolts which are exposed to heat. The bolts
are also under considerable stress, since
they hold the pipe together. If these bolts
should deform, or creep, at the temperature
involved, they would not hold the steamlines
tightly together, and leaks would develop.
To test the tendency of a metal to creep
under given conditions, ~ specimen is held
at constant temperature and subjected to a
constant tensile load for a suitable length of
time. The load causes the specimen to
elongate gradually, and this elongation is
measured over a period of time which may
range from hours, weeks, or months to
years. From these data, the creep behavior
of steel under varying service conditions
may be forecast. Fig. 1:20 shows an ac-
cepted method of creep testing.
MECHANICAL PROPERTIES UNDER
IMPACT-LOADING
Static Loading Whenever a structure or
piece of machinery is de-
signed, the designer must seriously consider
the question of the type of loading to which
21
700 1100 1500 1900
Temperature 0 F
the various components will be subjected.
The possible types of loading are static and
dynamic. To understand the difference be-
tween them consider first a work bench upon
which some heavy parts have been placed
for storage. The load does not change in
value nor does it move. It exerts a con-
stant force on the bench and in one direction
only. The bench may then be said to be
statically loaded.
Dynamic Loading In contrast, consider
the piston rod of a
single-headed pump which, by means of a
suitable valve arrangement, is designed to
displace water on both the front and return
strokes. On the front stroke, the rod pushes
the piston head against the water, forcing it
out of the cylinder. The rod is therefore
loaded in compression. On the return stroke,
the rod pulls the head againSt the water and
the rod is loaded in tension. This type of
loading is dynamic loading. Dynamic load-
ing includes not only those conditions where
the nature of loading changes alternately,
that is, from tension to compression and
back to tension, etc., but also those types of
loading in which there are frequent and/ or
rapid changes in the amount of load.
Shock Loading A further type of dynamic
loading is shock loading.
To distinguish shock loading from the normal
22

Fig. 1:20. Typical Creep Testing Apparatus.

The operator is measuring the amount of creep through a
window in the small furnace surrounding the test specimen (1) ·
Tension is applied through the overhead lever arm (2) by
means of weights (3).
(Courtesy Bethlehem Steel Co.)

cyclic loading, consider the automobile.
When the car is in motion, many moving
parts in the engine are subject to rapid
changes, not only in the amount of load but
also in the direction of load. This is normal
cyclic dynamic loading. In contrast, consid-
er the loading which occurs when the car
hits a hole in tile road or rams a fixed ob-
ject. This type is shock loading.
The behavior of the different types of
steel under certain types of loading varies
drastically. It is, therefore, necessary to be
able to predict their behavior under varying
service conditions.
The evaluation of steel under shock load-
ing is done to a large extent with the impact
testing machine. This device consists es-
sentially of a frame from which is swung a
weighted pendulum. The pendulum is raised
to some predetermined height H and, when
released, will swing through an arc and
raise to a certain height h, as in Fig. 1:21a.
The height h is somewhat less than the
original height H because of frictional and
other losses. If we place a specimen of
steel in the path of the swinging pendulum,
the pendulum must break the piece of steel
if it is to continue in its natural arc. The
amount of energy necessary to break the
 23

Free Swing Swing Reduced

In Fracturing Specimen
Fig. 1:21. How (Izod) Impact Values are Determined.

specimen is absorbed from the pendulum,
thus reducing the height to which the pendulum
can rise, as in Fig. 1:21b. The difference
between the height to which the pendulum can
rise unimpeded (h) and the height to which
the pendulum rises after breaking the spec-
imen (h-i) represents the energy absorbed
in breaking the impact specimen. This en-
ergy i is usually reported in foot- pounds
and is called the impact strength of the steel.

10 Iftlll.
SQUARE
., r

CHARPV SPECIMEN
IZOO SPECIMEN

DIRECTION OF
BLOW

OIREcnON OF
BLOW

CHARPY TEST
IZOD TEST

Fig. 1:22. The Two Major Impact Test Specimens, Charpy and Izod.
24
The two most commonly used impact
tests in this country are the Izod and Charpy
impact tests, which differ mainly in the
manner in which the bars are broken. Fig.
1:22 shows the specimens used in these tests
and the manner of breaking them. It will be
noted that the specimens are notched. This
reduces the effective cross section at the
notch and concentrates the energy of the
blow at that point. In this manner the loca-
tion of the break is controlled. Various
types of notches are employed but the pur-
pose in any case is as described. From the
information secured with this test it is pos-
sible to predict intelligently the vehavior of
steel under shock type of loading.
The susceptibility of steel to "brittle
fracture" may also sometimes be predicted
from impact testing, particularly if the
testing is done at low subzero temperatures.
Brittle failure is described as failure which
Fig. 1:23. Closeup of a Typical Fatigue Fa.ilure ShOWing Characteristic
Radiating Type of Fracture.
occurs without warning as compared to duc-
tile failure, which exhibits prior signs of
warning such as buckling or twisting. The
distr'essing aspect of brittle failure is that it
can occur under essentially static conditions
and in steels normally considered not sus-
ceptible to this abrupt type of failure. In
any case, a single isolated impact test is of
little or no value in predicting the behavior
of steel in service. A series of tests run at
different tempera~res is usually neces-
sary for any reasonable appraisal of the
qualities of a steel. This testing procedure
will be described more fully in a later lesson.
Fatigue Occasionally, service failures
are encountered in parts which
are subjected to repeated cyclic stress.
These failures occur in spite of the fact that
the stress applied may be considerably be-
low the tensile strength of the steel. Fig.
1:23 shows a typical failure with its
(Courtesy Bethlehem Steel Co. )

characteristic 'radiating fracture. This type
of failure is brought about by a condition
called fa~e, which is defined as the tend-
ency of a metal to break under conditions of
repeated cyclic stressing considerably below
the tensile strength of the material.
The cause and appearance of fatigue fail-
ure may be understood when we consider a
bar which is loaded, as in Fig. 1:24. As the
bar bends, the surface metal at the bottom
of the bar is in tension, and that at the top
is in compression. When the bar is turned
180 degrees, this condition is reversed so
that the surface metal which originally was
in tension is now in compression, and that
which was in compression is now in tension.
Thus, for every complete turn of the bar all
the surface metal will be at one time in ten-
sion. If there is a small notch (crack or im- .
perfection) in the surface of the steel bar,
and the surface is in tension, the stress will
be concentrated at the notch, as indicated in
Fig. 1:25. The concentrated stress at the
notch will be many times the stress away
from the notch and may be raised to such a
point as to actually exceed the tensile
strength of the steel at this localized point.
Thus, a minute failure may be started at the
scene of the stress concentration, and every
time the bar revolves and places this point
in tension, more localized failure occurs,
In
w
Fig. 1:24. Effect of Loading on the Distribution of Stress as it Pertains
to Fatigue.
25
resulting in a gradual growth or propagation
of the original small failure. This propaga-
tion continues until the failure becomes
large enough to reduce the effective cross
section considerably. The remaining solid
metal then becomes severely overloaded
and the bar will break in abrupt fashion. The
development of a fatigue failure is readily
evident in Fig. 1:23 as indicated by the pro-
gressively larger circles radiating from the
origin of failure.
In order to evaluate the ability of the
steel to resist fatigue, a test specimen is
loaded as in Figure 1 :24. The specimen is
then revolved until failure occurs. This pro-
cedure is repeated, using progressively
smaller loads until at some point of loading
failure will not occur, regardless of the
number of cycles (up to a practical limit of
several million) to which the specimen is
subjected. If the various stress values and
the corresponding number of cycles reqUired
to produce failure are plotted, a curve such
as that in Fig. 1:26 will result. This dia-
gram is called an S-N diagram (stress -
number of cycles).
From this diagram it is seen that at some
limiting stress, A, no failure will result re-
gardless of the number of cycles of stress
reversal. This limiting stress is called the
 26

"- r
''" fr.
"fur.
,~r
~

t----------I~ •

.....--------..... -... ~
/

(0) Stress (b)

Lines

High Concentration Of Stress

At Root Of Crack
Fig. 1:25. Schematic Drawing Showing Block of Steel Under Tension.
(a) With normally uniform distribution of stress
(b) With concentration of stress caused by presence of notch

: Endurance Limit
Q)
'-
+-
en

Number Of Stress Reverse Is Or Cycles

•
Fig. 1:26. Typical S-N Diagram Showing How EndUrance Limit is Evaluated.

endurance limit or fatigue limit and is the
maximum stress that a metal can withstand
without failure, regardless of the number of
cycles of stress reversal that may be ap-
plied.
The endurance limit may also be esti-
mated from the tensile strength, and it is
true that for steels of 50,000 to 80,000 psi
tensile strength, the endurance lim it is
about half of the tensile strength. For steels
above 200,000 psi tensile strength, the av-
erage endurance limit is below this 50%
value. If notches are present, these ratios
are considerably reduced, and with severely
notched specimens the above ratio of
 en
(l.
130
o 50% Ratio
o
o
-100
...-e
I

:J
CD 70
(.)
c
e
::3
-c
c
lLJ

,01li.ilEEE3111.
40 80 120 160 200 240
Tensile Strength - 1000 PSI
Fig. 1:27. Relationships of Tensile Strength and Endurance Limit, and the
Effect of Notch Upon that Relationship.

Test

Fig. 1:28. Typical Fatigue Testing Machine.

(Courtesy Baldwin-Lima-Hamilton Corp-)
28

endurance limit to tensile strength does not
hold at all. This ratio is referred to as the
"endurance ratio." The general relationship
of endurance limit and tensile strength and
the effect of notch on this ratio are shown in
Fig.. 1:27. It may also be noted here that
since hardness is proportional to tensile
strength, it too is proportional to the en-
durance limit. A typical fatigue testing
machine is illustrated in Fig. 1:28.

ANSWERS TO SEI.;F CHECK QUIZ PROBLEM

~~
V
V
501--+---~~~~V~,,:J--r----J--+----r----r--t---+---+----r-'t-
~ e
.-: )V a =proportional limit (300~O psi)
en
.:It: V
40 J--~-'-"-+---+--+-If---+----+--+---;-
b = upper yield point (34000 psi)
-

g
~ 30~P
al~~~ c = lower yield point (32000 psi)
-
f- d =ultimate tensile strength (56 000 pSi)
en U~, e = fracture stress (48000 psi)
201--l~-4---+----J--+--4---+-----I-+--+I_t--+---t---+--;--r--r
(.
I O&-nJ~~-+----I----I---+--+--+--+---+--t-+--t---r--r--t--t-

of c"
I

'"
o 2 3 4 5 6 7 8 9 10 II 12 13 14 15 16 17 180

STRAIN, 0.001 in/in

f. % elongation = final l~~~r~gin:length x 100

origin en

- 2.360 - 2.000 100 = 18 01

- 2.000 x 70

g. 01
70 re duc ti on 0 f area = original. area-final
. al area x 100
arlgm area

- .250 - .164 100 - 34 4 01

- .250 x - · 70
= --=c'.;......-=--
.. ' _

". .. BASIC FUNDAMENTAL PRINCIPLES USED IN CALCULATING HARDNESS NUMBERS.

1.BRINELL BHN-== Load in Kilograms

Area. of Impression in mm2

L
BHN:=c - - - - - - - _

~o (0-'10 2 -d 2 )
Ls::Load, Kg.
D=Ball diameter, mm.
d=Impression diameter, mm.
7f =3.1416

2. DIAMOND PYRAMID DPH (DPN) =-= Load in Kilograms

Area of Impression in mm 2
1.8544 L
DPH(DPN)~------
d2
L=Load, Kg.
d =Average diagonal, mm.

3. :KNOOP KHN == Load in Kilograms

Projected Area of Impression in mm 2

KHN= __L_ _
f2 xCp
L=Load, Kg.
( =Long diagonal, mm.
Cp=A constant, 0.07028

4. ROCKWELL Rockwell numbers for the diamond cone ("Brale") indenter are read from
a dial indicator. The Rockwell number is indicated by 100 units minus the
number of units of depth of indentation (each unit of depth being equiv-
alent to 0.00008").

When using the 1/16" diameter ball indenter J the Rockwell. number is in-
dicated 'by 130 units minus the number of units of depth of indentation.

NOTE: ROCKWELL IS NOT CALCULATED FROM A LOAD/AREA RELATIONSHIP

 ASM Intemational@
MATERIALS ENGINEERING INSTITUTE
~~.
C10L1, C11L10, and C41L2
~~ Test Series F Steel and Its Mechanical Properties
Page 1 of2

PLEASE PRINT YOUR NAME AND ADDRESS BELOW RETURN TO: ASM Intemationafil
MATERIALS PARK, OH 44073-0002
NAME REGISTRATION NO. _

STREET _

CITY STATE ZlP ' DATE _

MULTIPLE CHOICE

Place the appropriate letter (A, B, C, D) in the box. (Only one answer is correct and
each question has a value of10 points.)

1. Stress values in the tensile test are ratios of load to cross-sectional area and are
expressed as: D
(A) foot-pounds.
(B) kilograms per millimeter.
(C) pounds per square inch.
(D) inches per inch.

2. Figure 1:27 indicates that for most polished ductile steels, the endurance limit is
approximately one haIfthe: D
(A) Brinell hardness.
(B) tensile strength.
(C) elastic strain.
(D) Charpy-impact value.

Using the formula KHN =U12 x Cp, the Knoop hardness for a steel part tested with
3.
a 0.500 kg load is 588 kglmm2 • The constant for the Knoop indenter is 0.07028. The
length of the long diagonal of the indentation would be approximately:
D
(A) 0.121 mm.
(B) O.110mm.
(C) 0.155 mIn.
(D) 0.012 mrn.

4. When conducting the Shore Scleroscope hardness test, the operator measures:

(A) the indentation with a microscope.

D
(B) the height of the rebound of the "tup."
(C) the depth of the indentation.
(D) the impact energy in foot-pounds.
 ASM International@
MATERIALS ENGINEERING INSTITUTE
C10L1, C11L10, a~d C41L2
Test Series F Steel and Its Mechanical Properties
Page 2 012

5. A tensile test bar with a 51 mm (2 in.) gage length measured 62 mm (2.44 in.) after
fracture. The percent elongation would be: D
(A) 44%. (C) 24.4%.
(B) 12.2%. (D) 22%.

6. Figure I :22 indicates that the Charpy and Izod V-notched impact tests differ mainly:

(A)
(B)
(C)
in cross section of the specimens.
in position of the specimen and notch in relation to direction of blow.
in that energy is not absorbed in the Izod test.
D
(D) in that a pendulum-type machine is not used for the Charpy test.

7. The Rockwell table onpage 16 indicates thatto perfonn aRockwell hardness test on
the B scale, one would use a:

(A) 1.5875 mm (O.0625-in.) diameter baIl and 100 kg load.

(B) 1.5875 mm (O.0625-in.) diameter ball and 150 kg load.
D
(C) Brale and 60 kg load.
(D) Brale and 100 kg load.

8. Tensile strength = load/area. Areaofacircle = (3.1416xdiameter2)/4. The maximum

load supported by a 9.07-mm (O.357-in.) diameter tensile test specimen was 5,455
kg (12,000 lb.). The tensile strength for this steel would be: D
(A) 689 MPa (100 ksi). (C) 83 MPa (12 ksi).
(B) 231 MPa (33.5 ksi). (D) 828 ~a (120 ksi).

9. Using the fonnula DPH =(1.8544 x L)/d2, the DPH for a hardened steel part tested
on the Vickers machine using a 20 kg load with the average length of the two
diagonals measuring 0.22 mm would be approximately:

(A) 440 kg/mm2• (C) 566 kg/mm2 •

D
(B) 766kg/mm2 • (D) 866kglmm2 •

10. Figure 1: 17 indicates that ifa Rockwell hardness test on apiece ofsteel gave a value
ofHRC35, a comparable Brinell hardness value would be approximately:

(A) 275kg/mm2 •
(B) 4OOkg/mm2•
(C) 325 kglmm2•
(D) 250 kglmm2• D r---
 ASM International~ is a Society whose
mission is to gather, process, and
disseminate technical information.
ASM fosters the understanding and
application of engineered materials and
The Materials their research, design, reliable manu-
Infomation Society
facture, use, and economic and social
benefits. This is accomplished via a
unique global information-sharing
network of interaction among members
in forums and meetings, education
programs, and through publications
and electronic media.

Printed on DocuTech February 1994

(-
 The Materials
Information Society

Microstructure and
Mechanical Properties
by
George F. Melloy, B.S.

Course 10
Lesson, Test 2
Course 11
Lesson, Test 11
Course 41
Lesson, Test 3
 Metric Conversion Factors
To convert from To Multiply by

in. nun 25.4

in. m 25.4 x 10-3
mil J.U11 25.4
J.1in. J.U11 25.4
in.2 m2 6.45 x 1()-t
in. 3 m3 1.64 x 10-'
ft. m 3.048 x lerl
ft. 2 m2 9.29 x 10-2
ft. 3 m3 2.831 x ler2
oz. g 2.834 X 101
lb. kg 4.536 x 10-1
Btu J 1.054 x 1()3
Btullb.• of Jlkg· K 4.18 X 103
Btulft. • hr • of W/m·K 1.730
10 inJin.JOF 10"' mmlIIlIIlFC 1.8
psi Pa 6.895 x 1()3
psi kPa 6.895
ksi kPa 6.895 x 1()3
ksi MPa 6.895
ksi. in. 11l MPa-mlll 1.099
ksi~in. MPa·...Jm 1.099
oz.f gf 28.4
lbf kgf 4.536 x 10-1
Ibf N 4.448
lbf· ft. N - m (or J) 1.356
Ibflin.:Z kgf/cm2 14.223
Ibflin. 3 kgf/m3 2.768 x 1()C
Ib.lft. 3 kglm3 16.019
Ib./in. 3 g/cm3 2.768 x 101
Ib./in. 3 kglm3 2.768 x 10C
gal (U.S. liquid) L 3.785
gal (U.S. liquid) m3 3.785 x 10-3
IbJgal gIL 119.826
ft./gal mIlL 748
of °C (OF - 32)/1.8
of K (OF + 459.67)11.8
°C of (OC· 1.8) + 32
°C K °C + 273.15
K °C °C-273.15

Multiple and Submultiple Units

1018 exa 10-1 deci
101' peta 10-2 centi
1012 ••••••••••••••••••••••••••• tera 10-3 •••••••••••••••••••••••••• milli
1(JJ giga 1()-6 micro
1()6 mega 10-9 •••••••••••••••••••••••••• nano
1()3 kilo 10-12 pico
1()2 •••••••••••••••••••••••••• hecto 10-1' ••••••••••••••••••••••• femto
101 ••••••••••••••••••••••••••• deka 10-18 ••••••••••••••••••••••••••• atto

Abbreviations
J joule m .•............•.•..... meter Pa pascal
kgf kilogram force mm ........•.•. millimeter K kelvin
L liter N newton w wan

by
George F. Melloy, B.S.

e
19n American Society for Metals
This material may not be reproduced in whole or in part in any form whatsoever. Fully protected by copyright.
 Microstructure and
Mechanical Properties
The fact that steel can possess a wide
range of useful mechanical properties is of
extreme economic importance. This ·is
clearly illustrated in the railroad industry,
for example. To move a train from one
location to another we use a locomotive
which has the ability to pull a given total
load. This load is composed of the weight
of the cars or deadweight, and the weight
of the freight being transported, or the
payload. If the freight cars are made of
low strength steel, the structural members
must be relati vely large to support the'
load, and hence heavy. If a freight car is
made of high strength steel, the structural
members can be relatively small and the
car will be lighter as well as stronger.
This means that the amount of freight can
be increased, the ratio of payload to dead-
weight becomes greater, and more profit
can be realized per trip. Similar situa-
tions concerned with the mechanical proper-
ties of steel exist in the other industries.
Effect of Alloy- If we are to make in-
ing Elements telligent use of the
varying m e c han i cal
properties of steel, we must learn the
Table 2: 1. Effect of Carbon Upon Tensile Strength (Normalized)
Percentage of Carbon
of the Steel 21 -Inc
. h
0.30 77,500
0.40 88,250
0.50 111,500
0.60 113,000
0.80 150,000
scientific reason for these properties and
how to reproduce and control them. For
instance, if we desire' to make a lighter and
stronger freight car, it would be useless to
specify a stronger steel for its structural
members unless we know how to make the
steel stronger.
A large percentage of the steel produced
is supplied' to the consumer in the "natural"
or as-forged or as-rolled condition. In the
rolling or forging operation, steel is h.eated
to roughly 19000 F to 2200 o F, where it is
still solid but quite weak and plastic. While
in this condition the steel is forced through
shaped rolls or hammered between dies to
some intermediate fornl (such as a bar)
which will later be machined or formed to
some specific shape such as a gear, etc.
After the forging or rolling operation, the
steel is frequently allowed to cool in air to
room temperature. Steels of the type proc-
essed in this manner will usually have mod-
erate strength and hardness. For- plain
carbon steel, the as-rolled or as-forged
strength will vary with the anloW1t of carbon,
as shown in Table 2: 1.
Tensile Strength, psi,
for Diameter of Bar Processed
I-inch 2-inch 4-inch
77,500 74,000 72,500
85,500 84,250 83,500
108,500 106,250 100,000
112,500 110,000 108,250
146,500 141,000 134,000
 2
The values shown in the table are not
actually those of the steels in the as-rolled
or as-forged condition but are for steels
which have been normalized. Normalizing
consists of heating the steel to a "red-hot"
condition (to temperatures substantially
below those usually used for rolling or
forging) and then allowing it to cool in air
to room temperature. Normalizing is one
of the heat treating operations and makes
more uniform (homogenizes) the micro-
structure found in steels in the as-forged
or as-rolled conditions. The mechanical
properties which result from the normaliz-
ing treatment are approximately the same
as those obtained in as-forged and as -rolled
steels. It, as well as the other heat treat-
ing procedures described briefly in this
lesson, will be considered more fully later
in the course.
From Table 2: 1, it is apparent that the
strength of the steel is a function of
the amount of carbon present. Alloying
elements also have a similar effect, as
shown in Table 2:2. In this table, there
are listed a series of tensile strengths.
which resulted .when various alloying ele-
ments were added to a plain 0.40% car-
bon steel. These steels were normalized,
as were those in Table 2: 1.
It is apparent that the strength obtained
with any given carbon or alloy content de-
creases as the diameter of the bar is in-
creased. This effect is known as the effect
of mass. It is important for the student to
recog~ize that in addition to chemical
composition, mass (size) has an important
effect upon the hardness and strength which
can be attained in steel.
From Tables 2: 1 and 2:2, it is apparent
Table 2:2. Effect of Alloy Conten~ on the Tensile Strength
of Normalized 0.40% Carbon Steel
% Alloy Total
Alloy
Mrl Ni Cr Mo Content
1.77 1.77
0.90 1.21 0.62 2.73
0.40 0.94 0.21 1.55
that, the tensile strength of as-rolled or as-
forged steel can be increased considerably
by adding relatively small amounts of carbon
and other alloying elements. It is also
apparent that some alloying elements are
more effective than others in causing an
increase of tensile strength. For instance,
in a 0.40% carbon steel a total alloy con-
tent of 2. 73% resulted in a tensile strength
of 129,250 psi (1" diameter), while in a
steel of the same carbon content a total
alloy content of 1.55% (manganese plus
chromium plus molybdenum) resulted in a
tensile strength of 148,000 psi in a bar of
the same ·size, an increase of about 15%.
From these results it would appear that
the solution to securing a stronger steel is
quite simple, i. e. , add more carbon and/or
alloy to the steel. Unfortunately, the alloy-
ing elements are relatively costly, and to
develop a high strength steel by alloying
alone would be quite expensive. The econ-
omics involved would defeat the purpose of
such steels except for unusual circumstances.
Quenching and Fortunately, the strength
Its Effect of steel can be increased
through another method.
From observation of a blacksmith, or read-
ing the technical literature, we know that if
a piece of soft steel is heated to a "red hot"
condition and then plunged into a bucket of
cold water (or other liqui~ quenching medium)
the steel will become both harder and
stronger. The temperature to which the
steel is heated prior to rapid cooling in
liquid is usually lower than that used in
normalizing. This procedure is called
quenching, and its effect is demonstrated in
Table 2:3, where the relative tensile
strengths of steels of varying carbon
Tensile Strength, psi,
for Diameter of Bar Tested
! -inch I-inch 2-inch 4-inch
132,000 121,250 120,000 120,000
148,500 129,250 119,250 118,000
148,500 148,000 140,000 117,000
 3

Table 2:3. Effect of Quencning on the Tensile Strength of Steel

(I-inch Diameter Bars)

Carbon Content of Steel Tensile Strength in psi

% Before Quenching After Quenching

0.15 61,500 75,500

0.20 64,000 89,000
0.30 77,500 122,000
0.40 65,500 130,000
0.55 110,000 162,000

content are shown before and after thermal
treatment.
From these figure~ it is evident that the
procedure of quenching results in a gain of
strength which may well overshadow the ef-
fects of alloying alone. For instance, a
steel of only 0.40% carbon content can, after
quenching, have a tensile strength slightly
greater than a normalized O. 40% carbon
steel to which alloying elements up to
2.73% have been added. It should be noted,
however, that the gain in strength realized
by quenching is not limited to carbon steels
only · The alloy steels will also show a
large increase in tensile strength if they
are quenched. In Table 2:4, the hardness
of a plain 0.30% carbon steel is compared
to that of a similar carbon steel containing
some alloy. B.oth steels are in the as-
quenched condition.
It should be noted that the hardness
data listed in Table 2:4 represent the high-
est values attainable with the steels and
sizes processed. In other words, quench-
ing produces the highest hardness and
strength values attainable. No other heat
treatment can produce higher values.
Thus, both the plain 0.30% carbon steel
and the alloy steel containing 0.30% carbon
achieved the same maximum hardness,
Rockwell C 50. No higher hardness value
can be secured with a steel containingO.30%
carbon, regardless of the alloying content
or thermal treatment. For each carbon con-
tent which may be present in steel there is
a related maximum attainable ha.rdness.
This hardness value may not be realized in
commercial quenching because of mass ef-
fect, but if a small enough piece is processed
and the quenching conditions are severe

Table 2:4. Comparative "As-Quenched" Hardness of a Plain Carbon and Alloy Steel
(Water Quenched)

Percentage Composition Hardness Rockwell C

Location
of the Steel
of Diameter of Bar Processed
C Cr Mo Sample 1· h
~-lnc I-inch 2-inch 4-inch
0.30 Surface 50 46 30 20
1/2 Radius 50 23 13 7
Center 23 21 9 3
0.30 0.91 0.20 Surface 50 50 47 45.5
1/2 Radius 50 50 32 25
Center 50 44 31 24.5
 4
enough, it can be attained. This fact leads
to one of the very important rules of heat
treatment: maximum attainable hardness
is a function of the carbon content of the
steel. Alloy content does not contribute
to the maximum attainable hardness.
A t this point, the student will consider
Table 2:4 and observe that in larger
size categories the alloy steel is harder
than the plain carbon steel both at the
surface and interior positions. This is a
fact, but it does not contradict the previ-
ous statement that maximum attainable
hardness is a function of the carbon content
independent of the alloy content. It will be
noted that, for the alloy steel, nowhere
has a hardness in excess of Rockwell C 50
been realized. What has happened is that
the alloy content has reduced the effect of
mass. The presence of the alloy content
in the steel has allowed it to approach more
closely the maximum attainable hardness
indicated by the carbon content. The abil-
ity of steel to harden throughout its cross
section is a field of study in itself and will
be more completely considered later in the
course.
It should be noted that in the vast ma-
jority of cases, steel is not used after
quenching without being subjected to further
thermal processing. In the "as quenched"
condition, steel is very highly stressed and
quite brittle, conditions not conducive to
good service life.
Prior to placing a hardened part in
service it is usually tempered. Tempering
consists of heating the steel to a relati vely
low temperature (considerably below that
used for hardening) and then air cooling it.
Such procedure reduces the hardness slightly
but renders the steel a great deal more
ductile and eliminates a large portion of
the stress incurred during the quenching
operation.
Having established the fact that high
strength can be developed in a steel byadd-
ing carbon and then quenching, let us turn
to the problem of utiliZing such a material.
In order to be of value to us, the steel
must be in some useful form such as a
gear, an axle, etc. This generally means
that ,the steel must be machined to the de-
sired shape. As noted in our discussion of
mechanical properties in Lesson 1, the
stronger and therefore harder a piece of
steel is, the more difficult it is to machine.
Thus, it would appear that the designer is
"between the devil and the deep blue sea".
If he hardens the part to gain maximum
strength, he cannot machine it, and if he
can machine it, the part may not have suf-
ficient strength for its intended application.
In some grades of steel, the as-forged or
as-rolled hardness exceeds the bounds of
economical machining, apparently adding
to the problem.
Annealing and Fortunately, there is an
Its Effect escape from this seem-
ingly impossible situa-
tion. If steel is heated to a "red hot"
temperature (somewhat below that used for
hardening but higher than that used for
tem.pering) and, instead of being quenched,
is very slowly cooled to room temperature,
the tensile strength will be lower than that
resulting from quenching or normalizing.
The steel has been made more ductile and
weaker by a process called annealing and is
now quite readily machined.
Annealing is any process of heating and
cooling intended to reduce the hardness of
steel and restore its ductility. With the
steel in the soft condition it is readily ma-
chined to shape, and after machining may
then be further heat treated and made
strong and hard.
Heat Treatment Thus, for a single
composition of steel a
wide range of mechanical properties can
be secured by the relatively simple pro-
cedure of heating to a "red hot" condition
and then cooling at different .rates. For
example, the mechanical properties of a
0.40% C, 0.94% Cr, 0.21% Mo steel vary
with the thermal tre'atment, as shown in
Table 2:5.
Table 2:5. Variation in Strength and
Hardness by Heat Treatment.
Condition Tensile
of Steel Strength Hardness
Annealed - soft 95,000 psi 197 Bhn
Normalized 148,000 psi 302 Bhn
Quenched - hard 266,000 psi 534 Bhn

Other values of strength could also be
secured by using cooling rates intermediate
to those described. The general term ap-
plied to these procedures of hardening and
softening of steel is called heat treatment.
Heat treatment is defined as an operation,
or combination of operations, involving the
heating and cooling of a metal or alloy in
the solid state for the purpose of obtaining
certain desirable properties.
It should be emphasized that the proper-
ties of steel can be altered at Will, and
the effects of any single heat treatment
operation are not final and irrevocable. If
the steel is hard, it can be made soft for'
machining, and after machining it can again
be hardened. If a mistake has been made
in machining and not detected until after
hardening, the part may be annealed, the
error corrected, and the ·part again hard-
ened to the desired strength.
The value of heat treatment is by now
apparent. What is not apparent is how the
heating and cooling cycles cause the me-
chanical properties to change. If we place
two machined pieces of steel side by side,
one in the hard condition and the other in
the soft condition, no difference would be
apparent to the eye. Since the character-
istics of the steel are vastly different,
something must have occurred internally
during heat treatment to account for the dif-
ference in mechanical properties. The prob-
lem of what occurs during heat treatment
was a troublesome one for many centuries.
Iron was known and used in tools and
ornaments as early as 4000 B. C. The in-
tentional addition of carbon to iron, quench-
ing, and the general advantages of heat
treatment were later developments but were
. employed many centuries before the Chris-
tian era. While it was recognized in early
times that carbon (the oldest intentional al-
loy addition made) was necessary for hard-
ening, the method' by which the steel actually
hardened when quenched was not known.
As a result, the hardening of steel be-
came an art, the secrets of which were
closely guarded. There is a legendary
methcxl of heat treating the fabulollsDamas-
cus sword blades which well illustrates the
mystery and "hocus-pocus" connected with
hardening in these early times. The hard-
5
ening procedure consisted of tying a slave
face down on the sacrificial block of the god
Bal-hal. Then, while the artisan recited a
prayer, the sword was heated in cedar wood
coals until its color was that of "the red of-
the rising sun". The heated sword was
passed through the buttocks of the sla ve
until it became "the color of the purple of
the king". The sw~rd was then tested and
considered acceptable if, without cracking
or breaking, the sla vets head could be lopped
off with one stroke and the sword could be
bent around his body.
With the passage of time and the rise in
the cost of slaves, their use as a quench-
ing medium passed out of the picture, but
the belief that hardness was due to some
mystical property of the material Into which
the steel was quenched persisted. All sorts
of weird concoction..; were used and great
emphasis placed upon the value of water
from specific places~ In 1558, the preacher
Mathesius observed in one of his sermons
that one water would harden steel better
than another and this was the reason that
armor made at Innsbruck was so good. As
late as the early 1900' s, hardening was con-
sidered· an art and a secret craft, and at
Solingen the apprentices had to take an oath
not to leave the country and not to teach
the craft to anyone but their sons.
THE CONSTIT·UTION OF STEEL
Although the basic reasons for the changes
in mechanical properties in heat treatment
were not understood, a useful workingknowl-
edge of how to harden and soften steel was
gradually developed by trial and error. It
was not until the early part of the 20th
century that scientific understanding of the
fundamental changes which occur during heat
treatment was developed. Since these
changes d9 not manifest themselves in the
external appearance of the steel, they must
be internal and invisible to the unaided eye.
From our preceding discussions, it is
established that the basic component of steel
is the element iron. The ability of the iron
to become stronger and more useful is due
to the presence of carbon. The actual
changes in the strength of steel are pro-
duced by a procedure of heating and cool-
ing called heat treatment. It logically
 6
follows that the changes in mechanicalprop-
ertie"s which occur upon heating and cooling
must be due to changes induced in the iron
and carbon. Therefore, if we are to un-
derstand heat treatment, we must under-
stand the effect of heating and cooling upon
iron and carbon, the basic constituents of
steel.
Atomic Ar- In order to accomplish this
rangement aim, it is necessary to
consider iron in terms of
its smallest unit particle, the atom. While
the atom is an extremely small unit, it can
be measured with the assistance of preci-
sion apparatus. For our purpose, it is
convenient to think of each atom as a
small ball of iron, having a definite weight
and a definite attractive force for all neigh-
boring atoms. In metals this attractive
force is such that the atoms arrange them-
selves in definite geometric patterns. The
arrangement of the atoms with respect to
each other is called the crystal structure
or lattice. At room temperature the iron
atoms arrange themselves into what is
known as the body-centered cubic lattice
illustrated in Fig. 2: 1. In this unit of
construction, an iron atom is located at
each of the eight corners of a cube and a
ninth atom is centrally located within the
cube. The attractive force between the
Ferrite
(0<:. Iron)
Fig. 2: 1. Body-Centered Cubic Structure of
Alpha Iron.
Each ball represents an atom of iron.
atoms and their neighbors keeps them
"balanced" at these locations.
Iron is an allotropic element, that is, it
can assume several different lattice modifi-
cations depending upon temperature. The
body-centered cubic modification, commonly
called the alpha modification, exists to a
temperatureOf1674 F. After a tempera-
ture of 1674 F has been ~xceeded, the
atoms in the iron rearrangethemselves into
the face-centered cubiC, or gamma modifi-
cation, shown in Fig. 2:2. Here it is seen
that the same basic cubic structure remains
with an atom located at each of the eight
corners of the cube. However, instead of
a single atom being located in the geometric
center of the cube, there is now an atom
located in the center of each face of the
cube. The gamma modification exists be-
tween 1674 F and 2541 F. A third modifi-
cation, called delta, exists between 2541 F
and 2800 F. If the iron is heated higher
than 2800 F, it becomes molten and is no
longer solid. Fig. 2:11 (page 17) depicts the
changes that occur in pure iron as it cool~
from the molten state to room temperature.
The phase changes, and also the -magnetic
change, cause horizontal i~gs or lags in the
cooling curve illustrating that heat is evolved
when these changes occur. All of the heat
treating operations. we will consider take
place well below 2541 F so that the delta
Austenite
(r Iron)
Fig. 2:2. Face-Centered Cubic Structure of
Gamma Iron.

modification is of no importance to us. We
will be concerned only with the alpha and
gamma crystallographic structures of iron.
The lattice modifications of iron, being
temperature - dependent, are reversible.
A lpha iron changes to gamma iron upon heat-
iog, and the gamma iron changes back to the
alpha form upon cooling. It should be noted
that one unit of the lattice structure is ex-
tremely tiny. In even a small piece of
steel, millions of such units are arranged
row upon row, in three dimensions, to
make up the mass of the steel. The effect
may. be compared to bricks neatly stacked
into a ~rge piI.e , the individual bricks being
comparable to the individual lattice cubes
and the pile of bricks being comparable to
the piece of steel.
Solid Solution The effect of temperature
upon iron has now been
explained. What is the role of carbon dur-
ing these changes? Before considering
this, it is necessary to understand the
phenomenon of solid solution. Weare all
familiar with liquid solutions. When a
lump of sugar is placed in a cup of coffee,
the sugar dissolves and apparently disap-
pears. However, by tasting the coffee we
know the sugar is present in liquid solution.
When iron is molten, it too takes certain
elements into liquid solution in the same
manner. When the iron cools and be-
comes solid, it is still capable of retaining
certain elements in solution. This solution
in the solid stat~ is called solid solution
and is as true a solution as that formed in
the liquid state. Solid solutions are not
unusual in metals. Iron takes many other
metals into solid solution.
Solid solutions can result from solidifi-
cation of a liquid solution, or they can be
formed in the solid state. For instance, if
suitable carbonaceous material is packed
around a piece of pure iron and the temper-
ature raised until the iron is "red hot", the
iron will take the carbon into solution al-
though the iron remains solid at all times
and does not even approach the melting
point. In similar fashion it can be demon-
strated that iron, while hot and solid, can
dissolve numerous other elements and form
a solution with these elements.
The degree of solubility of elements in
solid solution may vary considerably. Some
7
elements may be completely soluble in all
proportions at all temperatures. Some may
be completely soluble at high temperatures
but only partially soluble at lower tempera-
tures, and still others may be soluble at
high temperatures but completely insoluble
at low temperatures.
Carbon is soluble in both alpha and gam-
ma iron in various degrees, depending upon
both the lattice structure and the tempera-
ture. These solutions are, of course, solid
solutions. It is important to note that as
carbon is added to iron, the temperature at
which gamma iron can exist is lowered, as
indicated by the line GS in Fig. 2:3. When
0.77% carbon has been added, the maximum
effect of carbon with respect to loweringthe
temperature at which gamma iron can exist
has been reached.
'600
Alpha Iron
and
Gamma lro"
0.0218%
0.7741f. 1340.6 F
Alpha Iron and Cementite
0.2 12
0.4 0.6 0.8 LO
PERCENT CARBON
Fig. 2:3. Effect of Carbon Content on the
Alpha - Gamma Transformation.
For plain carbon steels, as represented
in Fig. 2: 3, it is. apparent that alpha iron
may exist between room temperature and
some elevated temperature ranging from
1340.6 F to 1674 F, depending on the carbon
content. Alpha iron is capable of dissolv-
ing very little carbon. At room tempera-
ture, only 0.008% C is soluble; at.1340.6 F,
the solubility increases but only to 0.0218%
C. At temperatures above 1340.6 F, the
carbon solubility decreases.
The solid solution of carbon in alpha
iron is called ferrite (after ferrum, the
Latin word for iron; the chemical symbol Fe
8

also derives from the Latin word). Most
steels which are heat treated contain more
than 0.0218% carbon. The carbon in excess
of the small amount held in solid solution
in the alpha iron is present as the inter-
metallic conlpound iron carbide (Fe3 C),
which is usually called cen1entite. The
cementite is present as small particles
more or less evenly distributed throughout
the ferrite. The cementite particles are
clear ly visible under the microscope if the
steel is properly prepared.
When ferrite and cementite are heated above
1340.6 0 F, some of the atoms in the alpha iron
rearrange themselves into the gamma modifi-
cation. Gamma iron can hold much more carbon
in solid solution than alpha iron. As soon as
gamma iron is formed, it can hold up to 0.77%
carbon in solid solution. As the temperature of
the gamma iron is increased above 1340.6F, its
ability to hold carbon in solid solution also
increases and reaches a maximum at 2066 0 F,
where it can hold 2. I 1% carbon. The solid
solution ofcarbon (and other elements) in gamma
iron is called austenite (named for W. C. Roberts
-Austen, pioneer English metallographer).
When austenite forms, it usually does so
at a cementite-ferrite interface. Since
there are many such interfaces in steel,
the formation of austenite is initiated at
many spots. Under conditions fa vorable to
continued formation of austenite, each
small grain formed at an interface acts as
a nucleus around which growth continues in
all three dimensions until an expanding sur-
face of an austenite grain contacts the
surface of another e~pandingaustenite grain.
At a temperature fa vorable for the steel to
be completely austenitic, it consists of a
great number of grains of austenite. The
various regions in which ferrite, cementite,
and austenite exist in the plain carbon
steels is diagrammed in Fig. 2:4. This is
c~lled a constitution diagram and its values
are determined under near equilibrium conditions
of heating and cooling. This diagram is actually
the same one shown in Fig. 2:3 but relabeled to
reflect the new terminology. An additional line

1700.--------------------------~
16741.
Austenite <y)
Limit of Solid
lL.
o
in Gamma Iron~
~1500
L..

~
::3
~ Ferrite and ===:a..~ ~~ Austenite and
~ Austenite ~ ~v~ ~ Iron Carbide

~
'P s
~~~ 0.0218 % C 1340.6.r~
1300~~
~Ferrite (0)
Ferrite and Iron Carbide (Cementite)

1200
0.2 0.4 0.6 0.8 1.0 1.2 1.4
Per Cent Carbon
Fig. 2:4. Partial Constitution Diagram.
This shows regions of existence of ferrite, austenite, and cementite.

showing the limit of solid solubility of carbon in
gamma iron has also been added.
It is interesting to note that for a steel
ha ving 0.77% carbon content, the change
from ferrite to austenite occurs at a single
temperature, 1340.6 F, where all of the
carbon is taken into solid solution on heat-
ing. Such a steel is called eutectoid. For
steels containing more than 0.77% carbon,
the change from ferrite to austenite also
occurs at 1340.6 F, but not all the iron car-
bide can be dissolved. Higher tempera-
tures represented by the Acm line must be
reached before complete solution of iron
carbide can occur.
Iron-carbon alloys containing more
than 0.77% carbon are known as hyper-
eutectoid steels. If a steel contains less
than 0.77% carbon, the change from ferrite
to austenite takes place -over a temperature
range as indicated in Fig. 2:4. Such steels
are called hypoeutectoid steels. Again, it
should be emphasized that the diagram is
for plain carbon steels heated and cooled at
very slow rates. As the modifications which
occur in steel, such as change of lattice and
solution of carbon, are not instantaneous,
they require time in which to take place.
Very slow rates of heating and cooling allow
ample time for the reactions to occur and
proceed to the degree of completion indicated
by the lines in the diagram. Faster rates
of heating and cooling, a$ well as alloy addi-
tions, affect the construction of the diagram,
as we will see shortly.
When austenite is cooled, the reverse
transformations occur, and as the tempera-
ture drops, the ability of austenite to re-
tai,Q carbon in solution decreases from 2.11 %
at 2066 0 F tOO.77o/i,at 1340.6 F. Below 1340.6 F,
austenite transforms to ferrite and cemen-
tite precipitates, since ferrite can hold only
0.0218% C or less, as shown in
Fig. 2:3.
MICROSTR UCTURE
When the carbon separates from the solid
. solution, it forms cementite particles which
may be located either in the grain bound-
aries of the original parent austenite grain
or within the confines of the grain. Simi-
larly, ferrite may be found in the grain
9
boundaries as well as within the grain. In
either case, during the cooling of austenite,
the formation of ferrite and cementite may
assume any of several patterns and be
present in some definite proportions. This
internal arrangement of the microconstitu-
ents in metals constitutes its microstruc-
ture.
As previously stated, a careful exam.i-
nation of a piece of polished metal, either
with the eye or the microscope, will not
reveal a microstructure. This is because
the smooth surface of the metal reflects
the light and acts as a mirror so that we
cannot see behind the flat surface layer.
In order to see the internal microstruc-
ture of a nletal, it is first necessary to
prepare the metallic surface by grinding and
polishing it until smooth. Then, if we ap-
ply an acid to the surface of the polished
metal, certain phases or areas will be at'-
tacked more readily than others, resulting
in "hills and valleys" on a microscopic
sc~.le. Grain boundaries, for instance, are
attacked more severely than the grains; and
the grains themselves are attacked to vary-
ing degrees because of orientation differ-
ences.
When an acid-etched metal is placed
under the microscope, the shadows caused
by the .depressions in the metal make the
grain boundaries clearly visible. By a
similar procedure, other particles or phases
within the grains may also be made visible.
On account of the extremely fine size of
some of the constituents, the microstruc-
tures must necessarily be observed under
the microscope, using magnifications rang-
ing from lOOX to lOOOX or more. The
science of determining the internal struc-
ture of metals is called metallography and
is a fi eld of effort in itself.
Steel When steel has been cooled from
above A 3 to room temperature, any
of four possible microstructures can result,
depending on the nature of the carbide
distribution in ferrite. These are spher-
oidite, pearlite, bainite, and martensite.
(Martensite is usually reheated to impart
some ductility and is then referred to as
tempered martensite.)
10
~ach micrqstructural constituent has a
particular range of hardness, strength, etc.
The various microstructures and the cor-
responding hardness and strength values
for a o. 80% plain carbon steel are shown
in Fig. 2:5. Although the same microstruc-
tures can be developed in all steels, the
corresponding values of hardness and
strength will vary somewhat according to
the chemical composition of the steel. It
should also be noted that in a given steel
there are no sharp lines of demarcation
between the various microstructures, but
they blend into each other gradually. As a
result, the individual microstructures pos-
sess a range of hardness and strength, rather
than a single specific value, as presented
in the simplified data in Fig. 2:5.
It should be clear from the preceding
how the mechanical properties of the steel
are controlled by the processes of heat
treatment. Steel before heat treating has
its carbon distributed in some manner which
results in a microstructure with less than
optimum properties. In order to eliminate
the undesired microstructure, the steel is
heated to some elevated temperature where
austenite forms and the carbon is taken into
solid solution. The previous microstruc-
ture, along with its undesired properties,
is thereby n erased". In order to form a
new microstructure having desired proper-
ties, the austenite is cooled to room tem-
perature. During cooling, the austenite
transforms to different products of decom-
position. This causes the amount of car-
bon which can be held in solid solution to
be greatly reduced. By controlling the
conditions of cooling, the carbon which
comes out of the solid solution can be
made to assume a new manner of distribu-
tion, different from that which originally
existed. The new microstructure is, of
course, selected to afford the properties
desired in the steel.
TRA NSFORMA TION TEMPERA TURES
Obviously, if the heat treater is to con-
trol the manner in which the carbon is dis-
tributed in the steel and hence the final
mechanical properties, it is imperative that
he know the temperature at which carbon
enters solid solution in austenite, as well
as the temperature at which austenite trans-
forms and redistributes the carbon in the
steel. " In Fig. 2:4, the temperatures at
which the various constituents can exist
are clearly outlined. This diagram illus':
strates the changes which occur in plain
carbon steels under conditions of very
slow heating and cooling. The tempera-
tures at which the phase changes occur
are called the t~ansformation temperatures
or the critical temperatures, the former
being the preferred term.
Under conditions of slow heating, it will
be noted that the hypoeutectoid steels have
a lower transformation temperature (indi-
cated by line AI) where austenite begins to
form, and an upper transformation temper-
ature (indicated by line A 3 ) above which the
transformation from ferrite to austenite is
complete. Between these two temperatures
a mixture of ferrite and austenite exists.
This region is called the transformation
range or the critical range. Under similar
conditions of heating, the eutectoid steel
and the hypereutectoid steels act somewhat
differently in that the phase-change of the
former is completed at a single tempera-
ture; this is represented by the A 1 line.
In hypereutectoid steels, the line Acm
must be exceeded in temperature if all
of the carbon is to go into solid solu-
tion. At temperatures between the A em
and the A 3, 1 line, the steel will consist of
a mixture of austenite and cementite, as
indicated in Fig. 2:4. In the hypoeutectoid
steels., the carbon in the cementite goes
into solution in the gamma iron when the Al
line has been passed. At this low temper-
ature, only a small amount of austenite
has been formed, and since it "holds all of
the carbon it is relati vely rich in this ele-
ment. As the temperature is raised, more
austenite forms which is initially lacking in
carbon. However, by a process of diffu-
sion, some of the carbon in the high-carbon
content austenite which was first formed
passes over to the newly formed, carbon-
poor austenite, so that eventually this
element is evenly distributed throughout the
steel.
 11

Typical Hardness - 160 Bhn Typical Hardness - 235 Bhn

Tensile Strength - 82,000 psi Tensile Strength - 116,000psi

Typical Hardness - 495 BIm Typical Hardness - 653 Bhn

Tensile Strength - 247,000 psi Tensile Strength - 324,000 psi

Fig. 2:5. Microstructures Showing Effect of Varying Distributions of Carbon on the

Hardness and Strength of 0.80% Carbon Steel.
Spheroidite - an aggregate of spheroidal carbides in ferrite.
Pearlite - so called because of its lustrous appearance.
Bainite - named for E. C. Bain, its discoverer.
Martensite - named for A. Martens, pioneer metallographer.
12

Diffusion The diffusion mechanism has employed are much more rapid than those
been observed by -all of us at used ~o determine the constitution diagram
one time or another and can readily and shown in Fig. 2:4. A wide variety of steels
clearly be demonstrated. If a small puddle other than plain carbon steels are also fre-
of white paint is placed in a container and quently· processed in most heat treatment
several drops of red paint are added in the shops. These steels may, in addition to
middle of the puddle, it appears that we the full range of carbon contents shown in
ha ve two entirely different materials, each the constitution diagram, contain one or
with clearly defined area limits. several alloying agents in amounts ranging
As time passes, the sharp lines of de- from relatively small to relatively large.
marcation blur, and the white paint in the It is natural, therefore, for the student to
vicinity of the red drops begins to assume inquire just how various rates of heating
a reddish tinge. With the passage of more and cooling, as well as different amounts
time, we can no longer distinguish red and types of alloys, will affect the temper-
from white but have a homogeneous pink atures at which austenite forms and trans-
puddle. We can then say that diffusion has forms.
taken place in that the two dissimilar colors
have blended and assumed an average color Both the rate of heating and cooling and
value of pink. It will be noted that diffusion the type and amount of alloy addition have
is a relatively slow process requiring time an appreciable effect on the location of the
in which to arrive at the end result. Under transformation temperatures. When ad-
the conditions of slow heating described here, vanced rates of heating are employed, both
sufficient time for diffusion to occur is al- the lower and upper heating transformation
lowed. Homogeneity of carbon content in temperatures are elevated. In similar
the austenite developed in eutectoid and fashion, advanced rates of cooling lower
both the upper and lower cooling transforma-
hypereutectoid steels is also achieved by
the process of diffusion. tion temperatures. The effect of advanced
rates of heating and cooling is illustrated
Under normal heat treating conditions, in Fig. 2:6. New terminology is used here
the rates of heating and cooling usually to designate the lines representing the
1700 ~,,---.,.----~----r-----r------,r-----....------­
~

" "-
"
"" "
1600 ""
"
""
"
4 " ",,~
" '~
,,
,. 1500
""
,
"
",
1&1
a: " , ,.
:)
"-
ct- 1400 "
-~!. -- - ~- -' ~ - ~~...I- -
a:::
1.&.1
Q.
- - - ---- - -- - - - - - - - - - - - - __
~ AI A
...
1&.1 ___________A!J :: ~}s1 _

1200 - - -_ _..a-. --'- ~ ...... _'_ .....J. ...J

0.2 0.4 0.6 0.8 1.0 1.2 1.4

PERCENT CARBON
Fig. 2:6. Effect of Advanced Rates of Heating and Cooling on the
Transformation Temperatures of Steel.

limiting temperatures of the various phase-
changes. On heating, the lowel- trans-
formation temperature of the hypoeutectoid
steels is designated as the AC1 ' and the
upper transformation temperature as the
A c . In the eutectoid and hypereutectoid
ste~ls, the line A 3 ,1 becomes A C3 • 1 •
When the steel is cooled at advanced
rates, the upper cooling transformation
temperature of the hypoeutectoid steels is
designated as the A r ,and the lower cool-
ing transformation timperature as the Ar ·
The cooling transformation temperature hi
the eutectoid and hypereutectoid steels is
designated as A r3 1. The Acmline, rep-
resenting the maXImum solid solubility of
carbon in austenite, is relatively unaffected
by changes in the rate of heating and cool-
ing.
The addition of alloying elements to the
steel has a pronounced effect on the loca-
tion of the transformation temperatures.
Depending upon the type and amount of
these elements, the constitution diagram
as shown for the plain carbon steels may
be altered considerably in appearance. For
the time being, the specific effects of alloys
need not concern us.
The first step in any heat treating oper-
ation consists of heating the steel from
room temperature to some preselected ele-
vated temperature. For a successful heat-
ing procedure the operator should know
the following:
1. The lower heating transformation
temperature (A C1 )where the austenite
begins to form.
2. The upper heating transformation
,. temperature (A C3 ) where the steel is
entirely austenitic.
3. The heating transformation range and
the degree to which austenite trans'-
formation has proceeded.
4. The extent and uniformity of carbon
solution in the austenite.
Dilatometric This information is neces-
Method sary for the particular
steel being processed and
for the particular rate of heating which is
to be employed. Numerous methods exist
13
for the determination of the transformation
temperatures. Probably the most Widely
used method is the dilatometric methOd,
which consists of placing a piece of the
steel in a suitable furnace where the rate
of heating can be accurately controlled. A t-
tached to the steel sample isa sensitive de-
vice which measures and magnifies any
longitudinal or volumetric changes which
may occur. The equipment is usually de-
signed so that both the temperature and
c:
-eno
c:
&
x
lLJ
-g
Q;
E
::J
~
o
~
Q)
c
:.::;
Temperature' - - - -
Fig. 2:7. Normal Thermal Expansion.
Note proportional expansion of steel
when heated below the critical range.
volume changes are recorded, or the values
plotted on a graph.
As the steel is heated above room tem-
perature a normal thermal expansion occurs.
The thermal expansion is proportional tothe
increase in temperature and may be graphi-
cally represented as in Fig. 2: 7· This
behavior continues until the lower trans-
formation temperature is reached and the
alpha iron alters to the gamma form; the
rearrangement of the atoms from body-
centered cubic to face-centered cubic then
causes the lattice to be more closely
packed, and a contraction occurs in the
steel. The extent of the contraction is suf-
ficient to offset the normal thermal expan-
sion and causes the plotted line to assume
a downward slope" as in Fig. 2:8. The
overall contraction continues throughout the
transformation range in spite of the in-
crease in temperature. After the alpha
to gamma transformation is complete, the
steel resumes its normal thermal expan-
sion.
 14

Contraction Due To /
versus dimensional change assumes an up-
Phose Change ~_. ward slope. This continues throughout the
c cooling transformation range. When the
o
o
·en
c::
c..
x
LIJ
. .~. 1
Resumption Of Normal
cooling transformation is complete, the steel
again contracts and the plotted line assumes
Thermal Expansion After its original downward slope. The A r3 and
Completion Of Phase Ar are respectively represented by the
ChanQe
terhperature where the curve first deviates
from normal thermal contraction and where
it resumes normal thermal contraction.~

Temperature - - - -

Fig. 2:8. Contraction vs. Expansion.

Diagram shows how contraction caused
by transformation offsets normal thermal
expansion during heating. .

From the plotted line representing tem-

perature versus length or volume change,
it is a simple matter to determine the
lower heating transformation temperature
(A C1 ). This is done by noting the tem-
perature where the first deviation from
,.. .,rmal thermal expansion occurs. The
upper heating transformation temperature
(A c 3) is represented by the point where the
contraction due to phase- change ends, and
the plotted line representing temperature
versus dimensional change resumes its
normal upward slope.
The significance of the designations ACl
and A c 3 can be appreciated readily. The
letter A stands for the French word "arr@t",
which means stop or arrest. The letter c
represents the word "chauffage" or heating.
The numerals 1 and 3 indicate the sequence
of the arrests. The numeral 2 following Ac
designates the temperature at which a change
in magnetic properties of alpha iron occurs.
It is of little consequence to the heat treater.
When the steel is cooled, the reverse
changes occur. Upon cooling, a normal
thermal contraction takes place proportion-
ally to the drop in temperature. When the
temperature is reached at which the gamma
to alpha modification occurs, the rearrange-
ment of the atoms froI.Il face-centered cubic
to body-centered cubic causes a lattice ex-
pansion. The lattice expansion is sufficient
to offset the normal thermal cooling con-
traction, and the line plotting temperature
Temperature - - -
Fig. 2:9. Schematic Dilatometer Curve Show-
ing Relationship Between Heating
and Cooling Criticals.
The letter r in ·the designation represents
the French word "refroidissement" for cool-
ing. It will be noted in Fig. 2:9 that the
range of heating transformation is higher
than the cooling transformation range.
When the transformation temperatures
are determined at very slow rates of heat-
ing and cooling, the gap between the A c
and A r temperatures no longer exists, and,
as was demonstrated earlier, the lower
(and upper) heating and cooling criticals
coincide and are designated by the letter A
followed by the suitable numeral or numer-
als. Frequently, the transformation tem-
peratures are indicated by the designation
A e followed by one or more numerals. In
thIS case, the letter e designates the word
equilibrium and indicates that heating and
cooling have taken place at very slow rates.
Heat Meas- In addition to the method
urement described, several other
procedures are available
for determining the transformation tem-
peratures It is one of our natural laws

that in order to bring about any change in
an existing' system, a certain amount of
energy must be supplied to the system or
taken away from it. This is true whether
a page is turned, ice is made, a nail
driven, or a lattice structure altered from
body-centered cubic to face-centered cubic,
or vice versa. The energy involved in
creating the change may be in many forms,
one of which is heat. To trigger the alpha
to gamma modification, heat must be sup-
plied to the steel.
When the reverse reaction occurs, heat
is given off. By externally heating and
cooling the steel at a uniform rate and
measuring the heat' content of the steel, it
is possible, by suitable data plotting meth-
ods, to note accurately where the transfor-
mation changes occur.
Magnetic A third method of determin-
Properties ing transfo;rmation tempera -
tures is based on the fact
that steel has certain magnetic properties
which change with the phase changes.
(Alpha iron is magnetic while gamma iron
is nonmagnetic.) By heating the steel and
noting where the ferromagnetic character-
istics change, it is possible to locate the
transformation temperatures. This infor-
mation was put to direct practical use years
ago by the heat treaters, who lacked the
scientific temperature measuring devices
and information available to modern heat
treaters. The old-time operators would
suspend a magnet in the heat treating
furnace in close proximity to the work. A
steel object was placed against the magnet
and held there by normal magnetic attrac-
tion. When the furnace was heated and
reached a temperature where austenite
formed, the magnet would become austenitic,
lose -its magnetic force, and drop the steel
object. The falling of the steel object was
the signal to the heat treater that the steel
was at, or close to, a temperature from
which he must start his cooling procedure.
A fourth procedure for determining trans-
formation temperatures consists of studying
the microstructure of specimens which have
been heated to different temperatures and
then quenched. In this manner, it is possi-
ble to tell when the heating transformation
15
temperatures have been reached and to fol-
low the progress in the transformation
range.
Transformation temperatures are of
major importance to the heat treater in that
they help him determine the proper tempera-
ture to which to heat the steel. This heat-
ing temperature is usually referred to as
the austenitizing temperature. Correctly
speaking, the term austenitizing temperature
should be used only in those cases where
the steel is completely austenitic. However,
the term is commonly used even when the
steel is only partly austenitic. A s we shall
see later, the selection of the austenitizing
temperature has an important bearing upon
the final heat treating results.
Transformation temperatures are of little
use to the heat treater in determining proper
cooling practice for the transformation of
austenite. The next lesson describes more
pertinent information which allows the heat.
treater to select, with high precision, a
cooling procedure that will result in a de-
sired microstructure and hence the required
mechanical properties.
Effect of Microconstituents on Mechanical
Properties
From the preceding discussions, it is
apparent that metals, and particularly
steels, poss,ess an internal structure which
can be observed under proper conditions.
This microstructure changes with the amount
of carbon present in the steel and consists
of varying amounts of ferrite, cementite,
and lamellar pearlite. Each of these con-
stituents has a set of mechanical proper-
ties all its own, as in Table 2:6.
Hardness and tensile strength of several
microstructures of O. 40% C low alloy steel
are shown in Table 2:5. This represents
an easy-to-understand "framework" within
which the properties of most steels must
fall. Higher strengths, however, are being
obtained with quenched steels of higher car-
bon content. The effect of increasing carbon
content on the as quenched hardness of mar-
tensite is illustrated graphically in Fig. 2:10
16

Table 2:6. Mechanical Properties of Ferrite, Cementite and Lamellar Pearlite

Microstructural Tensile Strength Elongation Brinell Hardness

Constituent psi % Number
Ferrite 40,000- 50,000 40 90
Lamellar Pearlite . 125,000-150,000 15 275
Cementite 325,000 Negligible 650

and also the relative effects of varying
amounts of martensite.
The role of microstructure in deter-
mining the mechanical properties of steel
should now be obvious. In a hypoeutectoid
steel where much ferrite is present, we
might expect the steel to be relatively soft
and weak but quite ductile. In a hyper-
eutectoid steel where much cementite is
present, we might expect the steel to be
strong, hard, and brittle. By altering
the amount of the various constituents in
the microstructure, practically any degree
of strength, hardness and ductility, within
the limits shown, can be obtained.
In Table 2:6, the microstructure has
been varied by the amount of carbon in
the steel, with the cooling rate of the
steel through the critical ranges being very
slow, i. e. , a condition of near-equilibrium.
The complete iron-cementite phase diagram
is shown on the back cover as Fig. 2:12.
It would be well to become thoroughly
familiar with it so that it can be redrawn
from memory.
As demonstrated in Table 2:5, the me-
chanical properties of steel can also be
varied considerably by the process of heat
treatment. When steel is cooled at rates
faster than equilibrium, the ·structural
changes which take place are more drastic
and less well understood than those noted
here. These additional structural changes
have a considerable effect upon the me-
chanical properties of steel and will be
discussed in greater detail in succeeding
lessons.

60
."",
" "
.....
."..-- ------_-
.
•: 50 \,e
~fiJ#
~" •• ,
r •••
#fIII'"
~......
• .,,-
.'
.--.--.
••• ••• __ .--..... -._.~

c ~\e aoto ,,"... ~.

~~ -A? ,-: •••
~ ~ ... ~OOlo .".
".
a
::E: / ... ~~ .'."
/1IItr.f.t"I~.·
4 ~"J..
,-=-~. , pJo ."
(.) ~ ••• 0 010
•• ,.~ ~O
-a; /
-
~
g
a::
30

10....._ ..... ---1_ _......._ _..... .....

0.1 0.2 0.3 0.4 Q5 0.6 0.7 OS

Carbon, %

Fig. 2:10. Average Relationships Between Carbon Content, Hardness,

and Percentage of Martensite in Quenching.
(Transactions, AIM:E, 1946, V. 167, pp. 627-642.)
 17

1538 C

1394C

~
00
1674
oU- 00
-
CD Of:)
...
~
o
1418
.•.•..•..........•...•.......................••••..••.•••.•..
..:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:
••••••••.•••...•..•............................•..•.•...••••.
•.•.••.•••.•.•.•........................•.••..••.........•...•
'-
Q) •...•.•••.•.•.•.•.•..••.......•...•..........••...•...•••.....
.••.•.....•.•.......................•.•.••..•••..•.........•.••
.:::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::.
~
E ................................
:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.
••••.....•.•.....•....................•....•.•••.•.•.•••....••....
•.••.•.......•........•..•....•............••.•..•.......•.•..•...
~ .•........................•..••............•••.•.••..••.....•......
.•••.•.............•..•••.••............•••••••.•.•.•••....•...••.••
••••••.•••...•.........••.......••.............••••..••....•••••...••..
::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::
~~~~~~~~~: ~ ~:~.~:~~~.~:~~
•:.:.:.:.:.:.:.:.:••~agnetlc Iro'! :.:.:.:.:.:.:.:.:.:.:.:.
::::::::::::::::::••:: (Alpha Iron) ::::::::::::::::::::::::::::
··················~········b .cc···································
•••....•....•............... ..•.•••..•.•......•.•.•....•......••••
••••..•...•....•......................•...••••••.•..•.•..•...•.•.•...•••••
••.•.•.......•.......................••..••••....••....•.•.•••••••...•.•••.
••••••••........•.......•..•..•..•.........••....•.•........•••....•••••••••
••••••••••.....••...•...................•.•.••••.••••....•.•••••••••.•.••.•.
::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::.
•......................................
.••••••...•••.••.......•.•....••.•.....•..••.••••••.•....••..•.•.•.....•.••••..•
:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:
::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::
.•..••..•••.....•...•.••...•.••.••..•..•.•.••••.•.••.••••••.•.•...•...••..•.••..•.
:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:
.•.............................•.............•.......•..••.........................
.:.:.:.:.:.:.:::::::::::::::::::::::::::::::::.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:
:.:.:.:.:.:.........•..•...•................:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.
.
.••••••.•...............••.....••...........•.•.••....•.....••.......................
.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:.:..
Tem=ure ...::.::.:....::.:..::.:.:.:..::...::.:..::.::.:.::...:.:..:.=:.::~:.:..": ":.: .: ~:, ,: , : .:, : , : , :, ,: w:=": "·:": "=l.I : Lj: ..=..=:.=..:=....::1.1::..:..=:.:.:.:...:=..::.... ..-.
Time, Minutes

Fig. 2:11. Changes in Pure Iron as It Cools from the Molten State to
Room Temperature.
 3200

3000

:
J
I
L : 4- PERITECTIC
LIQUID 2723F

I
Y+L

.• ~,-----.,
em CEMENT.!IS..J
.~or
IRON CARBIDE
Fe 3C

EUTECTOID
1340.6 F

1000.:;
«+ em
8001::t:·--+---+-~-4--4---+--+--+-~--4--+-+--""--+----f--+--+---+-+---t"---t----t

600~:-~---+--+---I---+-~-~~~--4---I---4--+-~---+---+----+--t--+--t---t---t

300: o~ -+--4--...-4---4--~'-I---1--4-4-+-----4---4--4--I---I--~--+---1---+----i--+----t----t
o
oI
o 0.5 1.0 2.0' 3 4 5 6 6.67
·PER CENT CARBON

- - - - - - - - - - - - - - TYPICAL ANNEALED MICROSTRUCTURES - - - - - - - - - - -

. (--
Fig. 2: 12.. Iron-Cementite Phase Diagram.
The critical points for pure iron are shown on the left;
changes appear as lines progressing to the right as car-
bon is added.
 MATERIALS ENGINEERING INSTITUTE
C10L2
C11L11
C41L3
Test Series B Microstructure and Mechanical Properties Page 1

PLEASE PRINT YOUR NAME AND ADDRESS BELOW RETURN TO: ASM Intemationa/@
MATERIALS PARK, OH 44073-0002
NAME REG/STRATION NO. _
STREET _

CITY STATE ZlP DATE _

MULTIPLE CHOICE

Pli1ce the appropriate letter (A, B, C, D) in the box. (Only one answer is correct
and each question has a value of10 points.)

1. Figure 2:6 indicates that for a specific 0.40% carbon steel the Ac3 temperature is:

(A) below the Ars temperature.

D
(B) above the AC1 temperature.
(C) below the ~ temperature.
(D) equal to the Acm temperature.

2. Table 4 and Figure 10 indicate that the maximum attainable hardness for a steel
containing 0.30% carbon is approximately: D
(A) RC 60.
(B) RC 30.
(C) RC40.
(D) ·RC 50.

Figure 5 shows that the softest microstructure for steel ofeuteetoid composition is:
3.

(A) spheroidite.
D
(B) bainite.
(C) martensite.
(D) pearlite.

Iron-carbon alloys containing more than 0.80% carbon are known as:
4.

(A) hypereuteetoid steels.

D
(B) eutectoid steels.
(C) eutectic steels.
(D) hypoeuteetoid steels.
 MATERIALS ENGINEERING INSTITUTE
C10L2
C11L11
.C41L3
Test Series B Microstructure and Mechanical Properties Page 2

Definitions of ferrous microstructures are given on the inside back cover. Steel
5.
microstructures generally consist of two phase mixtures of ferrite and cementite.
The exception to this illustrated in Fig. 5 is:
D
(A) pearlite.
(B) bainite.
(C) martensite.
(D) spheroidite.

6. The A cm line on the Fe-Fe3C diagram (Figure 12) represents the limit of increasing
carbon solUbility in: D
(A) austenite with decreasing temperature.
(B) ferrite with increasing temperature.
(C) austenite with increasing temperature.
(0) cementite with increasing temperature.

The temperature to which steel is heated prior to quenching to martensite is called:

7.

(A) the fenitizing temperature.

D
(B) the austenitizing temperature.
(C) the Arl temperature.
(D) the annealing temperature.

A plain carbon steel containing 0.50% carbon could be referred to as:

8.

(A) a hypereuteetoid steel.

D
(B) a euteetoid steel.
(C) a hypoeuteetoid steel.
(0) a eutectic steel.

9. When euteetoid steel is transformed from pearlite to austenite on heating, the

eutectoid reaction can be observed by: D
(A) an increase in size.
(B) a decrease in size.
(C) an increase in ferromagnetism.
(D) the evolution of heat.

10. Definitions offerrous microstructures are given on the inside back cover. The steel
microstructure which contains cementite (Fe3C) is: D
(A) martensite. (C) austenite.
(B) bainite. (0) ferrite.
 IRON-CARBON-AlLOY SYSTEM
(Including Plain Carbon and Alloy Steels and Cast Iron)
PHASES

1. ferrite -«(alpha) - body centered cubic - a solid solution of carbon and/or alloying
elements in b. c. c. iron (carbon ranges from 0 to ,..,0.025%) -

2. austenite - Y(gamma) - face centered cubic - a solid solution of carbon and/or alloy-
ing elements in f. c. c. iron (carlx>n ranges from 0 to,..,2.0%) -

3. ferrite - 8 (delta) - body centered cubic - a solid solution of carlx>n and!or alloying
elements in b. c. c. iron (carbon ranges from 0 to~O.10%)

4. cementite - em - orthorhombic - iron carbide - Fe 3C - a chemical compound of iron

and carbon -

5. graphite - carbon in the free -state occurring in several geometric foms.

6. alloy carbide - a chemical compound containing alloying elements which form both
simple and complex chemical compounds with carbon - usually Mn, Cr, V, W, Mo,
Cb, and Ti -

MICROSTRUCTURES

1. ferrite - see phases 1 and 3

2. austenite - see phase 2

3. cementite - see phase 4

4. pearlite (coarse and fine) - a mixture of ferrite and cementite consisting of alternate
platelets with the thickness of the ferrite being about seven times the thiclmess of
the cementite.

5. spheroidite (coarse and fme) - a mixture of ferrite and cementite and!or alloy carbides
in which the carbides are in a spheroidal form.

6. bainite (upper and lower) - a mixture of fenite and cementite in which the carbides are
in a very fme rodlike or needlelike form resulting in an acicular appearance. - -

7. tempered martensite - a mixture of fenite and cementite in which the carbides are very
finely dispersed in a submicroscopic spheroidal form. --

8. martensite - a supersaturated solid solution of carbon in distorted body centered

cubic iron - the iron lattice is commonly referred to as body centered tetragonal

9. graphite - see phase 5.- free carbon whose shape either as a flake, nodule, or spheroid
describes the type of cast iron as either gray, malleable, or ductile.

10. alloy carbide - see phase 6.

 ASM Internationall!l is a Society whose
mission is to gather, process, and
disseminate technical information.
ASM fosters the understanding and
application of engineered materials and
The Materials their research, design, reliable manu-
Information Society
facture, use, and economic and social
benefits. This is accomplished via a
unique global infonnation-sharing
network of interaction among members
in forums and meetings, education
programs, and through publications
and electronic media.

Printed on DocuTech February 1994 I~

 ASH\.
\:.2
The Materials
Information Society

" Austenite and Its Transformation

and Classification of Steels
by
Howard E. Boyer

Course 10
Lesson, Test 3
Course 11
Lesson, Test 12
Course 41
Lesson, Test 4
 Metric Conversion Factors
To convert from To Multiply by

in. rom 25.4

in. m 25.4 x 10-3
mil 25.4
,.un. 25.4 x 10-3
in. 2 6.45 X 10-4
in. 3 1.64 X 10-5
ft. 3.048 X 10-1
ft. 2 9.29 X 10-2
ft. 3 2.831 X 10-2
oz. g 2.834 X 101
lb. kg 4.536 x 10-1
Btu J 1.054 x 1()3
Btullb. e of J/kg - K 4.18 x 103
Btulft. e hr e of W/m-K 1.730
10 in.lin.l°P 1~ mm/mrnl°e 1.8
psi Pa 6.895 x 1()3
psi kPa 6.895
ksi kPa 6.895 x 1()3
ksi MPa 6.895
ksi - in. l12 MPa e m l12 1.099
ksi"in. MPae"m 1.099
oz.f gf 28.4
lbf kgf 4.536 x 10-1
lbf N 4.448
lbf eft. Nem(orJ) 1.356
Ibf/in. 2 kgf/cm2 14.223
Ibf/in. 3 kgf/m3 2.768 x 1()4
lb.lft. 3 kglm3 16.019
Ib.lin. 3 g1cm3 2.768 x 101
Ib.lin. 3 kglm3 2.768 x 1()4
gal (U.S. liquid) L 3.785
gal (U.S. liquid) m3 3.785 x 10-3
lb.lgal gIL 119.826
ft.lgal mIlL 748
of °e (OP _ 32)/1.8
of K (OP + 459.67)/1.8
°e of (Oe e 1.8) + 32
°e K °C + 273.15
K °e °e-273.15

Multiple and Submultiple Units

1018 exa (E) 10-1 ••••••••••••••••••••• deci (d)
1015 ••••••••••••••••••••• peta (P) 10-2 centi (C)
1012 ••••••••••••••••••••• tera (T) 10-3 ••••••••••••••••••• milli (m)
109 giga (G) 1()-6 micro (Jl)
106 mega (M) 10-9 nano (n)
103 ••••••••••••••••••••••• kilo (k) 10-12 •••••••••••••••••••• pico (P)
102 hecto (h) 10-15 femto (f)
101 deka (da) 10-18 ••••••••••••••••••••• atto (a)

Abbreviations
J joule m meter Pa pascal
kgf kilogram force rom millimeter K kelvin
L liter N newton W watt

by
Howard E. Boyer
Consultant

Editor
Philip D. Harvey
Metals Engineering Institute, American Society for Metals

e 1981 American Society for Metals

This material may not be reproduced in whole or in part in any form whatsoever. FUlly protected by copyright.
 Austenite and Its Transformation
and
Classification of Steels
A definite relationship exists between the struc-
ture of steel and its mechanical properties. This
being the case, the ability to develop the mechani-
cal properties of steel is only as good as our ability
to control the microstructure.
When austenite transforms, two factors deter-
mine the structural and mechanical properties of
steel. These are the amount of carbon present in
the ste~l, or its composition, and the conditions of
cooling during the period of transformation, or
the cooling rate. By using very slow rates of cool-
ing, a ductile and soft structure results when the
austenite transforms If a rapid rate of cooling is
employed, a hard and strong structure can be
secured.
Whether the structure is controlled by the com-
position of the steel or the cooling rate, the changes
which take place when the austenite transforms
and the resultant structures are of concern. The
primary key to understanding all heat treating
procedures is the control of austenite and its trans-
formation products.
This lesson describes how austenite is formed
and the effect of carbon and alloying elements on
its transformation; how austenite transforms to
pearlite, spheroidite, bainite and/or martensite;
transformation of austenite under isothermal and
continuous cooling conditions, and the reasons
why all of the austenite may not fully transform to
other products and the effect on the structure.
Also discussed in the lesson is the classification of
steels, the means of identifying each classification
and the steels within a classification.
After studying· this lesson, the student should
have an understanding of the following:
• Austenite is a solid solution of one or more
elements in face-centered cubic iron that
exists between 2541 F (1394 C) and 1674 F
(912 C).
• How austenite, under varying conditions of
time, temperature and alloy content, trans-
forms to other products, and the relation-
ships of these products to the properties of
the steel.
• The effects of alloying elements other than
carbon on the transformation of austenite
and, in turn, on the properties of steel.
• Classification by composition of the more
common grades of carbon, alloy, stainless,
tool and maraging steels, and the method of
identifying them.
2
, I
0c
OF
2800 1538
FREEZING
POINT
LIQUID
, 1
2541 1394
FERRITE
( 6-Fe) Ci~
AUSTENITE
(,- Fe)
1674· 912
FERRITE
(a-Fe)
INC~EASING
32 0··..
, //////
IRON '0
(a) (IOO% Fe) (b) (PURE Fe)
Fig. 3-1. (a) Constitutional diagram for pure iron; (b)
effect of carbon on the iron diagram.
Formation of Austenite
In metallurgy, a constitutional diagram of an
alloy system containing two metals is a ruled sheet
whose ordinates are temperatures (shown verti-
cally) and compositions (shown horizontally, the
units being percentage of one of the metals). In
further simplification, the constitutional diagram
for a single metal or element is merely a straight,
vertical line. The constituti~nal diagram for pure
iron, and consequently its thermal history, is con-
tained in a single vertical line (Fig. 3-1a).
As iron cools from the liquid state, solidifica-
tion suddenly occurs at t~e single temperature of
2800 F or 1538 C. (Only in alloy systems does metal
freeze over a range of temperature.) The iron atoms
attach to one another in a specific crystal pattern,
such that a cube is formed having one atom at each
of its eight corners and a ninth atom at its center.
This crystalline pattern is called body-centered
cubic (bee), and is illustrated in Fig. 3-2a. This
crystalline form of iron is known as delta ferrite
(d-Fe).
On further cooling, these atoms rearrange
themselves by diffusing into a somewhat altered
arrangement, such that the atoms are spaced
around an atom at the geometric center of each of
the six cube faces. This arrangement is called face-
centered cubic (fcc), and is illustrated in Fig. 3-2c.
This crystalline form of iron is known as austenite
or gamma iron (y-Fe). T.he temperature at which
this change occurs is 2541 F or 1394 C (denoted as
FREEZING
POINT A 4 in Fig. 3-1a).
At 1674 F or 912 C, the austenite structure (fcc)
reverts to ferrite (bee), the stable form down to'
room temperature and below. The temperature at
which this crystalline change occurs is denoted as
A 3 in Fig. 3-1a. This body-centered cubic form is
known as alpha ferrite (a-Fe).
The technical name for the power to adopt more
CARBON
than one crystal form by a solid is allotropy. The
role of allotropy in elevating iron-base alloys to a
0.10 0.20
C, 0/ _
0
position of importance in engineering follows
from the fact that the face-centered cubic allotrope
known as austenite exists in a range of red heat
where the metal is workable. The metal then re-
verts again on cooling to body-centered cubic
ferrite. The atoms are less densely packed in fer-
rite, so this transformation causes an expansion in
all dimensions, and the structure may at times
have difficulty in accommodating this change in
size. More important, in this atomic shifting from
the face-centered cubic structure to the body-cen-
tered cubic structure, foreign atoms in the iron
matrix, particularly those in the interstices, might
become trapped and jammed, causing great distor-
tion and irregularities in the atomic architecture.
Such irregular atomic arrangements accompany
high strength and hardness.
Influence of Carbon
Returning now to the constitutional diagram for
pure iron shown in Fig. 3-1a, carbon is added as a
second variable, horizontally and to the right (Fig.
3-1b). Immediately, the transformation tempera-
tures which characterize pure iron are altered pro-
gressively as the carbon content increases. This is
indicated in Fig. 3-1b. For any fixed rate of cooling
the metal (with a carbon addition it is now steel)
remains liquid longer, transforms to delta ferrite
earlier, and hesitates in reverting to alpha ferrite
from austenite.
Carbon actually dissolves; that is, the individual
atoms of carbon lose themselves in the interstices
of the iron atoms - unlike metallic alloying ele-
ments which replace iron atoms in their positions
on the cube. It becomes evident that the interstices
within the face-centered cubic structure (austenite)

Fig. 3-2. Three possible crystal structures of steel. (a)
Bcc, body-centered cubic, characteristic of ferrite; (b)
bet, body-centered tetragonal structure:1 characteristic
oj martensite:1· (c) fcc, face-centered structure, charac-
teristic of austenite.
3
are larger or more accommodating to carbon than
are those of ferrite, the other allotrope. However,
this performance should not be interpreted only on
the mechanical basis of size of opening, for it is
also a fundamental matter involving electron bond-
ing and the balance of these attractive and repul-
sive forces which underlie the allotropic phenomenon.
In Fig. 3-3 the iron-iron carbide constitutional
diagram initiated in Fig. 3-lb is completed up to
70/0 carbon. In this illustration note:
1. The size and relative position of the austenite
field,
2. The changes imposed by carbon on the A3 line,
3. The carbon content at point E, which is the
eutectoid (0.770/0 C), and
4. Point B in the austenite field, which is 2.11 070
C of the maximum that dissolves in austenite
and at a temperature of 2098 F or 1148 C.
It should be noted that carbon is quite soluble in
austenite, while it is of very limited solubility in
ferrite; also, the very narrow terminal solid solu-
tion ferrite field at the left of Fig. 3-3. Here it is
seen that there exists very little solubility in ferrite
(approximately 0.022070 at 1341 F or 727 C), and
then further decreases until the solubility of car-
bon in ferrite is approximately 0.008070 at room
temperature.
The presence of (Jther elements in the steel also
affects the A 3 line as well as other areas in the iron-
iron carbide diagram, but to a much lesser amount
compared with the effects of carbon.
Hardening, full annealing and normalizing of
steels are achieved only by first heating the steel so
that it is austenitic. The control of mechanical
properties is achieved by controlling the transfor-
mation of austenite in terms of time and tempera-
ture into various transformation products.
Transformation
and Products Formed
Carbon will dissolve in austenite up to about
0.77070 at 1341 F (727 C), and in greater amounts
up to 2098 F (1148 C), as indicated in Fig. 3-3.
Since carbon is of very limited solubility in ferrite,
 4

°C
1600 r---..,...-----.....,....---r---,------,~-.,....._-_T_-_,._-~I
-1"""""---'---""-- OF

1500 ~153~.
;- :~4~=J ...,...---+--+--+---...,-----i:.---~-r_+_I-/__I_-__+_-_4_-__I 280 0
(S·Fe)" :A' L ,I

-r -+
I " ,

4 : '-- I 2600
1400 139~O ~~ I

1300 -.L---~~ :~ / 2400

i " ~ i ~ < "
1227
0

'.26~. '---------~-
I - ,

1200 -1--.... J. ! / 2200

: (Y-Fe)
8,",~2.08
0
1154 L+Fe,C
; I.,.. .....- '"'/
-~.:=;--
..------
.- -,.- - ... _~
i Austen~te ~ V 2.11 1148 4.30 6.69 0

II 00 t---+-,----+----t---g-#+---f---+---+--+--+----4---4----+-----l--+-.--d 2000
1 I ~/ I

1000
i I /
-.--l.----+-----;,.-.--+---+--+---+----+----4-----4--~--.l-----~~__1
II

! i I Austenite + Fe3C Cementite 1800

900 912~ I i l (Fe 3 C)--:

~31~ACM 1600
800 I\ 0.68' !)
??~.~. 1'1 E
._ 1400
1-, ~r/~", ~
.- -- ~, '-1-----
120b
600 t----t----;----t---t---+---+---+---f--I---+..---4---4---4-oI---d

1000
500 r---t----r----r----t---+--+--+--+---+---+--4---4---4--JJ---I
~(a-Fe)
Ferrite Ferrite + Fe! C 800
400 t---t---t---+---+---+---I---I---~--1----4----+---I-----+--I-~

300 r---t---t---t---+---+----+---+---+--~f------1---4-~----+--+-~

200
----,-_.- _.- _._-,_._-,_._-_._--_._-,-
600

t---;----t----+---+----+---+---+----+---+-----4-----+-----+----+..--I----=I
230 0
_._~. __._-- 40 a
I
Ferrite + Fe2.2C
100 t----t---r---+----+---+---I----+----+------:...-.+---.-.+----I----I----I----J----:t 2 00
;
I
I

o '----L.~'.w.l'1.Io..-----''----..-I-_-...._...L.-_.L-----l_---'-_-J.-_-L..-----J~_..l.......II----1
Fe 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0' 6.5 7.0
'Weight Pe'rcentage Carbon

Fig. 3-3. Iron-iron carbide constitutional diagram.

a major transformation must take place when area to the left of the Ps-Bs curve and above M s is
austenite changes to ferrite (y-Fe to a-Fe) and ce- shown as austenite. When austenite is cooled below
mentite (Fe3C). Art (Ps-B s) transformation begins, but very slowly
above about 1200 F (650 C). (Line A et in Fig. 3-4
The products that are formed from the...transfor- exists only under equilibrium conditions which are
mation of austenite depend mainly on composition not achieved on cooling.) At about 1000 F (540 C),
and time (cooling rate). The gamma-to-alpha plus or the nose of the TTT curve, transformation be-
cementite transformation takes place in a manner gins almost instantaneously (less than one second);
demonstrated by the time-temperature-transfor- then, at lower temperatures, the beginning of trans-
mation (TIT) curve given in Fig. 3-4. This curve is formation line is shifted far to the right, and not
for a eutectoid (0.770/0 C) steel. Temperature is until M s is reached does the beginning of transfor-
shown on the vertical scale and time is plotted mation become a horizontal line. The narrow gap
logarithmically on the horizontal scale. All of the between the vertical line at the left of the chart and
 MICROSTRUCTURES
927 (1700) •

871 (1600)

AUSTENITE

760 (1400)' i
4.

728 (1341) 4. Ae,------------~.

:+- 11
649 (1200)
32

38

538 (1000)
P 40
w
a:
:J
42
iw
Q. 427 (800) 0
:E 44 a:
w :r:
... I
47 en
en
w
316 (600)
t z
50 a
a:
c(
:I:
55

277 (530) Ms I ; aT
.. " .. , t • "'\ ~ ,,\ . • '" •• " , 58
" ... ,. - "' ... If ~ ..... "" ,- • ~ , • ~
.... " or - "' " v ,. -.., ,. ........ f1 I I 'J:" '(III ~II Lltll =
~~~~~~~~~;.~~'"'_vV"'_..·""vv 'c(+
~~~~~~~~~~~~~~~~ 62
.,-.,,.....,, V'" \""" f""'V ' V ~ ~ rv ...,...... rv V' .,......" 14.·~.I= ~ :;Q ..
v-v rv- V"' ~ rv rv- - r .......... v-v- -v-... ~ Y"" 'I· . t

v-vrv--v""~.rvrv__rv__."_..,y_.,r~orv_"""""..,....,,.,_...,r"""'""'
, v-v- ~ .....-. orv -rv- rv- ~ V"V -.r"'t ..".....,. -r"'t vv v-v -v-... vv.,-y rv- rv-o ""'V"\ r

-46 (-50) M, FeArn nrnrtllXlcn::n VFFFl;n=n;:tCnf)rrncrnrn"'nt I~_ . --' I 64

.-v~f""'\r"V-V~""",,",,,,""""~"""""'''''''''''''''''-v"'''\'''''' MARTENSITE • 66

1 SEC. 1 MIN. 1 HOUR 1 DAY

TIME

Fig. 3-4. Time-temperature-transformation diagram for a eutectoid (0. 77%) carbon steel.
UI
6
the beginning of transformation line (Ps-B s ) indi-
cates the critical cooling rate for the specific steel;
the critical cooling rate is that speed of cooling
which must be equaled or exceeded in order to
fully harden the steel. T"he \vidth of this gap also
indicates the hardening capacity or hardenability
of a given steel.
It is established that there are two basic methods
of controlling the structure and hence the mechani-
cal properties of a given steel; they are: steel com-
position and ~ooling rate. Regardless of whether
the structure is controlled largely by composition
or cooling rate, the resulting structures are of
n1ajor concern.
Products of Austenite l"ransformation
As stated previously, in simple carbon steels
austenite exists only above the A 3 line (Fig. 3-3)
under which conditions the tiny carbon atoms are
dissolved in gamma iron. When the temperature
drops below 1341 F (727 C), the carbon atoms will
escape or attempt to escape from the lattice as the
crystal changes from gamma to alpha. When
. cooling rates are such that an orderly escape is
permitted, the steel becomes annealed and the
microstructure is pearlite and/or spheroidite.
Lamellar pearlite is formed directly from austen-
ite. Figure 3-5 is a photomicrograph (a picture
taken with a metallurgical microscope) of this
constituent in a hypereutectoid steel (a steel con-
Fi~. 3-5. La/nellar pearlite in a hypereulectoid sleel;
40/0 picral, 1000 x.
taining more than 0.770/0 C). In a steel of this high
carbon content, the original austenite grains are
frequently outlined by cementite (iron carbide),
and the interior is' composed of the pearlite (alter-
nate plates of ferrite and cementite). In hypoeut~c.-
toid steels (less than 0.77010 C) the original austen-
ite grains are usually outlined with ferrite and the
grains are composed of pearlite. The eutectoid
steel (0.77070 C) will consist entirely of pearlite
regions. No grain boundary constituent will be
present, and the boundaries of the original auste-
nite grains are ordinarily difficult to determine
clearly unless special techniques are used.
In lower carbon steels the lamellar pearlite struc-
ture often may be used because it is considered as
one of the annealed structures and is generally
satisfactory as the prior structure for heat treating
and fabrication. However, for steels with h~gher
carbon content (about 0.450/0 and upwards, de-
pending upon the fabrication process), the type of
structure shown in Fig. 3-5 is regarded as poor for
forming, machining and heat treating operations,
and \vhat is known as a spheroidized structure is
far more desirable .
Spheroidite. The structure spheroidite (sphe~ical
cementite) may be formed directly by cooling aus-
tenite below AI. In a vast majority of cases, how-
ever, spheroidite is not formed in this manner but
is developed by modifying an existing structure,
such as martensite, bainite or pearlite. The meth-
ods of forming spheroidite from an existing struc-
ture may vary somewhat, depending to some extent
upon available equipment. One method consists of
heating the steel to a temperature just below where
austenite forms (note Fig. 3-3) for long periods of
time. Another method is to heat the steel just above
Al for a pre-established period of time, then just
below Al and continue for two or three of these
cycles. Spheroidizing techniques are described in
Lesson 4 of this course (Annealing and Normaliz-
ing of Steel). In any instance, spheroidizing is a
time-dependent reaction as well as being tempera-
ture dependent, and considerable time is required
to attain complete spheroidization. For example,
in spheroidizing mill heats of AISI 52100 bars,
cycles as long as 90 hours may be required.
Regardless of the (echnique used, the objective
is to attain a microstructure such as shown in Fig.
3-6. At the spheroidizing temperature the cemen-

Fig. 3-6. Spheroidite in a hypereutectoid steel; 5%
nital, 700 x.
ti te particles in the original structure change shape.
Instead of remaining as long, thin plates, as in
pearlite, by a diffusion process the cementite parti-
cles tend to rearrange into shorter pieces which
coalesce and form balls, or spheroids, of cemen-
tite. The resultant structure then consists of many
small rounded particles of cementite in a ferrite
matrix and is called spheroidite. A typical appear-
ance of this constituent is shown in Fig. 3-6.
The annealed structures, pearlite and spheroid-
ite, as shown in Figs. 3-5 and 3-6, are only typical
for fully annealed hypereutectoid steels. Depend-
ing on the steel and the condition of austenite for-
mation and transformation, the appearance of
these structures may vary. For instance, the length
and spacing (or thickness) of the cementite and
ferrite plates in lamellar pearlite may vary consid-
erably. The pearlite illustrated in Fig. 3-5 consists
of long plates, relatively thick and easily identi-
fied. Such pearlite is called coarse. On the other
hand, the plates may be very short and thin, so
that at the same magnification the lamellar struc-
ture would be hard to distinguish. Such pearlite is
called fine. In similar fashion, the size and distri-
bution of the spheroids can be altered so that they
may be small and close together or large and far-
ther apart. The degree of fineness or coarseness of
these structures has a definite bearing on mechani-
cal properties. A fine pearlite is harder and stronger
than a coarse pearlite.
7
High-carbon steels are usually furnished by the
mill in the spheroidized condition, ready for fur-
ther heat treatment or for various fabricating
operations.
Martensite. The two structures discussed in the
preceding paragraphs are forms of annealed struc-
tures and are developed by relatively slow trans-
formation of the austenite by diffusion, thus
permitting an orderly exit of the carbon atoms
from the gamma crystals and an orderly transfor-
mation to ferrite, or the alpha phase.
When the steel is cooleJ rapidly, entirely differ-
ent conditions prevail. l~he resulting microstruc-
ture depends upon steel composition and cooling
rate. If the cooling rate is sufficiently rapid so that
no transfo"rmation is permitted at the "nose" of
the TTT curve (Fig. 3-4), the carbon atoms are
entrapped in the austenite lattice. As shown in Fig.
3-4, if the tra~sformation can be avoided by rap-
idly cooling to below the nose of the TTT curve,
then there is appreciable time before transforma-
tion does begin, and as cooling proceeds at a
reasonable rate, austenite transformation for a
eutectoid steel begins at approximately 530 F (275
C), which is denoted as M s (martensite start) in
Fig. 3-4. When transformation takes place under
these conditions martensite begins to form at ·Ms '
and then transformation continues to M f (marten-
site finish).
Martensite is the quenched structure forined by
rapid cooling. Because of the carbon atoms being
entrapped the resulting. structure (shown in Fig.
3-7) is actually a supersaturated solution of carbon
·8

in alpha iron and is characterized by its sharp, It is again emphasized that a structure comprised
needle-like or acicular appearance. Because trans- of 100070 martensite can be formed only by cooling
formation from austenite to martensite is only so rapidly that no transformation from the auste-
temperature-dependent and not time-dependent, nite is allowed to take place until the M s tempera-
the transformation to supersaturated ferrite is by a ture is reached (Fig. 3-4). The width of the gap at
shear mechanism along crystallographic planes. the nose of the TTT curve varies over a broad time
Therefore, transformation to martensite is consid- range depending basically upon the composition of
ered as a diffusionless transfornlation. I'he mar- the steel. However, as a very practical considera-
tensite shown in f"ig. 3-7 is relatively coarse and is tion there now enters another variable, the mass of
used to illustrate the typical acicular structure. the steel being transformed. Obviously a large
Under normal conditions, nlartensite would be so mass of steel cannot be cooled as quickly as a
fine as to make the needles relatively indistinguish- small mass. Therefre, composition and mass must
able at magnifications as high as IOOOx. f"reshly often be considered together in establishing cool-
formed martensite is called 'Alhite martensite, or ing procedures.
quenched martensite, on account of its very light
etching characteristics. Since white martensite is a Bainite. Another quenched structure formed by
structure resulting from supersaturation, it is rela- rapid cooling past the nose of the l"'Tl" curve and
tively thermally unstable, and when subjected to held at some temperature above M s is shown in
warming or tempering, it tends to assume a more
stable form. "'his results in some of the carbon
being ejected from the supersaturated alpha iron
forming a carbide and causes the resultant struc-
ture, tempered martensite, to become darker etch-
ing. Figures 3-8a and 3-8b show the relative etching
characteristics of white martensite and tempered
martensite in a highly alloyed steel, where the mar-
tensite needles are so fine that they are indistin-
guishable at normal magnifications .
. ~ "~).'~"~"
y'. ~)~." ".....". .'J .;;... .

.f-.~;:..iP"/ :,;-,,;:~~,(f-.'': :,.:\..

!~~;:. {'..4-',_ (~-,-, )

~l, '~~,:.'. '",:_ . ')~~~~::.r~· ~~ M:

E';·';·-:;.r~·,~
·1' ~",. .'~ ,. ." ,:': .' /:~~:"'v--,"'r
~';$. ","'t .... J,....~
.... ~ '
~ . ~- .•\. '.." :'. n.~~",'~ "\
't'" •• z "',) .. ,
j .~~ ' . . ,
.,.1,
,
;- . . ,:" -_,..,,', .'I ' '."jJ.J -"',-...~' •
• 10(' .• (
.~. r
'\
...,
\.'1'\
;
.
~, ,,, - . f'"\ ~'.:-.•' :)f': ~ . ~ \:~~:'j' . _, .... .,'
• . ( ,., ~ . ./ ) -~: r .. . T • ~" ~
(~.l~~ . . ~ ~ . ~>: .".' ,J ,;~~. '. :",;,._,;)-.~. ~',
I" of'..,s.."" -.:,.,\- ":.! .....\ .' >~ -r·
I
~·'-·'I.'-~'··:·.·~
. -::'~r
.' ..' '\ . ~ \,... -,""""~
.J '''"'.'".
.. . ~ )l
!
4.

i)-·~ ..\.:
'. .v. ··,.J.f."
'r'" ; ...
/.
. ," - 'f::- ~-: . ~c: }.)~
~ 'J
I . !
.(/. -,.":./,,~
a
~
\"
:y -. , ",'
....
~ F",
'<
:' J 1
- ~~ ',! ' •
..... '"
"{~"'\.~
-- ~
."
' ~
,',
..
••
,)
'r'

Fig. 3-8. Martensite and bainite. (a) While as-quenched

martensite; (b) tempered martensite, in a hiRh-alloy
steel. The small white spots are undissolved carbides
(5% nital, 700 x); (c) bainite in a low-carbon Ni-Mo- V
steel which has been quenched and slightly tempered.
The bainite is thefeathery constituent and is surrounded
by tempered martensite (1000 x).

Fig. 3-8c. In contrast to martensite, bainite forms
by holding at some temperature .above M s and
indicated in Fig. 3-4. Bainite is another constituent
which has a range of appearances and hardnesses
and is generally classified as upper or lower bain-
ite. Lower bainite frequently resembles martensite,
and the two are difficult to distinguish. Lower
bainite, however, has a darker appearance when
etched and viewed with the microscope, while
martensite has a brownish tinge. Typically the
microstructure of bainite is characterized by its
"feathery" appearance, shown in Fig. 3-8c. Com-
pared with martensite, which is virtually always
tempered for use, bainite is a final structure, and
further heating would be objectionable.
Mixed Structures. The structures discussed above
are those which will be encountered in the study of
the transformation of austenite during heat treat-
ing operations. l"'heir presence will depend on the
carbon and alloy content of the steel and, more
important, on the time-temperature relationship
existing when the austenite is cooled. In all cases,
the heat treatment serves only as a means of alter-
ing the form in which the carbide is distributed in
the ferrite phase. Remember that carbon first com-
bines with iron and/or with other elements and is
dissolved and precipitated in the form of carbide'
(Fe3C) rather than in the form of elemental car-
bon. The effect of microstructure on hardness
(and hence on related mechanical properties) is
shown in Table 3-1.
On a practical basis, because of the variables
involved, a quenched workpiece may contain a
mixture of structures and is considered slack
quenched. 1"'his possibility is indicated in Figs. 3-19
and 3-20 and will be emphasized later in this lesson.
Table 3-1. Effect of Microstructure
on Hardness of 0.77% Carbon Steel
Microstructure
Spheroidite
Coarse. .... .. . . .. . .. . .. .
Fine. .. .. .. . . .... . . .. ..
Lamellar pearlite
Coarse. ..... . .... . .. . .. . ... .. . . .. . .
Fine.. . . . . .. . .. .. . ... .. . .. . .. . ... . .
Bainite
Upper.. . . . . .. . . ..
Lower.. . . .. . .. ..
Martensite, tempered
9
Transformation of Austenite
Under Isothermal Conditions
As discussed previously, the transformation of
austenite to pearlite, spheroidite or bainite is a
time-temperature dependent function, and the
resultant hardness depends not only upon the
structure formed, but also upon degrees of fine-
ness or coarseness of a particular structure. "·0
develop an understanding of what occurs in heat
treatment, the manner in which austenite trans-
forms under isothermal (canst an t temperature)
conditions must be considered.
A steel specimen about 1/4 in. in diameter and
1/16 in. thick (6.4 by 1.6 mm) is ideal for such a
study because such a small sample will respond
almost instantaneously to any temperature change.
Assume, then, that such a specimen of a eutectoid
carbon steel has been thoroughly and completely
austenitized at some temperature, 1" (see Fig. 3-9),
above the critical temperature AI. 3. When this
specimen is quenched in water to room tempera-
ture, transformation will occur, and as-quenched
(or white) martensite will result.
Now assume that several similar specimens are
austenitized in the same fashion as the first. If,
after being austenitized, the specimens are plunged
into a bath of molten salt held at temperature T},
the specimens will, because of their small size,
immediately assume the temperature of the bath,
T I. I f one of these specimens is then removed from
the salt bath after a lapse of time of a seconds and
then quenched in water at room temperature, it is
seen that the structure (as-quenched) consists en-
tirely of white martensite. This indicates that hold-
ing the specimen at temperature TO} for a seconds
has done nothing to the austenite, and it transforms
as normally would be expected if the steel were
Typical quenched directly from the austenitizing temperature.
hardness,
HBN If, after being in the salt pot for a period of b
seconds, a second specimen is removed and quenched
.. 156 in water, a somewhat different microstructure
.. 174 might be observed. Now, a small amount of lamel-
lar pearlite might be seen in the white martensite
170
375 field. This indicates that holding for b seconds at
temperature T 1 results in a small amount of auste-
. 400 nite transforming to pearlite. The balance of the
. 575 austenite then transforms to white martensite
650 when the specimen is quenched.
10

a
Il~11
5 Specimens
Austenitizing Temperature...,
T

4 Specimens

Transformation Temperature T1
in Molten Salt Bath
% Transformed
I 25 50 75 100 100

o o a o o
80% Pearlite 100% Pearlite
20OJoMartensite

Instantaneous
S
Log Time - - - - - -...~
Fig. 3-9. Typical transformation of austenite at a constant temperature. The above illustration is jor a 1080 steel
where T j is about 1150 F (620 C).

The microstructure of a third specimen, removed
from the salt bath after c seconds and quenched in
water, shows more lamellar pearlite in the white
martensite background. This indicates that hold-
ing the additional amount of time at temperature
T 1 results in an additional amount of austenite
transforming to lamellar pearlite. If additional
samples are removed from the salt bath at increas-
ing intervals of time and then quenched, it is found
that the quenched structure shows increasingly
greater amounts of lamellar pearlite and less and
less white martensite. This indicates that the longer
the specimens are held at temperature T 1, the more
austenite transforms to lamellar pearlite and less
remains to transform to white martensite on the
quench. At some time d, it would be found that
holding at temperature T 1 results in all of the aus-
tenite transforming to lamellar pearlite, as evi-
denced by the fact that the as-quenched microstruc-
ture consists entirely of lamellar pearlite with no
white martensite visible. Such progressive trans-
formation, occurring over a period of time at
constant temperature, is shown in Fig. 3-9.
This type of austenite transformation is known
as isothermal transformation. In Fig. 3-9, time

slightly greater than a or s shown on the lower
horizontal line represents the start of transforma-
tion, and time atfrepresents the finish of transfor-
mation at the isothermal transformation tempera-
tureofT I ·
Now if the process just described was repeated,
with the exception that the isothermal temperature
was raised to T 2 (as indicated in F"ig. 3-10), it ~ould
be found that the beginning of transformation
would occur somewhat later than in the previous
case. The product formed would be lamellar pearl-
ite, but a coarser lamellar pearlite than was formed
at temperature T 1. The passage of time at T 2 would
result in increasing amounts of austenite trans-
forming to coarse lamellar pearlite, and at some
later time, completion of transformation would be
realized. The time necessary for completion of
transformation at T 2 would be greater than that
required at T 1•
I.f the process was repeated using a transforma-
tion temperature T 3 (lower than T 1) it would be
found that lamellar pearlite again forms with the
passage of time. This lamellar pearlite would be
finer than that ,formed at temperature T l' and the
time necessary for both the beginning and comple-
tion of transformation at the lower temperature.
would be less than that encountered at T I .
At some still lower isothermal transformation
temperature, T4 , it would be found that the begin-
ning and completion of transformation occurs at
somewhat later times and that the product formed
is upper bainite. If the isothermal transformation
temperature w·ere dropped even lower, to tempera-
ture T s , lower bainite would form after an even
longer time interval. The "tudent should note that
all of the above also relates to Fig. 3-4.
Transformation
Tcl-------- ....... Product
T2 Coa,..
I~
T1
T3
T.
- - - - - - - - Fine
J..----.....iIiiiiIIIlI...----
Medium
Upper
Pearlite 30
~ ~~
ie T5 Lower
~ conditions, the spheroidite can be formed directly
Lo; Time - -......
Fig. 3-10. Effect oj time and temperature on transfor-
mation of austenite to lamellar pearlite and bainite.
11
By plotting the times of beginning and comple-
tion of transformation for each temperature inves-
tigated, and then joining these points, as in Fig.
3-10, a composite diagram can be constructed
which will show the start and finish of transforma-
tion at any intermediate temperature. In this fig-
ure, it will be noted that the transformation curves
become asymptotic to some temperature, Tc . This
is actually the critical temperature for the eutectoid
composition steel, and if the steel is held above this
temperature no transformation will occur regard-
less of the time of holding.
From these studies it is evident that the pearlite
and bainite reactions depend upon both time and
temperature. They occur relatively slowly over a
period of time and if the temperature is maintained
constant, a homogeneous structure will result. If,
however, the steel were allowed to transform over
a temperature range of T2 , T 1,T3 , a mixed struc-
ture of coarse, medium and fine lamellar pearlite
would result, the hardness of which would depend
on the relative amounts of each type present (see
eel" curves later in this lesson). It should be em-
phasized that transformation is terminal in nature
and irreversible. If austenite has been completely
transformed to a coarse lamellar pearlite at tem-
perature T2 , dropping the temperature to T 1 will
not cause the coarse pearlite to become medium
pearlite, nor can the coarse pearlite be changed to
fine pearlite or bainite by lowering the tempera-
ture. In a similar fashion, if austenite has been
completely transformed to a fine pearlite at tem-
perature T 3' it cannot be made into a medium
pearlite at T 1 or a coarse pearlite at T 2.
The only manner in which an existing structure
can be made into another transformation structure
is to reheat the specimen until it is austeni tic and
Rc then allow it to transform under the conditions
Hardness
15 which will result in the desired structure. While it
36
42
is true that spheroidite can be formed from either
pearlite or bainite, it should be emphasized that
spheroidite formed in this manner is merely a modi-
fication of the form of the cementite in an existing
transformation structure and is not a result of
~7 direct transformation from austenite. Under proper
by. transformation of austenite, in which case
spheroidite is the final and irreversible transforma-
tion structure and will not change to lamellar
pearlite as the temperature is lowered.
12

Ms Final Structure
I % Martensite - - - I I %Martensite 99% Bainite
It ..
20% ------#-~
40%-----""
20%
40%
.. 80cro
60% II
II
80% It
20%
~8f----, 90% .. 10% '1
100% . 0% II

100 %> Martensite

Mf

Time
Fig. 3-11. Effect of-remperature on transformation of austenite to martensite for carbon and alloy steel.

If the process used to determine the pearlite and ally form bainite, as indicated. However, with
bainite reactions is repeated using a still lower high carbon or alloy content the retained austenite,
temperature of transformation, T 6 (Fig. 3-11), the though theoretically unstable, may persist. The
product of transformation will be found to be martensite reaction is also irreversible, and if aus-
martensite. When the martensite forms, it does so tenite is transformed at such a temperature as to
instantaneously, and there is no incubation period form 40070 martensite, this amount of martensite
or diffusion, as in the case of pearlite and bainite cannot be reduced to 1% by reheating to a temper-
reactions. At T 6 , it will be noted, the small amount ature just below the Ms. In like fashion, the mar-
of martensite which has formed instantaneously tensite cannot be made into bainite or pearlite
does not increase with the passage of time. If the unless the martensite is first converted to austenite.
transformation temperature is lowered to T 7 (Fig.
3-11), an additional amount of martensite will It is sometimes confusing to understand how the
form instantaneously but will be unaffected by the martensite reaction is observed, since both the
passage of time. If this procedure is repeated over product of transformation and the matrix, or
a series of progressively lower transformation
'temperat~res, it will be found, as shown in Fig.
3-11, that the amount of martensite formed de-
pends upon the temperature of transformation and
is independent of time. The martensite reaction is
temperature-dependent only, as compared to the
pearlite and bainite reactions, which are both tem-
perature and time-dependent.
Ms and M f Temperatures. On all isothermal
transformation charts the temperature at which
martensite first begins to form is designated as M s
(martensite start), and the lower temperature at
which transformation is completed, or final, is the
M f (martensite finish). Some charts do not show
M f thus indicating the existence of retained austenite.
In Fig. 3-11 it is also seen that when a limited Fig. 3-12. Dark-etching tempered martensite in a ma-
amount of martensite forms, the balance of the trix of white martensite. (Micrograph by Anton des
austenite, when held isothermally, should eventu- Brasunas.)
 13

background, are martensite. When the specimens Austenite (Stable)

are removed from the isothermal transformation
bath, they contain some white martensite; the w
::c
remainder is austenite. If such a specimen were (1)
0
enc: Transformed
~

quenched in water to room temperature, the auste- :J 2

Q,)

nite would also transform to white martensite and 0

~
c:
Q,)

be indistinguishable from that formed first. If, Q)

0..
Cf)
:::I
however, the specimen is removed from the trans- E 4

formation bath and heated to some higher temper- (1)

t- M
ature, the white martensite, which formed initially,
will form dark-etching tempered martensite, while
the remaining austenite will be unaffected, and .
will transform normally to white martensite on Time
Fig. 3-13. Schematic isothermal transformation diagram
cooling to room temperature (note Fig. 3-12).
for eutecloid carbon steel.

It should be emphasized that the methods de- Ae3 A

scribed are not the only ones available for the study
of the changes which occur when austenite trans-
forms. When this happens, several properties of Ael
steel are altered. Among these is the magnetic
permeability, which varies considerably between
austenite and its products of transformation. By
t
I
CD
means of suitable electronic equipment, the changes
in magnetic permeability which occur during the
...e
~
::J
A
CD
. transformation of austenite can be measured. aa
E
Transformation is also accompanied by volume ~
changes which can be seen in buoyancy tests or Ms
translated into dimensional changes, and the prog-
ress of the transformation thus noted. Mf
I
(a) Time -..
Isothermal Diagrams. If Figs. 3-10 and 3-11 are
plotted as a composite, as in Fig. 3-13, the result is Acm
A
. ..: :.::.:: : :;: ::. :......... . ... .... .... _....
"'' ,.... ';:';;;:'A~C~':'!:'?::';~
~ ~:.:: ; ~::: ~ ;

a complete diagram of all the time and/or temper-

ature relationships which exist when austenite :1
transforms isothermally. Referring to the descrip- A el
I
tion of such diagrams involving Fig. 3-4 earlier in
this lesson, they are commonly termed as time- tI
CD
I
I
temperature-transformation (TTT) curves or S
curves from their resemblance to the letter S, or ...e
~
:J A I
IT (isothermal transformaton) curves. By using CD I
aa I
such diagrams as a guide various isothermal heat E
treating procedures can be designed to result in ~ I
Ms
Mflllllllll~~1
specific hardnesses and structures. In referring to
the isothermal transformation diagram, the upper
portion of the curve where pearlite is formed is I
called the pearlite nose, or more simply the nose of (b)
Time----..
the curve. (This area may also be called the knee of
the curve.) The area immediately above Ms and to Fig. 3-14. Schematic isothermal transformation diagram
the left of the line of beginning of bainite forma- for (a) a hypoeutectoid steel and (b) a hypereutectoid
tion is called the austenite bay (see Fig. 3-13). steel.
14

The isothermal transformation diagram devel- Effect of Alloys on Transformation

oped and shownin Fig. 3-13 was for a eutectoid
The isothermal transformation diagrams shown
steel. A similar diagram for a .hypoeutectoid steel
are schematic and are intended only to serve the
(a steel having less than 0.770/0 C) differs only
purpose of general description. The actual shape
slightly in general construction. It is known already
and time-temperature relationships of such dia-
that when a hYP0eutectoid steel transforms under
grams vary considerably with the carbon and alloy
equilibrium conditi,?ns above the nose or knee of
content of the steel. In general, the addition of an
the curve, ferrite is the first product of transfor-
alloying element to the steel makes the austenite
mation. This is also true when the austenite trans-
more reluctant to transform. As a result, transfor-
forms isothermally, so that to find an area of ferrite
mation may occur at lower temperatures and may
formation in the isothermal transformation dia-
require more time. The net result is that the iso-
gram for a hypoeutectoid steel is expected. Such a
thermal transformation diagram is shifted down-
diagram is shown in Fig. 3-14a. When a hypereu-
ward and to the right, as shown in Fig. 3-15, which
tectoid steel transforms isothermally (a hypereu-
depicts the effect of nickel content on the begin-
tectoid steel contains more than 0.80010 carbon), a
ning and ending of transformation. With the ex-
similar diagram will result, Fig. 3-14b, with the
ception of aluminum, cobalt and copper, other
exception that cementite now appears in the general alloying elements have this effect on transforma-
area where ferrite appeared in the hypoeutectoid tion. They may also cause various changes in the
steel. In the diagrams for both the hypoeutectoid general form of the TTT curves. These changes
and the hypereutectoid steel (Figs. 3-14a and b~, it fall into one of fOUf types of diagrams, as shown
is noted that the lines representing the beginning of
in Fig. 3-16.
either ferrite formation or cementite precipitation
become asymptotic to the upper critical tempera- The various alloy additions, except cobalt and
ture. The lines representing the beginning and aluminum, also have a depressing effect upon the
completion of pearlite formation become asymp- M temperature. This has been studied at length by
totic to the lower critical temperature. se~eraI investigators. Various methods, induding

LL'400
o

~IO
I 00 ~~ 0.94% Ni
1.94%Ni
~ 3.88% Ni
-t--

e 600~~~"""----'
Q)
0-
~ 200 L---~-+---......----,
I- O~-'----'--..- 0
10 100 1000 10,000 10 100 1000 100,00
Beginning of Transformation Ending of Transformation

tC = (tF -32) x 0.555 Time - Seconds

Fig. 3-15. Effect of increasing alloy content (nickel) on isothermal transformation of o. 60% carbon steel.

Transformation Begins
Time (Log Scale)
Fig. 3-16. Four general types oj isothermal transformation diagrams.
formulas, were devised so that the exact location
of M s could be determined for any alloy steel.
The approximate location of Ms and M f as a func-
tion of percent carbon may be obtained from the
graph below Table 3-2. Table 3-2 shows the rela-
tive effect of various alloys on the Ms ; carbon is
most effective. The M f temperature is even more
strongly depressed by alloy content. The effect of
carbon content is shown in the graph which accom-
panies Table 3-2, for both Ms and Mf •
The addition of alloying elements has one more
effect that is worth noting. In a plain carbon steel
the eutectoid composition is 0.770/0 carbon. With
15 .
Form
the addition of alloying elements, the carbon con-
tent of this eutectoid composition is lowered to the
extent indicated in Fig. 3-17. While alloying ele-
ments lower the eutectoid carbon content, they do
not affect the reaction or the product which forms
at the eutectoid temperature, although tempera-
ture itself may be affected, as in Fig. 3-17.
Transformation of Austenite
Under Continuous Cooling
The isothermal transformation diagram is a very
useful tool for designing heat treatirig procedures
using isothermal transformation methods. Because
 16

.Table 3·2. Effect of Alloy on the M s Temperature of limitations of equipment and other factors,
of a 1 % Carbon Steel for Each 1 % of Alloy Added however, it is not always possible to perform the
Change in Ms heat treatment under conditions of isothermal
temperature
transformation, and in most processes a continu-
Element F C ous type of cooling, at some given rate, must be used.
Carbon. . . . . . . . . . . . . . . . . . . . . . . . . . .. -510 -285
Manganese. . . . . . . . . . . . . . . . . . . . . . .. - 60 -33
Chromium. . . . . . . . . . . . . . . . . . . . . . . .. - 40 - 22 Derivation of Continuous Cooling
Nickel. . . . . . . . . . . . . . . . . . . . . . . . . . . .. - 30 - 17 Transformation Diagram
Molybdenum. . . . . . . . . . . . . . . . . . . . . .. - 20 - 11
Tungsten. . . . . . . . . . . . . . . . . . . . . . . . .. - 20 - 11
A significant amount of information required to
Silicon. . . . . . . . . . . . . . . . . . . . . . . . . . .. - 20 - 11
Cobalt. . . . . . . . . . . . . . . . . . . . . . . . . . . .. + 10 + 6 construct a continuous cooling transformation
Aluminum. . . . . . . . . . . . . . . . . . . . . . . .. + 30 +17 (CCT) diagram is obtained by metallographic in-
vestigation of the microstructures formed along
1000,----.-,.....-....,....-,---,-_r--........---.,....-...,....-... °c the surface of an end-quenched hardenability bar.
~ (Lessons 6 and 7 provide complete information on
800
LL 400 laboratory techniques and application of data for
0' .. 600 the end-quenched hardenability bar.) The progres-
e 300
~

:z 400 "- sion of decreasing cooling rates along the bar re-
'0 "- "- 200 sults in forming various microstructures ranging
i.E zoo " ',,-~. 100 from as-quenched martensite at the 'quenched end
x
~ 0 "- o to softer mixed structures such as ferrite-pearlite-
"- bainite at a distance of several inches from the end.
"-
-200~~_.r----&.._.a....-
......._,~--'-_6000..--'"----J ~OO
02 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 I. f 1.2
0
Original investigators constructed the CCT dia-
Carbon, /0
gram from end-quench data by superimposing it
on the IT diagram to depict the shift downward

Table 3·3. Formation Temperatures of

Austenite Transformation Products (a)
Eutectoid Eutectoid Ferrite Pearlite Bainite Martensite
Steel Fe Fe F C Fe
Carbon-ok Temperature - 0 C
3140 1180 640 1100 595 970 520 625 330
130 a ,.--.,.---r----l~-....-.----..-....,..--, 4024 1400 760 1200 650 1060 570
2200 lL..
4047 1240 670 1100 595 1100 595 625 330
4140 1290 700 980 525 670 355
Iroo '--+----+--~2000 ~ 43BV14 1280 695 890 475
1800 ~ 4340 890 475 560 295
900 p..-........-+---t 1600 15 4340 + Si 840 450 740 395 545 285
~
Q)
4640 1100 595 970 520 900 480 570 300
1400 E 4815
5140
1240
1250
670 1000 540
675 1170 630 . 940 505 635 335
1200 ~
5160 1170 630 1100 595 970 520 535 280
500 ~--'---'----'--.-......-~~ 1000 52100 1250 675 1250 675 - 470 245
0.80 8620 1380 750 930 500 -
~~d--~C+-----+--~ 8640 1220 660 1060 570 675 355
86B40 1070 575 880 470 610 320
0.40 9260 1310 710 1190 645 1110 600 480 250
o Ti Mo 9840
MoB
950
1370
510
745
870
1190
465
645
610 320

o 4 8 12 (6 55
01
1480 805 760 405 11 00 595 515 270
Alloying Elements - '1'0 1330 720 1130 610 1090 585 530 275
(a) The figures indicate temperatures of initial formation of
Fig. 3-17. Effect of alloying elements on eutectoid tem- transformation products 0.5 in. (13 mm) from the quenched
perature and eutectoid carbon content. ends of Jominy hardenability specimens.
 17

END-QUENCH HARDENABlllTY TEST

I
o·
~ /

'"oz /
'3
a:/
c
.~ ~O I.

" . 0 / 05 I 0 / I 5 2. 25 0
.~ __/._I_' ._t~~ANCE.~.~C~~.ND-IN~HES __ . • __ • .

1400

12
I
....,

800 -

®~ ®- .....©
200 - MartenSITe Martensite, Ferrite Martensite, Ferrite
and Bainite and Bainite
Cooling Transformation diagram
Isather mal transformation diagram
o - Cooling curves
TransformaTion during cooling

2 5 10 10 2
Ti me - S eco~ds

Fig. 3-18. Correlation of continuous cooling and isothermal transformations with end-quench hardenability test
data for 4140 steel.

and to the right of isothermal temperatures and
time, as shown in Fig. 3-18. More recent experi-
mentation has resulted in construction, of separate
CCT diagrams as shown in Fig. 3-19. The cooling
curves derived from the cooling rates at various
positions along the end-quenched bar are desig-
nated by the linear distance from the end of the bar
in fractions of an inch, such as 1/16 in., 1/4 in.
or 1/2 in.
uous cooling in a number of alloy steels. To pro-
duce pearlite in the microstructure, it is necessary
to process under isothermal conditions. Similar
data have been obtained experimentally for other,
positions along the bar. Compilation of the data
is then used to construct the CCT diagram and
heat treating procedures can be developed to
produce a required microstructure and control the
resulting specified mechanical properties.

Table 3-3 provides temperature data for the
initial formation of austenite transformation prod-
ucts at the ~-in. position on the end-quench bar.
Microstructural transformation products are
ferrite, pearlite, bainite and martensite. It can be
noted that pearlite does not form during contin-
Continuous Cooling Transformation
Diagram for AISI 4024 Steel
For AISI 4024 steel, decreases in cooling rate
have the following effects which may be observed
in curves of Fig. 3-19a: Fs (ferrite start) is essen-
18

1600 ....-.-~~~ _ _~~-.:;~~~_ _-+-- -+-_ _--+ 0.24 C, 0.88 Mn, 0.33 Si, 0.23 Mo
- Austenitized at 1700 F
Grain size No. 8
AC3 = 1520 F
ACI= 1380 F
1400 I----+-~l____\_---+--+---+----d~-_... ..............~~:__:_:-__+--___+----r__-~

Cooling curves from 1700 F

at indicated distances
from quenched end
I
~ 1000 t - - - - -
.2
~
Q)
c::4
E
Q)
I--

800 t------+-------...~~.....;..._

600

A-austenite
F-ferrite
400 P-pearlite
B-bainite
M-martensite

200
1 2 5 10 20 50 100 200 500 1000
(a) Cooling Time, Sec
260
0.24 C, 0.88 Mn, 0.33 Si, 60
0.18 Ni, 0.23 Mo
I I
I 0.24 C, 0.88 Mn, 0.33 Si. 0.23 Mo
220 Normalized, 1750 F Quenched from 1700 F
Austenitized, 1700 F
0v;
0... 180
Quenched in agitated oil
Tempered, 1 hr ---+------f
\
o
<::>
C>
I I
Tensi Ie Strength \ \.
~ 140 ~.......~Red~ction
Zi5

100
in Area

~-';I-----+-.;;;II~-~ 40 ~ .:c:::
80
"'C
c:::
~c::

~
CD 0
:~
~.
Q.)

10
'" ~ .......
---N--
I
I

i --H-- ~
c::: :.=
o <.:J
L::j.g
I
II
60 _-----''--- _ Q.)
c:: u I
o 4 8 12 16 20 24 28 32 36 40
(b) (c) Distance From Quenched End. 1/16 'no

Fig. 3-19. Cooling curves for medium carbon steel. (Homer Research Laboratories, Bethlehem Steel Corporation.)
 19

tially constant beyond the 1/ 2-in. posItIon; Ps The photomicrograph in Fig. 3-20a shows that
(pearlite start) reaches a maximum at 3/8 in. and XO% transformation occurred within 15 sec at the
then drops; M s (martensite start) remains steady ~-in. position. (The micrographs are correlated
up to about the 5/16-in. position and then drops with numhcrs on the CCT diagram in Fig. 3-19a.)
sharply. These changes are reflected in the micro- Transforn1ation was 95% complete in 30 sec at
structures shown in Fig. 3-20. The curves in Fig. the '~-in. rosition (photomicrograph b). Photomi-
3-19a indicate that transformation to 10% and 50% crographs c and f show that the ferrite becomes
nonmartensitic products occur at about the 1/ 8-in. more blocky and massive as the cooling rate de-
and 3/16-in. positions, respectively. The dotted creases.
portions of the F s , P s ,. Bs and Ms curves were de-
termined by metallographic data.

...".' "
••. <•. ~(. > y'" ,. ,. ,,,...,,~' ~-...-

... A....

i
"
.., ....
~<:
~/
...... -"

'·;(d")
i .. J.I,. ~ /

~/, .. ' . ..-,.

./

:.1 .'............",''-.

,:=~...
(
.,.,......
.....
_". ./

~ ~..
y ": >JII._
~.~i~\·.·" ,,~~,
ri....'".
'¥..
. .•'.. ~.?S
~~... ..-~,..
",> ~. '..• .• .1:;'r£~
._~ *~.~
£ " •... . . <- f.:• .~~.'~
.•.'.•..•...•. ,•. .:~. ',-'
... ;~.~~~'!:i.<...;»""'''''''f:.
J
. ..
.•:

1'.~,t,· \'- "-~).~"«''; ;,:(~ •.~..;

~e:~~~}~~~~
Fig. 3-20. Micrographs of structures at points indicated in Fig. 3-19. (a) Point 1 at 114 in., 30 sec. (b) Point 2
at 114 in., 100 sec. (c) Point 3 at 318 in., 15 sec. (d) Point 4 at 7116 in., 15 sec. (e) Point 5 at 1 in., 40 sec. (f) Point
6 at 1 in., 50 sec. (Courtesy, Homer Research Laboratories, Bethlehem Steel Corporation.)
 ·20
The hardenability curve in Fig. 3-19c indicates a
relatively low hardenability value. Further, the
tensile and yield strengths rise to a maximum upon
tempering between 600 and 700 F (315 and 370 C)
and then drop (see Fig. 3-19b), probably because
some retained austenite is being transformed.
Retained Austenite
In a discussion of austenite and its transforma-
tion, the subject of retained austenite must also be
considered. Retained austenite is that austenite
which has survived a heat treating cycle in which it
would, for lower carbon and low.alloy steels, be
expected to transform to some other product, al-
though this condition changes when carbon and/or
the alloy content is increased. After the transfor-
mation reaction is partly complete, the retained
austenite is found to exist along with the partially
transformed products which formed during cooling.
Reasons for Retained Austenite
Austenite may be'retained in a microstructure as
a result of carbon and/or alloy content, quenching
temperature and austenitizing temperature.
Effect of Alloying Elements. Earlier discussion
noted that the addition of carbon and/or alloying
elements resulted in lowering the Ms-M f tempera-
tures. If sufficient carbon and/or alloying elements
are added, the M f temperature will be below room
, tempera~ure. In this case the bainite reaction is
also very slow, so that if martensite does not form
the austenite remains untransformed for long peri-
ods of time. Since most heat treating procedures
are terminated at room temperature, or slightly
above, the effect of carbon and/or alloying ele-
ments in lowering the M f to below room tempera-
ture can readily be understood in its relationship to
retained austenite. The effect of alloying elements
is so strong that in some cases the M s temperature
can be suppressed so far that the steel is entirely
austenitic at room temperature. Such steels are
referred to as austenitic steels, and should not be
confused with steels which contain retained auste-
nite unintentionally.
Effect of Quenching and/or Austenitizing Tem-
perature. If a steel is quenched from a series of in-
creasingly higher austenitizing temperatures, the
Table 3-4. Effect of Austenitizing
Temperature on Retained Austenite
Austenitizing temperature,
F C Retained austenite, 0/0
1550 845 6.4
1700 925 12.4
1900 1035 47.0
amount of retained austenite increases as the aus-
tenitizing temperature increases. The effect of
austenitizing temperature upon retained austenite
is shown in Table 3-4 for a \vater-quenched 0.85010
carbon steel containing about 2070 of other alloying
constituents.
Increasing the austenitizing temperature for a
given steel r.esults in increased amounts of retained
austenite because of the greater degree of solution
of carbon and other elements in the austenite real-
ized at the higher austenitizing temperatures. If,
on quenching, the cooling rate in the Ms-M f range
is slowed down, causing stabilization, there is a
greater tendency to retain austenite than if the
cooling is continued at a rapid rate. This explains
why more retained austenite is often found in oil-
quenched steels than in water-quenched steels of
the same grade (the oil is slower cooling in the
lower temperature ranges), and why retained aus-
tenite is sometimes found in relatively slowly cooled
parts. The amount and distribution of stress in a
quenched piece may also have a bearing on the
amount of retained austenite.
Problems Associated With
Retained Austenite
Many instances exist where a reasonable amount
of retained austenite is not significantly detrimen-
tal, and in some rare instances some retained aus-
tenite has proven to be advantageous.
For the most part, however, for one or more of
at least three reasons, significant amounts of re-
tained austenite are undesirable.
First, the primary purpose of quenching a steel
from the austenitic state is to form a hard, strong
structure, and the presence of any retained auste-
nite, which is softer than martensite, will lower the
overall hardness of the steel. The presence of re-
tained austenite may lower the hardness by several
points, a considerable amount of retained auste-

nite in a quenched part where, for example, the
difference between 63 and 59 HRC may render a
piece useless for service.
Second, the presence of retained austenite may
be the cause of dimensional instability in heat
treated tools, gages, or working parts. The insta-
bility is brought about because, under certain con-
ditions at room temperature and over long periods
of time, the retained austenite can transform to
some extent to martensite. This transformation is
accompanied by volume changes which in turn
affect the dimensional stability, an undesirable
condition in many applications.
Third, under certain' conditions of use, the re-
tained austenite can transform to martensite. This
is, of course, white martensite which is very hard
and brittle and may be the cause of cracking in
operations such as grinding. It is suspected to be
the source of spontaneous cracks and other service
failures.
Elimination of Retained Austenite. Several
means are available for eliminating or at least min-
imizing the amount of retained austenite. Temper-
ing at 300 F (150 C), or higher if permitted by
mechanical property requirements, will at least
reduce the amount of retained austenite. For the
higher alloy steels double or triple tempering helps
to reduce the amount of retained austenite.
Another procedure, usually used in conjunction
with tempering, is to cool the steel to subzero tem-
peratures - often as low as -150 F (-100 C). The
steel is thus cooled near or below its M f tempera-
ture, and transformation will proceed.
Mechanical deformation of the steel has also
been used as a· means of transforming retained
austenite because it is in a metastable condition. In
this case, martensite is formed by applied stress.
Effect of Alloying
Elements on Transformation
The subject of alloying elements and their ef-
fects upon mechanical properties and the transfor-
mation of austenite has been recurrent in these
discussions. The question might well arise as to
exactly why various elements are added to the
steel. Certainly their presence complicates matters
21
of transformation and some benefit must be real-
ized or one would not go to the expense of adding
them to the steel. Alloying elements are added to
the steel to impart some property which could not
be secured by adding carbon alone and/or by heat
treatment. How the alloys behave in the steel is a
. complex matter, and a thorough consideration of
it is beyond the scope and space.limits of this text.
It will, however, be profitable to review briefly the
more common alloying elements and their major
effects in steel. I
Alloying elements may be presen t in steel in
several forms, that is, they may (a) enter into solid
solutions with the ferrite, (b) combine with carbon
and form both simple and complex carbides, (c)
form intermetallic compounds, or (d) be present as
nonmetallic inclusions. The presence of alloying
elements affects both the heating and cooling criti-
cal temperatures as well as the general transforma-
tion characteristics. It is this fact which makes heat
treatment as complex as it is, for each composition
of steel must be considered as having its own trans-
formation characteristic, and the heat treating
cycle is tailor made to fit. In all cases, however, the
principles remain the same, although certain minor
modifications may be necessary.
Most of the alloying elements considered here
will increase the depth to which a piece of steel will
harden. This is brought about mainly by the influ-
ence of the alloy on the transformation character-
istics of austenite, allowing the steel to form mar-
tensite with slower rates of cooling. Individually,
the alloys have other diversified effects which are
discussed in the paragraphs that follow. Th.e vari-
ous alloys (or additions) made to conventional
steels are discussed in their alphabetical order.
Aluminum is not used as an alloy per se in any
but a few highly specialized compositions. These
special steels contain approximately 1.0070 alumi-
num and are used for surface hardening by the
nitriding process because under certain conditions,
aluminum combines with nitrogen to form hard
nitrides in the surface layers. However, the amount
of aluminum used in these specialty steels does not
significantly affect the transformation characteristics.
The major importance of aluminum in steel is its
ability to control grain size. Aluminum, when
added to molten steel, combines readily with any
·22

oxygen present in the molten state and therefore is
used as a deoxidizing agent to rid the steel of unde-
sirable dissolved oxygen and ferrous oxide. In the
process of deoxidizing (sometimes referred to as
killing the heat), many small, solid aluminum oxide
particles are formed. These have a high melting
point and remain as solids in the liquid steel. When
the steel solidifies, the oxides are mechanically
entrapped and act as nuclei around which the indi-
vidual grains form. Since the nuclei are numerous
in aluminum killed steels, many grains form, and
the average size of the grain is much smaller than
would be the case in another steel where fewer
nuclei are present. The steel is then said to be fine-
grained as compared to coarse-grained steel. The
effects of grain size on properties are ~umerous
and have an influence in steel whether it is auste-
nitic, annealed or quenched.
Grain size is so important in steel that a compre-
hensive method of measuring and reporting it has
been developed. This procedure is shown in Fig.
3-21 (see also Table 3-5).
Chromium affects the properties of steel in a
number of ways. It has a marked effect in slowing
down the rate of transformation of austenite; that
is, it greatly increases the hardenability of any
steel. In addition to this effect is the effect of rela-
tively large percentages of chromium on resistance
to corrosion and oxidation at elevated tempera-
tures. The resistance of chromium steels to corro-
sion and high temperature oxidation is intensified
by the addition of nickel. The stainless steels all
contain large amounts (up to 28070) of chromium.
Chromium also readily forms carbides which are
hard and wear-resistant. It is for this reason that
chromium is used in many working steels for such
applications as bearings, drills, chisels and dies.
Chromium is also used in steels which are exposed
to service at high temperature, since the chromium .
confers some high temperature strength. Chro-
mium is one of the most universally used alloying
elements and finds very broad application, gener-
ally in conjunction with other alloying elements.
Cobalt. The effect of cobalt on the transforma-
tion of austenite is similar to that of nickel (see
Fig. 3-15). The use of cobalt is usually restricted to
high alloy tool steels or to heat-resisting steels. It
adds materially to the life of a tool by increasing
the red hardness. Red hardness is a property that
maintains hardness and cutting ability although
the tool is heated to a dull red during the machin-
ing operation. A second use of cobalt is in certain
permanent magnet steels which are noted for their
ability to retain their magnetism.
Copper is not used extensively as an alloy in steels
except for the HSLA grades (high-strength low-
alloy), where it is used to increase resistance to

0000000 atmospheric corrosion and to increase strength due

to precipitation hardening. The amounts of copper

I 2345678 Table 3·5. Effect of Grain

Size on Properties of Steel
ASTM Grain Size Num b er Condition Improvement in property caused by

(Mean Grain Size atlOO X) of steel Fine grain

Austenitic .. Plastic form-
Coarse grain

Grain Number of Grains per

ability
Range of Grain Size
Size Sq. Inch at lOOX Decomposition
per Sq. Inch at 100X
of austen ite
1 1 0.75 to 1.5 (more rapid)
2 2 1.5 to 3.0 Annealed ... Cold working Machinability
3 4 3 to 6 Ductility Hardness
4 8 6 to 12
Impact Creep resistance
5 16 12 to 24 resistance
6 32 24 to 48 Quenched .. Impact Depth of harden ing
7 64 48 to 96 resistance
8 128 More than 96 Retained auste-
nite (less)
Fig. 3-21. Standard for evaluating grain size in steel. Distortion (less)
The grains shown are idealized. Cracking (less)

usually used do not significantly affect austenite
transformation.
Lead does not dissolve in steel and cannot, there-
fore, be correctly termed as an alloy. However,
amounts of lead in the range of 0.15 to 0.35070 are
commonly added to steel to improve machinabil-
ity. The lead is present in the form of a dispersion
of fine particles. Since it does not dissolve it does
not affect transformation of austenite.
Manganese is an important alloy in steel for sev-
eral reasons and is present in virtually all steels in
amounts of 0.30070 or more. Manganese is a car-
bide former and has a marked effect on slowing
down the gamma-to-alpha transformation. Fur-
ther, manganese in steel is important because of its
ability to counter hot shortness; that is, the tend-
ency to tear when being hot formed. In steel, iron
and sulfur combine to form a sulfide which has a
relatively low melting point. When the steel freezes,
this sulfide solidifies in the grain boundaries. Upon
subsequent reheating for rolling or forging, the
sulfide melts. The steel is then weakened, and
when it is hot worked by rolling or forging it tears
or breaks apart.,· If manganese is added, it com-
bines preferentially with the sulfur and forms a
manganese sulfide of higher melting point, which
by its distribution and nature eliminates hot shortness.
Manganese in large amounts forms a very useful
alloy steel known as Hadfield's steel. This steel,
which is austenitic at room temperature, has the
ability to work harden drastically. It is useful in
applications where hard materials are crushed or
conveyed.
Molybdenum, like chromium, greatly slows
down the gamma-to-alpha transformation and is
generally more powerful than chromium. In addi-
tion, molybdenum enhances the corrosion resist-
ance of stainless steels, raises hot strength, affords
good creep resistance, and gives the steel high red
hardness. The high temperature properties of
molybdenum steel result in its being used to a large
extent in dies and tools intended for the hot work-
ing of metals. Molybdenum increases the tough-
ness of steels.
Nickel has a marked effect upon the transfor-
mation of austenite by shifting the nose of the TTT
curve to the right, and further, by significantly
lowering the gamma-to-alpha transformation tem-
23
perature even to the point where the steel is auste-
nitic at room temperature when large amounts of
nickel are used.
Nickel is a very versatile alloy. It provides an
added degree of uniformity to quenched steels and
strengthens the unquenched or annealed steels. It
toughens ferritic-pearlitic steels, especially at low
temperature, and gives good fatigue resistance.
Nickel greatly increases corrosion and oxidation
resistance of stainless and heat resisting steels.
High-nickel steels are also used extensively in cryo-
genic applications.
Phosp~orus, in the amounts found in carbon
and low-alloy steels, has no significant effect on
the transformation characteristics of austenite.
Phosphorus strengthens low-carbon steels and
increases resistance to corrosion, for which reason
it is deliberately added to certain HSLA steels. In
most instances, however, phosphorus is an inci-
dental element in steels and generally regarded as
an impurity. It is sometimes added to increase
machi~ability of the resulfurized free-machining
steels (see Table 3-5).
Silicon, to a relatively mild extent, slows down
the rate of beginning transformation from auste-
nite. Silicon is used as a deoxidizer in making steel
- singly or in conjunction with aluminum. When
present in relatively high percentages (as much as
5.0070), it has important effects upon the electrical
and magnetic properties. The high silicon steels
have high magnetic permeability and high electri-
cal resistance and form several grades of trans-
former steel. The presence of silicon improves the
oxidation resistance, notably in stainless steels. In
low-alloy steels silicon has a marked strengthening
effect, and the silicon structural steels have seen
considerable use. Silicon also improves shock re-
sistance and is used where impact is a problem, as
in chisels, gouges and punches.
Sulfur is usually found in steel as an incidental
~lement and, like phosphorus, is generally regarded
as an impurity. Sulfur does not influence transfor-
mation, but usually combines with manganese or
iron to form sulfides which exist as finely dispersed
segregates.
Sulfur is deliberately added to some grades of
steel in amounts up to about 0.35070 to improve
machinability.
24
Tungsten, like chromium and molybdenum, has
a marked effect in slowing down the beginning of
the gamma-to-alpha transformation. Thus, signif-
icant amounts of tungsten can slow down the criti-
cal cooling rate to the point where the steel may be
fully hardened by air cooling from the austenitic
phase. As a rule, tungsten is used only in tool or
other specialty steels.
Tungsten in steel forms hard, abrasion-resistant
carbides. Tungsten also promotes red hardness
and hot strength. This combination of properties
makes tungsten steels very useful for high-speed
cutting tools. They can cut ordinary steels when
the tool is hot enough to be a dull red in color (red
hardness).
Vanadium is a strong deoxidizing agent and pro-
motes fine grain size, which accounts for the fact
that up to 0.25070 vanadium is commonly added to
water'-hardening tool steels. Vanadium is also a
powerful carbide former, resulting in extremely
hard, abrasion-resisting carbides. Hardened vana-
dium-containing steels also resist softening from
reheating as during tempering. Vanadium is sel-
dom used in any significant amounts alone, but is
usually used in combination with other alloying
elements such as chromium and/or tungsten. Sig-
nificant amounts of vanadium make the gamma-
to-alpha transformation more sluggish.
Summary
While alloying elements. are sometimes used
alone, they are more often used in combinations of
two or more because it has been found that the use
of smaller amounts of two or more elements is
more effective than large amounts of a single ele-
ment. For instance, chromium is generally used
with nickel, molybdenum, or some other element.
Regardless of the type or amount of the alloy
added, the general principles 'of austenite forma-
tion and transformation remain the same, although
the reference points, such as the critical tempera-
tures and the time and temperature of transforma-
tIon, vary with the type and the amount of alloying
element.
Classification of Steels
Products made from steel find extremely varied
use in the modern world. As a result of the widely
differing service applications, a diversified range
of physical and mechanical properties is required.
This is obtained by adding alloys to the steel and
then subjecting it to various heat treatments. No
one composition of steel could possibly fit all ser-
vice applications and, over the year, a great num-
ber of different grades have been developed, each
having a definite field of usefulness in certain
types of service applications. The many different
grades of steel are assigned designations both gen-
eral and specific in nature. These may be listed as
carbon steels, alloy steels, tool steels, and the stain-
less and heat resisting steels. Each broad classifica-
tion includes numerous steels of specific chemical
composition.
Carbon Steel
Steel is considered to be a carbon steel when:
1. No minimum content is specified or required
for aluminum, boron, chromium, cobalt,
columbium, molybdenum, nickel, titanium,
tungsten, vanadium, zirconium, or any other
element added to obtain a desired alloying
effect.
2. The specified minimum for copper does not
exceed 0.40070.
3. The maximum specified content does not
exceed the following limits: manganese,
1.65070; silicon, 0.60070; and copper, 0.60070.
Steels are sold to definite chemical limits, and
each grade of plain carbon steel is assigned a code
number which specifies its chemical composition.
There are several coding systems, but the AISI
(American Iron and Steel Institute) system is the
most comprehensive and widely used. It uses a
code of four, or in some cases five digits. The first
two always indicate the general type of steel. For
instance, consider a steel identified as AISI 1040
grade. The digits 10 indicate it to be a plain carbon
steel with a manganese content of no greater than
1.0070. The next two digits, 40, indicate the mean
carbon content, in this case 0.40070. Thus, when
the code 1040 is seen, the reader knows the steel is
a plain carbon grade containing 0.40070 carbon.

Because of differences in properties, and in par-
ticular their response to heat treatment, the higher
manganese carbon steels have been removed from
the 10xx series and assigned to the 15xx series.
These steels are marketed as carbon steels, but
because of their higher manganese content, they
respond to heat treatment as do some alloy steels.
The same method of designation prevails for the
15xx series; that is, the digits 15 denote higher
manganese (up to 1.65070), and the last two digits
designate the mean carbon content. For example,
steel 1547 has a carbon content of 0.43 to 0.51 070.
In carbon steels, for each element specified there
is a permissible range of composition, or a maxi-
mum limit, rather than a single specified value.
This is because it is difficult to melt to the exact
desired chemical percentage of each element; some
leeway must be allowed for practical purposes.
Manganese, phosphorus and sulfur appear in the
composition of every commercial steel either as
residual elements or alloys. In the case of manga-
nese, the amounts present usually require that this
element be purposely added. Phosphorus, sulfur
and silicon are useful in their own right in special
applications. Generally, however, they are unde-
sirable and are considered impurities which have
been carried along in the materials used to make
up the steel. Since their complete removal is quite
expensive, their presence is reduced to some nomi-
nal amount (0.040/0 maximum or less for P and S),
which renders them practically ineffective as far as
the properties of the steel are concerned. They are
then carried along in this reduced amount rather
than eliminated entirely.
Free-Machining Carbon Steels. There are pres-
ently two different series of free-machining carbon
steels. First, is the series assigned AISI llxx, re-
ferred to as resulfurized carbon steels. The carbon
content ranges vary from 0.08 to 0.13070 carbon
for AISI 1109 to 0.48 to 0.55070 carbon for AISI
1155. Several members of this series also contain
a relatively high manganese content (as high as
1.65070 for 1144), and sulfur contents as high as
0.33070. Phosphorus content for this series is,
however, restricted to a maximum of 0.040070,
and these specific steels are normally available
with1ead additions.
A second series of free-machining carbon steels
is identified as rephosphorized and resulfurized
carbon steels, and is coded as AISI 12xx.
25
All of the 12xx grades are low in carbon content
(0.15070 maximum). Also, the maximum manga-
nese content is "generally lower compared with the
l1xx series. Sulfur content is also high in the 12xx
steels (0.35070 maximum for two grades), but the
outstanding difference between the 11 xx and 12xx
series is that the steels of the latter series have been
rephosphorized as well as resulfurized, for better
machinability, than can be provided by resulfuri-
zation alone. One grade, 12L14, also contains
lead, which further enhances machinability.
The nonresulfurized grades of carbon steel are
available with lead additions of 0.15 to 0.35070 for
improved machinability. Although lead additions
do not provide the degree of improved machinabil-
ity that can be provided by resulfurization and/ or
rephosphorization, neither is there as great a sacri-
fice of mechanical properties for lead additions
compared with sulfur and/or phosphorus addi-
tions. In all instances leaded steels are designated
by inserting the letter "L" between the second and
third digits of the AISI number. For example,
1045 steel with a lead addition is written as 10L45.
Boron Treated Steels. Because of the very small
amount of boron used (0.0005 to 0.003070), steels
containing boron to increase their hardenability
are usually termed as boron treated instead of
regarding boron as an alloying element. Carbon
steels which contain boron are identified by insert-
ing the letter "B" between the second and third
digits of the AISI number. For instance, a boron
treated 1035 would be written as 10835.
Alloy Steels
Steel is considered to be an alloy steel when
1. The maximum of the range given for the
content of alloying elements exceeds one or
more of the following limits: manganese,
1.65070; silicon, 0.600/0; or copper, 0.60070.
2. A definite range or a definite minimum quan-
tity of any of the following elements is speci-
fied or required within recognized limits:
aluminum, boron, chromium up to 3.990/0,
cobalt, columbium, molybdenum, nickel,
titanium, tungsten, vanadium, zirconium, or
any other element added to obtain a desired
alloying effect.
·26

Table 3-6. Numerical Designations of When lead is specified, the lette.r L is inserted
AISI and SAE Grades of Alloy Steels after the first two numbers; for example, 41L30.
Series Type and approximate percentages
designation of identifying elements Many of the alloy steels have a more or less spe-
13xx Manganese, 1.75 cific field of usefulness and serve as specialty steels
40xx Molybdenum, 0.20 or 0.25 for certain types of application. For instance, the
41xx Chromium, 0.50, 0.80 or 0.95; molybde-
n.um 0.12, 0.20 or 0.30
52100 grade is used almost exclusively for antifric-
43xx Nickel 1.83, chromium 0.50 or 0.80, mo- tion bearings, and the 9200 series for springs and
lybdenum 0.25 applications where shock resistance is a factor. In
44xx Molybdenum 0.53 many instances, however, a single composition of
46xx Nickel 0.85 or 1.83, molybdenum 0.20 steel will serve a wide range of applications; for
or 0.25
47xx Nickel 1.05, chromium 0.45, molybdenum
example, the 4340, 8640, and 8740 grades.
0.20 or 0.35
48xx Nickel 3.50, molybdenum 0.25 H-Steels. Many of the standard grades of alloy
50xx Chromium 0.40 steels can be obtained that are guaranteed to meet
51xx Chromium 0.80, 0.88, 0.93, 0.95 or 1.00 the requirements of the AISI hardenability bands.
5xxxx Carbon 1.04, chromium 1.03 or 1.45 Such steels carry the suffix letter H; as 4140H.
61xx Chromium 0.60 or 0.95, vanadium 0.13 or Therefore, if a steel must meet standard harden-
0.15 min.
86xx Nickel 0.55, chromium 0.50, molybde- ability requirements, it should be so specified.
num 0.20 However, in most instances the standard grades
87xx Nickel 0.55, chromium 0.50, molybde- will meet, or at least come very close to meeting,
num 0.25
the hardenability limits established by AISI.
88xx Nickel 0.55, chromium 0.50, molybde-
num 0.35 Stainless Steels
92xx Silicon 2.00
50Bxx Chromium 0.25 or 0.50 The AISI coding system for stainless steels con-
51Bxx Chromium 0.80 sists of three digits. The first digit of the code des-
81Bxx Nickel 0.30, chromium 0.45, molybde-
num 0.12 ignates a general grouping, and the second two
94Bxx Nickel 0.45, chromium 0.40, molybde- digits indicate certain information on the alloy
num 0.12 content, as gi,ren in Table 3-7.
B denotes boron steel.

Tool Steels
The alloy steels are also identified by the AISI
coding system, using the same general procedure Tool and die steels encompass a broad range of
described for carbon and resulfurized steels. The compositions. The standard AISI classification
first two digits of the four-numeral series for the groups them into seven major headings, as shown
various grades of alloy steel and their meaning are in Table 3-8. Tool steels are usually melted in rela-
given in Table 3-6.
The last two digits of the four-numeral series are Table 3-7. Numerical Designations
of AISI Grades of Stainless Steels
intended to indicate the approximate middle of the
Series
carbon range. It is necessary, however, to deviate designation Group
from this rule and to interpolate numbers in the
2xx Chromium-manganese-nickel steels; com-
case of some carbon ranges, and for variations in parable to the 300 series.
manganese, sulfur,·chromium or other elements.
3xx Chromium-nickel steels; nonhardenable,
austenitic, nonmagnetic.
The prefix letter E is used to designate steels
made by the basic electric furnace process. Steels 4xx Chromium steels; hardenable, marten-
sitic, magnetic.
without the prefix letter are normally manufac-
5xx Chromium steels; nonhardenable, ferritic,
tured .by the basic open hearth or basic oxygen magnetic. .
process. 6xx Precipitation-hardenable steels; stainless
steels which incorporate the further
When boron is specified, the letter B is inserted' addition of AI, Ti, Cu and/or other al-
after the first two numbers; for example, 41B30. loying elements to strengthen by aging.
 27

Table 3-8. AISI Groupings purpose or general type. This classification is

and Types of Tool Steels shown in Table 3-8. Some alloy and carbon steels
AISI resemble the tool steels very closely from the view-
Description designation Comments point of chemical content. Actually, the tool steels
High speed tool are made to more exacting standards and are in-
steels . M Molybdenum types
T Tungsten types
spected more closely than other grades. Therefore,
while carbon and alloy steels may resemble tool
Hot work tool steels. H1-H19 Chromium types
H20-H39 Tungsten types steels in composition, their quality in other respects
H40-H59 Molybdenum types may not be comparable. Because of the highly
Cold work tool steels D High carbon-high specialized nature of the implements made from
chromium types tool steels, this extra attention to quality is necessary.
A Medium alloy air
hardening types Maraging Steels. Table 3-9 lists the composition
o Oil hardening types
of three steels that do not bear AISI coding, but
Shock resisting tool are proprietary structural and tool steels. In con-
steels . s
Mold steels . P trast to all other tool steels, carbon is omitted from
Special purpose tool these steels. Maraging steels utilize instead the
steels . L Low alloy types soft, ductile, iron-nickel martensite, which can be
Water hardening tool
steels . W age hardened by other alloy additions to the steel.
These steels are basically martensitic in the solu-
tively small electric furnaces and are produced to tion annealed condition and in this condition are
meet special requirements for application in cer- easily machined. Hardening occurs by aging this
tain tools used by hand; for tools and dies used in martensitic structure in a temperature range of 840
mechanical fixtures for cutting, shaping, forming to 930 F (450 to 500 C).
and blanking of materials, either hot or cold; and
for precision gages. The description is not neces-
sarily intended to include that type of tonnage Summary
production open hearth steel used in the manufac-
ture of ordinary mechanic's hand tools, nor steel Austenite is a solid solution of one or more ele-
used in the manufacture of hammers, picks, files, ments in face-centered iron. Upon cooling molten
hollow drill steel, mining bits, cutters, and large pure iron it will solidify at 2800 F (1538 C) form-
rolling mill rolls. ing a-Fe or bcc ferrite. On further cooling to 2541
While identification of tool steels by codIng F (1394 C), the d-Fe transforms to y-Fe or fcc
system is practiced, the chemical content of indi- austenite. The iron remains austenitic until 1674 F
vidual grades of tool steels is usually not as rigidly (912 C) at which temperature it changes to a-Fe
fixed as in the carbon and alloy steels. To a·large or bcc ferrite. Although the a-Fe and a-Fe are
extent, this is because the tool steels are highly structurally the same, the two names are used to
specialized and designed for very definite pur- help identify the transformation temperatures.
poses. Although the various steels sold under trade
When carbon (or other alloying elements) is
names for a special use broadly resemble each
added to pure iron, the transformations no longer
other in chemical content, there are usually minor
occur at one temperature (except at eutectoids) but
differences in composition.
over a temperature range. The breadth of this
The principal types of tool steels have, however, temperature range depends on carbon or alloy
been classified according to their ultimate intended content.
Table 3-9. Compositions of .Maraging Steels
Identifying elements, per cent
Type C Mn Si Al Ti Mo Co Ni
18 Ni (200) ..... 0.03 max. 0.10 max. 0.12 max. 0.10 0.30 3.25 8.50 18.00
18 Ni (250) ..... 0.03 max. 0.10 max. 0.12 max. 0.10 0.50 4.85 7.75 18.00
18 Ni (300) ..... 0.03 max. 0.10 max. 0.12 max. 0.10 0.10 5.00 9.00 18.00
28
The rate at which a metal is cooled, particularly
at critical temperatures, has a marked effect on the
products of austenite transformation. Composi-
tion of the metal also has an effect. These effects
are demonstrated on isothermal transformation
and continuous cooling diagrams.
Austenite may be retained in a microstructure as
a result of carbon and/or alloy content, quenching
temperature and austenitizing temperature. Since
austenite is a relatively soft phase, the presence of
retained austenite will lower the overall hardness
of a steel. Retained austenite also may cause di-
mensional instability in heat treated parts. Marten-
site that transformed from retained austenite over
a long period of time is very hard and brittle and
may cause spontaneous cracking.
Most alloying elements are added to increase the
depth to which a piece of steel will harden. Some
elements are added to control grain size, others to
improve machinability, and others to enhance
electrical and magnetic properties.
Steels have been classified into carbon steels,
alloy steels, stainless steels and tool steels. The two
latter classifications are special-purpose alloy
steels. The American Iron and Steel Institute (AISI)
has devised a numbering system as a means of
identifying the different types and grades of steels.
 MATERIALS ENGINEERING INSTITUTE
C10L3
C11L12
C41L4
Test Series F Austenite and Its Transformation and Classification of Steels Page 1

PLEASE PRINT YOUR NAME AND ADDRESS BELOW RETURN TO: ASM Intemational®
MATERIALS PARK, OH 44073-0002
NAME REGISTRATION NO. _

STREET _

CITY STATE ZlP DATE _" _

MUL TIPLE CHOICE

Place the appropriate letter (A, B, C, D) in the box. (Only one answer is correct
and each question has a value of10 points.)

1. In cooling of pure iron, the face-centered cubic y-Fe (austenite) reverts to body-
centered cubic a-Fe (ferrite) at a temperature of: D
(A) 912°C (1674°F).
(B) 1394°C (2541 oF).
(~) 723°C (1335°F).
(0) 876°C (1610°F).

In AISI steel designations, the higher manganese (greater than 1.00%) carbon steels
2.
are indicated by the digits: D
(A) lOXX.
(B) 4OXX.
(C) l1XX.
(D) 15XX.

3. In addition to alloy content, a major cause of retained austenite is:

(A) rate of heating to the austenitizing temperature.

D
(B) increased austenitizing temperature for hypereutectoid steels.
(C) tempering too soon after quenching.
(D) condition of the quenching medium.

4. In addition to the steel composition, the products that form upon transformation of
austenite depend largely on the: D
(A) grain size.
(B) time of holding above 723°C (1335°F).
(C) cooling rate.
(D) holding time between Ms and Mr
 MATERIALS ENGINEERING INSTITUTE
C10L3
C11L12
C41L4
Test Series F Austenite and Its Transformation and Classification of Steels Page 2

5. The continuous cooling curve for AISI 4024 shows that transformation to 10%
ferrite occurs at the: D
(A) 6.35 mm (0.25 in.) position in 5 seconds.
(B) 4.7 mm (0.188 in.) position in 2 seconds.
(C) 1.588 mm (0.0625 in.) position in 10 seconds.
(D) ~.525 mm (0.375 in.) position in 11 seconds.

6. For 0.77% carbon steel, when austenite transformation is allowed to take place just
above about 540°C (1000°F), the product of transformation is: D
(A) bainite.
(B) retained austenite.
(C) lamellar pearlite.
(D) martensite.

The primary objection to the presence of retained austenite in a steel is that it:
7.

(A) lowers hardness.

D
(B) embrittles the structure.
(C) lowers resistance to corrosion.
(0) changes the magnetic properties.

8. . If a steel is cooled rapidly enough to suppress transformation at the nose ofthe TIT
curve, and if rapid cooling is continued to near room temperature, the product of
transfonnation is:
D
(A) lamellar pearlite.
(B) retained austenite.
(C) upper bainite.
(D) martensite.

Progressive transformation occurring over a period of time at constant temperature

9.
is known as: D
(A) continuous cooling.
(B) isothermal transfonnation.
(C) delayed transformation.
(0) temperature eqUalization.
 MATERIALS ENGINEERING INSTITUTE
vlOL3
C11L12
C41L4
Test Series F Austenite and Its Transformation and Classification of Steels Page 3

10. One means ofreducing the amount ofretained austenite in the heat-tt:eated product
is by: D
(A) using a higher austenitizing temperature.
(B) subjecting the quenched product to sub-zero temperature.
(C) increasing time at austenitizing temperature.
(D) decreasing quenching severity.

11. Most steel microstructures result from transformation of austenite by a diffusion

process that requires considerable time. The exception to this is the transformation
to:
D
(A) bainite.
(B) martensite.
(C) spheroidite.
(D) pearlite.

12. A diagram for a specific steel analysis which plots the microstructural constituents
of transformation of austenite at various cooling rates along an end quenched bar: D
(A) can only be related to water quenched steels.
(B) is known as a continuous cooling transformation diagram.
(C) is known as an isothermal transformation diagram.
(D) is known as a constitution or phase diagram.

13. If a plain carbon steel contained 0.50% carbon and its Ms temperature were 370°C
(700°F), its M f temperature would be approximately: D
(A) 400°C (750°F).
(B) 95°C (200°F).
(C) 200°C (390°F).
(D) 205°C (400°F).

14. Figure 3-15 indicates that the effect of increasing alloy content on an isothermal
transformation diagram for hypoeutectoid steel is to: D
(A) eliminate the proeutectoid ferrite field.
(B) raise the M s and M f temperatures.
(C) shift the curve to the left
(D) shift the curve to the right.
 Reviewers *
Kalinath Mukherjee
Department of Metallurgy, Mechanics and Materials Science
University of Dlinois

c. Marvin Wayman
Department of Metallurgy and Mining Engineering
University of Dlinois

* Affiliations listed were in effect at time of publication.

 ASM International® is a Society whose
mission is to gather, process, and
disseminate technical information.
ASM fosters the understanding and
application of engineered materials and
The Materials their research, design, reliable manu-
Information Society
facture, use, and economic and social
benefits. This is accomplished via a
unique global infonnation-sharing
network of interaction among members
in forums and meetings, education
programs, and through publications
and electronic media.

Printed on DocuTech November 1993

 The Materills
Infonnation Society

Annealing and
Normalizing Steel

Course 10
Lesson, Test 4
Course 41
Lesson, Test 5
 Metric Conversion Factors
To convert :from To Multiply by

in. mm 25.4
in. m 25.4 x 10-3
mil J.UD 25.4
JJin. J.UD 25.4
in.z
. )
rrr 6.45 x 1Q-4
m. m3 1.64 x 10-s
ft. m 3.048 x 10-1
ft.z nil 9.29 X 10-2
ft.3 m3 2.831 x 10-2
oz. g 2.834 x 101
lb. kg 4.536 x 10-1
Btu J 1.054 x 1())
BtuJIb.• OF JJkg. K 4.18 x ltY
BtuIft. - hr • OF W/m-K 1.730
10inJin.?F 1()"6 mmlrnrnJOC 1.8
psi Pa 6.895 x 1())
psi kPa 6.895
ksi kPa 6.895 x H)3
ksi MFa 6.895
ksi - in.l12 MPa-ml12 1.099
ksi...Jin. MPa-~m 1.099
oz.f gf 28.4
lbf kgf 4.536 x 10-1
lbf N 4.448
lbf- ft. N-m (orJ) 1.356
lbfJin.2 kgf/cm'2 14.223
Ibf/in. 3 kgflm3 2.768 x 10'
Ibjft.3 kglm3 16.019
IbJin. 3 g/cm3 2.768 x 1Q1
Iblin.3 kglm3 2.768 x 10'
gal (U.S. liquid) L 3.785
gal (U.S. liquid) m3 3.785 x 10-3
IbJgal gIL 119.826
fL/gal mIlL 748
OF °C (OF - 32)/1.8
OF K (OF + 459.67)/1.8
°c OF eC -1.8)+ 32
°c K °C + 273.15
K °c °C -273.15

Multiple and Submultiple Units

1018 •••••••••••••••••••••••••••• exa 10-1 _ ••••••••••••••••••••••••• deci
lOIS •••••••••••••••.•••••••••.• peta 10-: _..•••....•.....•........ centi
1012 ••••••••••••••••••••••••••• tela 10-3 •••••••••••••••••••••••••• D'lilli
1(fJ. ••••••••••••••••••_•••••••• giga 1()-6 •••••••••••••••••••••••• lI1icro
1()6 •••••••••••••••••••••••••• mega 10"'9 ..•.•.•••..••.••.•••.•••.• naDO
1()3 •••••••••••••••_ kilo 10-12 •••••••••••••••••••••••••• pico
1cr hecto 10-1s ••••••••••••••••••••••• femto
1Q1 ••••••••••••••••••••••••••• deka 10-13 ••••••••••••••••••••••••••• ano

Abbreviations
J .•....................... joule m meter Pa pascal
kg! kilogram force mm ........•... millimeter K kelvin
L _ liter N newton W •................•...... watt.

Technical Editor
Howard E. Boyer
Consultant

Editor
Philip D. Harvey
Metals Engineering Institute, American Society for Metals

e 1S79 American Society for Metals

This material may not be reproduced in whole or in part in any form whatsoever. Fully protected by copyright.
 Annealing And Normalizing Steel
Annealing
Annealing is used to reduce the hardness, im-
prove machinability, facilitate cold working,
produce a desired microstructur"e, or to obtain
desired mechanical, physical or other properties of
a metal. When applied to ferrous alloys, the term
annealing without qualification implies full an-
nealing. Full annealing is defined as annealing a
ferrous alloy by austenitizing and then cooling
slowly through the transformation range.
When other than full annealing is meant, the
following specific terms should be used to identify
the method of annealing or the condition of the
material after treatment: black annealing, blue an-
nealing, box annealing, bright annealing, flame
annealing, intermediate annealing, isothermal an-
nealing, process annealing, recrystallization an-
nealing, subcritical annealing, critical range an-
nealing, soft annealing, finish annealing, and
spheroidizing. Descriptions of these terms are
given in the following paragraphs.
Black or pot annealing is a term applied to an-
nealing of sheet, strip or wire in boxes or pots. A
black oxide coating is produced on the surface.
Blue Annealing. In this process, hot rolled
ferrous sheet is heated in an open furnace to a tem-
perature within the transformation range and then
cooled in air. The formation of a bluish oxide on
the surface is incidental.
Box annealing refers to heating steel in a sealed
container under conditions that minimize oxida-
tion. In this process, the charge is heated slowly
to a temperature usually below the transformation
range, but sometimes above or within it, after
which it is cooled slowly. This process is some-
times called close annealing.
Bright annealing is done in a protective atmo-
sphere which results in freedom from surface ox-
idation and discoloration.
Flame annealing is a process in which the heat is
applied directly by a flame. It usually is employed
for localized annealing.
Intermediate annealing is performed at one or
more stages of manufacture and may be
synonymous with process annealing.
Process annealing is performed by heating the
steel to a temperature below, but usually close to
the lower limit of the transformation range and
cooling slowly. The purpose being to soften the
metal for further cold working. Process annealing
and sub critical annealing are often synonymous.
Recrystallization annealing is used to produce a
new grain structure in cold-worked st~el without a
phase change. It is often referred to as process an-
nealing or subcritical annealing.
 2
Subcritical annealing consists of annealing cold-
worked steel to a temperature below the trans-
formation range and coolingby a convenient means.
This method can be the same as recrystallization
or process annealing.
Critical range annealing consists of heating the
metal to a temperature between the lower and up-
per critical temperatures (AI and A 3 ) followed by
cooling in a manner compatible with the material
composition. This process is discussed in more
detail later in the lesson.
Spberoidizing consists of controlled heating and
cooling to produce a spheroid or globular form of
carbide in the steel. See discussion later in the
lesson.
Finish annealing is a low-temperature (about
950 F or 510 C) heat treatment that is applied to
cold-worked steel having a low- to medium-carbon
content.
Isothermal annealing is a process in which the
steel is heated above the transformation range,
then cooling to and holding at a temperature at
which austenite transforms isothermally to a
.relatively soft ferrite-carbide aggregate.
mechanical properties are altered, one or more of
the other factors listed above may be affected. In
practically all cases, the alteration of any of the
properties is accomplished by an alteration of the
microstructure. It will therefore be appropriate for
the student to become familiar with the general
procedures of altering and controlling the micro-
structure.
Annealing Procedures
A desired microstructure may be obtained by
using one or more of the available annealing
pro.cedures. The choice of a specific procedure
depends on many factors, such as type of an-
nealing equipment available, weight and shape of
the workpiece to be annealed, accuracy of the tem-
perature controls, time available for the process
and end quality required.
The information in this lesson deals mainly with
the most widely used procedures or processes,
which are subcritical annealing, critical range an-
nealing, full annealing, isothermal annealing and
spheroidizing.

The above terms are all used in various segments
of the metal processing iridustry; however, some
overlap to a great extent in definition.
Purpose of Annealing
Annealing operations are performed to achieve
several purposes. In some instances, annealing is
done to achieve one purpose only, but two or more
of the following objectives may be achieved at one
time.
1. Induce softness.
2. Produce a definite microstructure.
Subcritical Annealing
Subcritical, or process annealing is the simplest
of the annealing processes. In this process, the
steel is heated to a temperature below the lower
heating critical temperature (Acl); after holding at
this temperature long enough to produce the
desired effect (time at temperature is usually
relatively short), the steel is cooled by any suitable
and convenient means. Table 4-1 illustrates the ef-
fect of time and temperature on hardness of a
medium-carbon low-alloy steel. Since no austenite
is formed, the manner of cooling is important only

3. Alter the mechanical, electrical or magnetic Table 4-1. Effect of Time,and Temperature on
properties. Brinell Hardness Subcritical Annealing (a)
4. Remove gases from the steel. Subcritical Brinell hardness resulting from
annealing holding for indicated time at
5. Remove stress. temperature annealing temperature
6. Produce grain reorientation. F C 2 Hrs 4 Hrs 6 Hrs 8 Hrs

7. Homogenize the microstructure. 1400 760 212 207 202

1350 730 229 217 212
8. Refine the crystal structure. 1300 705 241 223 217
The achievement of anyone of these objectives 1250 675 241 241 241
is no isolated accomplishment. For instance, if the a. Nitriding :135 mod. steel; A CJ = 1450 F 790 C.

A3 - - - - - - - - - - - - - - - -
Time---""
Fig.4-1. Schematic diagram showing method" of sub-
critical annealing.
in its effect on the internal stress condition of the
piece, rapid cooling producing relatively high
stress and slow cooling relatively low stress. If the
stress condition permits, any rate of cooling from
slow controlled to very rapid quench cooling may
be employed. Although no austenite is formed,
diffusion of some type may take place, gases can
be eliminated, softness induced, stress lowered,
cementite be made to spheroidize, the magnetic
and electrical properties altered, 'and grains
reoriented. No microstructural changes such as
result from the transformation of austenite can oc-
cur. The only possible change in microstructure is
the spheroidization of the carbides and the change
in size or shape of the ferrite grains.
For subcritical annealing, neither isothermal
transformation diagrams or continuous cooling
diagrams are usable because there is no trans-
formation. Further, the critical 'temperatures for
cooling <An and Arl) are not needed. The only
significant reference temperature is the lower
heating critical (Acl) so that the annealing tem-
perature can be kept below wher~ auste~ite begins
to form. Figure 4-1 shows schematically how a
subcritical anneal is performed with respect to the
lower heating critical.
In subcritical annealing, a degree of caution
must be exercised in selecting the annealing tem-
perature. In determining the location of Acl' it
should be borne in mind that the rate of heating
bas a defmite effect. If rapid rates of heating are
3
employed, the Ac1 will be higher than with slow
heating rates. Therefore, the rate of heating that
would be used in actual conditions of anneal
should be considered in determining the Ac1- If
the Acl has been incorrectly determined and the
real Ac.l for actual existing conditions is exceeded,
some austenite will form. This is particularly true
of certain manganese-nickel steels which can form
remarkably large amounts of austenite at relatively
low heating temperatures. The longer the holding
time at a specific temperature, the more austenite
will form up to some maximum amount for that
particular temperature. The austenite thus formed
mayor may not transform upon subsequent
cooling from the annealing temperature so that the
final microstructure may contain unwanted
retained austenite, untempered martensite, or
both. The presence of these constituents is
generally considered undesirable in subcritically
an:p.ealed steel and may raise the tensile strength,
increase hardness, lower the ductile properties and
lower the notch toughness.
When subcritical (process) annealing large
loads, it may be expedient to use a furnace tem-
perature' that is somewhat higher than the AC1 tem-
perature of the steel being annealed to speed up
the process. Such a method, however, demands
close control to prevent any of the charge becom-
ing austenitized. Refering to Figure 4-2, although
the maximum furnace temperature was above
815°C (1500°F), the top coil in the load did not
reach AC1 (approximately 727°C or 1341°F).
Critical Range Annealing
In this process, the steel is heated to a temper-
ature between the lower and upper heating criticals
<Acl and Ac3). Referring to Fig. 4-1 the holding
temperature for critical range annealing would be
just above the shaded area instead of within the
shaded area as is shown for subcritical annealing.
Since the annealing temperature is within the
critical range, some austenite is formed, the
amount depending on the temperature reached.
The method and rate of cooling from the an-
nealing temperature are therefore important, be-
cause the austenite may be transformed not only to
-soft lamellar pearlite but also to the hard micro-
structures of bainite and martensite. Therefore,
the austenite must be cooled in a fashion that will
 4

J 400 r--+-+---+---+--~~--t-----:::~--+---4-----t--~--I------+-
Stacked hei g ht, 144 in.
Weight, 217 tons
I

U-.IOOO
a>
~-

~ 800
Q)
~
E
~ 600 I
i
j Temperature o~
400 f----t---+----.;.---.;.--,' bottom coi J :'""----~--
I 1 I

I
2 00 It'"------!---+---t--+---:- _-+1_---:.1_--+-1-~--+---4--~---+-_..J..-------:-~
I I
o ~..___._......I _......I--_I..--~I_...:.----:.._.-..:.I_--J.-_.L...-~_~____L.
!
_ _J.__..:.._____J

o 10 20 30 40 50 60 70 80 90 100 110 i20 130 140 150 160

Time) hr

Fig. 4-2. Heating and cooling cyclejorprocess annealing a 217-ton load ojcoiled low-carbon steel.

result in a properly annealed microstructure. It-is
essential that the heating and cooling critical tem-
peratures be known so that the cooling procedure
can be properly controlled through the transfor-
mation range. The critical temperatures should be
determined by using rates of heating and cooling
similar to those employed in the actual annealing
operation in order to avoid incorrect evaluation of
these points.
Slow continuous rates of cooling are most
generally used in critical range annealing to insure
proper transformation of any austenite that may
have been formed. The austenite may also be trans-
formed isothermally. The Isothermal Transfor-
mation Diagram may be useful in this type of
annealing depending largely on the composition of
the steel being annealed. Continuous Cooling
: Diagrams have a very limited use here and need
not be considered.
Heating into the critical range has a pronounced
spheroidizing effect on existing cementite plates,
and this type of anneal may be used for some steels
to produce a spheroidized microstructure with
resultant lowering of hardness and mechanical
properties. Other effects of critical range an-
nealing are alteration of electrical and magnetic
properties, the removal of stress, grain reorienta-
tions and a limited degree of grain growth. This
method of annealing is most often used for low-
carbon steels under which conditions very little
austenite is formed.
Full Annealing
In this process the steel is heated to a tem-
perature at least 50 F (28 C) above the upper
heating critical. At this temperature, the steel is
completely austenitic and the m~thod and rate of
cooling the austenite are very important for proper
microstructure and other related properties. In full
annealing, it is essential to know one or more of
the following: (a) the heating and cooling criti-
cal points, (b) the Isothermal Transformation Dia-
gram, and (c) the Continuous Cooling Diagram. In
any case, the upper heating critical temperature
must be known so that the proper annealing tem-
perature can be selected. After the steel has been
austenitized, it must be cooled in a planned con-
trolled manner. Cooling to secure a given trans-
formation product can be done in several ways,
which will be considered from the viewpoint of
available information on transformation.
If the 'cooling criticals are available, it is known
that when the r\3 temperature is reached, the
austenite will begin to transform and transfor-
mation will progress with falling temperature
until the lower cooling critical temperature is
reached where transformation is completed. When
planning an annealing cycle around the critical

temperatures, remember that, as far as trans-
formation is concerned, the cooling practice in
the cooling critical temperature range (Ar~ to Ar :)
is the important part of the anneal. The rate with
which the steel passes through this range will deter-
m:ne the microstructure, hardness and other
properties of the transformed product. A very
slow rate of cooling will result in the softest
microstructure; which is spheroidite. Faster rates
result in lamellar pearlite of varying degrees of
coarseness and hardness. If the rate of cooling is
too rapid, formation of the soft products of trans-
formation will be suppressed and the harder con-
stituents, bainite and martensite, formed. These
latter microstructures are undesirable in the an-
nealed structure.
The rate at which the steel cools from the an-
nealing temperature to the upper cooling critical
temperature is unimportant as long as the proper
cooling rate is observed passing through the
cooling critical range. Because of the difficulty of
maintaining uniform temperature in a typical com-
mercial annealing load, common practice is to cool
from the annealing temperature, to and through
the transformation range at a single uniform rate.
In this manner, the danger of a faulty anneal from
nonuniformity of temperature and transformation
in the load is avoided.
In like fashion, the rate of cooling after trans-
formation is unimportant because it has no effect
on the microstructure and hardness. However, it is
common practice to maintain the uniform rate of
cooling well past the lower cooling critical to en-
sure a proper anneal in case the critical tem-
peratures were incorrectly located or there was a
temperature differential in the load. Figure 4-3
illustrates schematically how the cooling criticals
are used to design an annealing cycle.
The use of the cooling critical temperatures to
design ar. annealing cycle has the disadvantage
that the cooling rate necessary to secure a given
microstructure must be determined by, trial and
error, especially if fine gradations of hardness are
desired. Furthermore, as the rates of cooling are
altered, the locations of the cooling criticals are
also altered so that the heat treater must have con-
siderable experience and judgment to secure the
desired results.
If an Isothermal Transformation Diagram is
available, it can be very useful in securing a desired
·5
~~~nlrolled Slow
~Oling Rate
~ P---~ " ' - - - - - - A r3
e~II
~
E .......--~\------ . - - - - - - - - . .
C1>
r-
Tlme- Time-
a b
Fig. 4-3. Use of cooling critical temperatures for deter-
mining annealing cycle. fa) Slow cooling through the
critical range with arbitrary cooling both above and
below the transformation range. (b) Controlled con-
tinuous coofing from the annealing temperature,
through the transformation range, and below the trans-
formation range.
transformation product in full annealing. These
curves show the microstructure to which the
austenite will transform, the temperature at which
this transformation will take place, and the time
required; thus, their use makes possible a very
close control of the annealed product. It is only
necessary to cool the steel to the temperature
where the desired microstructure will form, hold
until transformation is complete, and then cool in
any convenient manner. The only precaution
required is to cool the steel to the desired trans-
formation temperature at a rate that will avoid
undercooling. This means careful planning of the
size, weight and distribution of the annealing load
as well as accurate temperature control of the fur-
nace. Figure 4..4 illustrates the use of the isother-
mal transformation diagram in full annealing. By
altering the transformation temperature a variety
of microstructures and hardnesses can be secured.
Continuous Cooling Diagrams may also be
useful in ·full annealing because the point where
transformation begins, where it ends and the
products formed with various rates of cool, have
already been worked out. It is only necessary to
select that rate of continuous cool which will result
in a given product. However, as a gener~l rule, the
Continuous Cooling Diagrams are more useful in
quenching than in annealing of steel. This is not to
imply that these diagrams are not useful in an-
nealing, because the data on the transformation of Heating cycles employing austenitiziIig tem-
austenite on continuous cooling is perfectly valid peratures in the upper part of the ranges given in
once it has been developed. Table 4-3 produce pearlitic structures. Predom-
inantly spheroidized structures are obtained when.
Annealing Temperatures. For many purposes of
the lower austenitizing temperatures are used.
annealing, simply specifying that the steel be
cooled in the furnace from a designated annealing
(austenitizing) temperature is sufficient. Tem- Isothermal Annealing
peratures and associated Brinell hardnesses for
full annealing of carbon steels are given in Table 4- The practice of annealing so that the trans-
2 and similar data for alloy steels in Table 4-3. formation occurs at constant temperature is re-
ferred to as isothermal annealing, as contrasted
Table 4·2. Recommended Temperatures and to continuous cooling annealing where trans-
Cooling Cycles for Full Annealing Small Carbon formation takes place over a range of temper-
Steel Forgings (a) ature. Both types are capable of producing a
full anneal and each has advantages and a field of
,-- Cooling ~
Annealing cycle(b) Brlnell usefulness. Regardless of the method of trans-
temperature, From, To, hardness
Steel F F F range forming the austenite, the purposes of full an-
1018 1575 to 1650 1575 1300 111 to 149 nealing are the same and include one or more of
1020 1575 to 1650 1575 1290 111 to 149 the following: induce softness, produce a definite
1022 1575 to 1650 1575 1290 111 to 149
1025 1575 to 1650 1575 1290 III to 187 microstructure, alter the mechanical properties,
1030 1550 to 1625 1550 1200 126 to 197 remove stress, produce grain reorientation,
1035 1550 to 1625 1550 1200 137 to 207 homogenize the microstructure and refine the
1040 1450 to 1600 1450 1200 137 to 207
1045 1450 to 1600 1450 1200 156 to 217 crystal structure.
1050 1450 to 1600 1450 1200 156 to 217
1060 1450 to 1550 1450 1200 156 to 217
1070 .•. 1450 to 1550 1450 1200 167 to 229 Isothermal annealing is most commonly applied
1080 1450 to 1550 1450 1200 167 to 229 to plain and alloy high-carbon steels, because with
1090 1450 to 1525 1450 1200 167 to 229
1095 1450 to 1525 1450 1215 167 to 229 these steels the greatest benefits of economy in fur-
(a) For forgings up to 3 in. in section thick- nace time and ability to obtain the desired
ness. Time at temperature 15 usually a mini- microstructure are obtained. The rules to be
mum of 1 hr for sections up to 1 in. thick;
;2 hr is added for each additional inch of followed in establishing an isothermal annealing
thickness. (b) Furnace cooliDg a.t 50 OF per hr. cycle may be briefly stated as follows:

Table 4-3. Recommended Annealing Temperatures of Alloy Steels (for Furnace Cooling)
Annea.ling Bhn. Annealing Bhn, Annealing Bhn,
Steel temperature. F ma.x Steel temperature, F max Steel temperature, F max

1330 ...... 1550 to 1650 179 50B40. . . .. 1500 to 1600 187 6150 . . . . •. 1550 to 1650 201
1335 ...... 1550 to 1650 187 50B44 ..... 1500 to 1600 197 81B45 ..•.. 1550 to 1650 192
1340 ...... 1~0 to 1650 192 5046 ...... 1500 to 1600 192 174
50B46. . . .. 1500 to 1600 192 8627 ...•.. 1500 to 1600
1345 ...... 1550 to 1650 8630 ...... 1450 to 1550 179
1500 to 1600 187 50B50 ..... 1500 to 1600 201 192
3140 ...... 50B60 . . . . . 1500 to 1600 217 8637 .. ~ ••. 1500 to 1600
4037 ...... 1500 to 1575 183 8640 ....•. 1500 to 1600 197
4042 ...... 1500 to 1575 192 5130 ...... 1450 to 1550 170 8642 ...... 1500 to 1600 201
4047 ...... 1450 to 1550 201 5132 ...... 1450 to 1550 170 8645 ...•.. 1500 to 1600 207
4063 ...... 1450 to 1550 223 5135 . . . . .. 1500 to 1600 174 86B45 ..... 1500 to 1600 207
5140 ...... 1500 to 1600 187 8650 ...... 1500 to 1600 212
4130 ...... 1450 to 1550 174 223
5145 ...... 1500 to 1600 197 8655 ...... 1500 to 1600
4135 ....•. 1450 to 1550 5147 ...... 1500 to 1600 197 8660 •..... 1500 to 1600 229
4137 ...... 1450 to 1550 192
4140 ...... 1450 to 1550 197 5150 ...... 1500 to 1600 201 8740 ...•.. 1500 to 1600 202
4145 ...... 1450 to 1550 207 5155 ...... 1500 to 1600 217 8742 ...•.. 1500 to 1600
5160 ...... 1500 to 1600 223
4147 ....•. 1450 to 1550 51B60 ..... 1500 to 1600 223 9260 ...••• 1500 to 1600 229
4150 ...... 1450 to 1550 212 174
94B30 ...•. 1450 to 1550
4161 ....•. 1450 to 1550 50100 ..... 1350 to 1450 197 94B40 ..••. 1450 to 1550 192
4337 ....•. 1450 to 1550 51100 ••... 1350 to 1450 197
223 52100 ••••. 1350 to 1450 207 9840 •..••. 1450 to 1550 207
4340 ....•. 1450 to 1550
 7
1. Higher austenitizing temperatures promote Sometimes two or more transformation· tem-
the formation of a lamellar microstructure; peratures are incorporated in the cycle. For exam-
lower austenitizing temperatures promote a ple, an isothermal treatment consisting of
spheroidal structure. austenitizing at 1425 F (775 C) and transforming at
2. The softest structure is obtained by using 1340 F (725 C) for 4 hr and at 1300 F (70S C) for 1
minimum austenitizing temperatures and hr is often used. This treatment not only can
maximum transformation temperatures. produce a microstructure superior to that obtained
by the conventional procedure of cooling at 10 F
3. Furnace time is saved by cooling rapidly (5 C) per hr from the austenitizing temperature
from the austenitizing ,to the transformation through the transformation range, but it can
temperature, and cooling rapidly (as by produce this structure in less than half the total
removing from the furnace) after the steel has time required for conventional annealing.
completely transformed.
The temperature control in isothermal annealing
Parts such as bearing races made from the high- must, however, be more accurate. This, in turn, in-
carbon, high-chromium steels are frequently an- dicates a need for better equipment and closer
nealed isothermally. For these steels, the ideal supervision; two conditions that often limit the
microstructure for machinability and for the hard- use of isothermal annealing in many shops.
ening treatment treatment is 100070 spheroidized.
The continuous furnace is ideal for isothermal
Cycles for isothermal annealing may vary over a annealing large quantities of parts of the same
broad range. The principal of isothermal an- grade of steel. The continuous furnace must be
nealing is illustrated schematically in Fig. 4-4. designed with several individual zones so that the

•....

·:::t~~~It·~":'"""~t-:::.--------/------------~=:----A I --

·\:!!t~~tt.. ;Isothermal Transformation

I Co 0 ling..::::~;:~:~:~:~:::~:~~:~:~:;:~:~:}~:. ::r::~:::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::;:;::::::. ::::::::::::::::::::::::::::::::::::::::::t~~~~!t~\ .

Transformation
Temperature
·:::t~t:.
·::tt:::..

/::~~t~!~~!tt~:: .
Cooling After ·:::~t:~j~jttk
M,--------~ Transformation . .:.: : : : : : :
Is Complete
Mf ------------

Time •
Fig. 4-4. Use of isothermal transformation diagram for annealing to lamellar pearlite. Time is counted only after the
upper cooling critical temperature has been passed.
 8
work as it passes through will be consecutively
brought to temperature, held at temperature, and
cooled at the desired rate. The speed through the
furnace and proper zoning must be coordinated to
obtain the desired cycle. The more complex alloys
and higher-carbon steels that generally require a
long period of controlled slow cooling may be
isothermally transformed successfully in con-
tinuous furnaces.
Annealing for Microstructure
In the majority of annealing procedures, the
primary purpose is to develop a specific
microstructure and the concomitant properties.
One of two major microstructures formed is
lamellar pearlite. Pearlite is desirable in many
machining operations such as forming, drilling
and broaching, especially in the medium carbon
steels (0.30 to 0.60070 C). This microstructure is
also desirable in cold drawing operations, partic-
ularly wire drawing of high carbon steel. Lamellar
annealing is almost without exception a full anneal
where the steel is heated above the Ac3. It is a
general rule that the more homogeneous the
austenite is during heating, the more completely
lamellar the microstructure will be after trans-
formation. Therefore, if a lamellar structure is
desired, the steel is heated well above the.Ac3 tem-
perature because formation of lamellar pearlite is
primarily a matter of the annealing temperature.
Temperatures at least 100 F (55 C) above the A.c3
are generally employed. In extreme cases, an ex-
cess of 200 to 400 P' (Ill to 222 C) is often neces-
sary to produce the desired lamellar pearlite.
Figure 4-5 shows the effect of the austenitizing
temperature upon the tendency to form pearlite.
While it is true that the annealing temperature is
the prime factor in produCing pearlite, the proper
cooling practice must also be used. When em-
ploying continuous cooling rates, the rate of
cooling through the transformation range must be
sufficiently slow to allow the pearlite to form, and
yet not so slow that the already formed cementite
particles will coalesce and form spheroids. The
transformation characteristics of steel vary widely
depending on the carbon and alloy content, thus
making it impossible to specify one general contin-
uous cooling rate that would result in the trans-
 formation to pearlite in all steels. As a means of
( comparison, however, for steels of medium carbon
and medium alloy content, a cooling rate of 30 to
40 F (i 7 to 23 C) per hour would result in a pearl-
itic microstructure, while a cooling rate of 5 to 20 F
(2.5 to 9 C) per hour would result in a spheroidal
microstructure. To secure a specific degree of fine-
ness of pearlite, the proper cooling rate must be
determined experimentally. Figure 4-6 illustrates
different degrees of coarseness of pearlite obtained
with variations in annealing practice.
Pearlite can be secured by means of isothermal
transformation. The degree of fineness can be con-
trolled closely by selecting the proper trans-
formation temperature, using an isothermal trans-
formation diagram as reference. Again, the car-
bon and alloy content of the steel will determine
the temperature at which a particular type of
pearlite forms and the time required for its for-
mation. In general, higher transformation tem-
peratures result in coarse, relatively soft pearlite
and lower transformation temperatures in fine,
relatively hard pearlite. In all instances, the nose
of the isothermal transformation diagram
represents the temperature where pearlite forms
most rapidly, this pearlite being medium in
microstructure and hardness. The important con-
siderations in transforming isothermally to
pearlite are: (a) cool the austenitized steel to the
desired isothermal temperature with sufficient
rapidity to avoid transformation to some other up-
per transformation product, and avoid un-
dercoating for the same reason; (b) after trans-
formation is completed at the isothermal tem-
perature, the steel should be cooled rapidly to
room temperature not only to conserve time but
also to prevent spheroidization of the previously
formed cementit.e lamellae.
Spheroidizing
A significant amount of steel is spheroidize an-
nealed to produce a microstructure consisting
essentially of spheroidized cementite particles in a
ferrite matrix. The spheroidized condition is the
true equilibrium state of .the steel and represents
the softest condition. The spheroidized micro-
structure is generally desirable for most machining
operations especially those involving high carbon
steels. The very high carbon and high alloy tool
steels are spheroidize annealed almost without
A
:'G'
a b
~,~~
- .-> .
._--~
~-.--. '-
......
-
~~:
c d
Fig. 4-6. Photomicrographs, at IOOOX, showing that
high annealing and trans/ormation temperatures favor a
coarse pearlite (a and b), while lower annealing and
transformation temperatures javQr a finer pearlite (c
and d), all samples were O. 73 % plain carbon steeL (5 %
Nitaletch)
exception, not only to provide good machinability,
but also to ensure good response to subsequent
hardening operations. The spheroidized micro-
structure also possesses good cold forming char-
acteristics and is generally specified for steels that
are to be formed cold such as by bending. Gener-
ally, the larger the spheroids and the more distance
between them, the greater the ability of the steel to
be cold formed.
A spheroidized microstructure can be developed
by numerous means, but in all instances, low an-
nealing temperatures are essential to the formation
of a spheroidized microstructure. The simplest
method of spheroidizing is to employ a subcritical
anneal. In this procedure, the steel is heated to a
temperature about 25 F (13 C) below the Acl and
held there for a prolonged period of time.' As a
result, the existing cementite particles coalesce, or
ball up, and form spheroids. For this type of
spheroidize annealing, a fine prior microstructure
such as martensite, bainite or fine pearlite is
desirable. Coarse cementite in the prior structure
should be avoided because the large cementite par-
ticles do not coalesce as readily as the finer ones.
Because of the comparatively long times required
to spheroidize at the subcritical temperature, this
procedure is used relatively seldom in commercial
practice.
 -10
A 3------------
Time-~
Hold
I
\(e)l Time • Time - -...
Fig. 4-7. Schematic diagrams showing various methods ofspheroidizing with relation to critical heating temperatures.
Holding time must be limited.
One frequently used method of spheroidize an-
nealing consists of heating to a temperature of 25
to $0 F (13 to 26 C) below Acl' holding at this tem-
perature, then increasing the temperature to a
point between Act and Ac3 and again holding.
Following the second holding period, the tem-
perature is decreased slowly. This procedure is
shown schematically in Fig. 4-7a.
A third frequently employed method of spher-
oidizing consists of heating to a temperature 10
to 50 F (5 to 26 C) below the Ac3' holding at this
temperature, and then increasing the temperature
to slightly above Ac3 followed by slow cooling, as
shown in Fig. 4-7b.
Another procedure that has been used ef-
fectively for obtaining spheroidized structures is
illustrated schematically in Fig. 4-7c. The steel is
first. heated to a temperature just below Ac3' held
Hold~
Slow Cool~·::·:;·:·::.,
;fUr.:.:.: . . .: :;:
Time--~p
for a time, cooled rapidly to below Act then heated
above Ac] and held for a time before slow cooling.
A slight modification of this procedure is shown in
Fig. 4-7d. Here, the steel is heated and cooled
alternately above and below Ac 1 after which it is
slow cooled.
In any of these procedures it is essential that
nuclei be present to ensure formation of spheroids.
The nuclei may be undissolved cementite, carbon
concentration gradients (inhomogeneous aus-
tenite) or, in some instances, nonmetallic inclu-
sions. Many devices are employed to ensure that
nuclei are present when the steel is spheroidize
annealed in or above the critical range. Common
practice is to preheat the steel at a temperature
about 25 F (13 C) below the Ac1 when annealing in
the critical range, and about 50 F (26 C) below the
Ac3 when full annealing.

...:::s
CD
e
CD
C.
E
Q)
I-
CI
Time
Fig. 4-8. Isothermal method ofspheroidizing. Note the long hold time after trans/ormation has been completed.
By preheating, some carbide from the prior
structure is spheroidized, and upon further heating
these spheroids resist being taken into solution and
are present to act as nuclei when the steel is cooled.
The reason the spheroids are more difficult to take
into solution, as compared to a long thin cementite
particle, is quite simple. As previously noted, when
austenite forms and dissolves cementite, it does so
by the process of diffusion. The larger the area ex-
posed to diffusion, the more rapidly the process
will occur, other conditions being equal. Thus,
when there are two cementite particles of equal
volume, one a sphere and the other a prism or cyl-
inder, the ratio of the surface areas to volume is
much greater in the prism or cylinder than in the
sphere. Thus, it is more difficult for the cementite
to diffuse into the austenite when it is in a
spheroidal form.
A second method of ensuring that nuclei are
present is to heat the steel to within or above the
critical range so that some austenite is formed,
then rapidly cool so that the austenite transforms
to some ime product such as martensite, bainite or
11
•
fine pearlite. The steel is then reheated as in a con-
ventional spheroidize annealing cycle where the
fine cementite particles formed during the rapid
cooling will readily spheroidize and serve as nuclei.
The various schemes described to aid nucleation
and spheroidization are used alone or in com-
bination and may be repeated cyclically to assure a
desired result (see Fig. 4-7). Whatever the process
used, the larger the spheroids formed and the
greater the distance between them, the so~ter will
be the annealed structure. This effect of larger
spheroids and greater distance is secured with
higher annealing temperatures, by 'repeated an-
nealing or annealing for very long periods of time.
At this point some caution must be used, for if ex-
cessively long annealing times are employed at
relatively high temperatures, a very coarse and ab-
normal agglomeration of the cementite particles
will result. This condition is extremely undesirable
from the standpoint of machinability. Fur-
thermore, cementite particles of this size are ex-
tremely difficult to dissolve in austenite and result
 12
in poor response in subsequent hardening
operations.
The spheroidizing treatments described thus far
consist of various heating and holding practices
followed by a slow continuous rate of cool.
Spheroidize annealing can also be done isother-
mally. In isothermal practice, the heating
procedure is the same as previously described
although the full type of anneal is generally used.
After proper heating, the steel is dropped in tem-
perature to slightly below the lower critical, as
shown in Fig. 4-8, and held at this temperature for
an extended period of time, after which it is cooled a ;i~~~!~
at any convenient rate.
Regardless of what procedure is employed to
produce a spheroidized microstructure, the micro-
structure of the steel prior to annealing has an
important bearing on whether spheroids will form.
To illustrate this point, consider a hypereutectoid
steel in the as-rolled condition. Such a steel is
generally slow cooled from the rolling temperature
and has a microstructure consisting of large grains
of pearlite. Surrounding these grains will be a solid
network of cementite called the carbide network
(see Fig. 4-9a). Within the grains, arranged along
certain crystallographic planes, will be solid spears
of cementite. When such a microstructure is sub-
jected to a spheroidize anneal, the fine pearlite
Fig. 4-9. Photomicrographs, at 700X, showing micro-
within the grains responds readily and spheroids structure of hot-rolled hypereutectoid steel. (a) Carbide
form. However, the coarse cementite in the car- network in the grain boundaries and (b) remnants of
bide network and spears resist spheroidization and network after improper annealing. (Nital etch)
are plainly visible among the spheroids in the final
annealed microstructure, as in Fig. 4-9b. This rem- steels. Regardless of the carbon centent, the pres-
nant of the carbide network is undesirable because ence of cementite plates, such as in coarse pearlite,
of the difficulty it causes in machining operations is undesirable because of their resistance to spher-
and generally because the cementite follows oidization. It is for this reason that eutectoid car-
definite planes, making the steel as a whole bon steel is often difficult to fully spheroidize
relatively brittle. anneal. In large sections of eutectoid carbon steel,
it is difficult to prevent the formation of coarse
Elimination of this condition may be achieved pearlite except with very rapid rates of cool. There-
by repeated or prolonged annealing, although such fore, it is generally recommended that such
procedure is not economica~ly desirable. A more material be quenched prior to spheroidize an-
satisfactory procedure is to normalize or quench nealing.
the steel before annealing. The rapid cooling rates
in these treatments re~~l~.t in a fine transformation
product which is homogeneous and free of carbide Miscellaneous - Purpose Annealing
network or spears. Such a microstructure is much There are several reasons for annealing other
more amenable to spheroidizing. than developing specific microstructures as
The importance of prior condition in spher- described in the foregoing. One involves alteration
oidize annealing is not confined to hypereutectoid of the properties with little or no regard to the

microstructure. Another procedure involves the
development of good machining properties in cer-
tain low and medium carbon, high-nickel steels,
where the excess ferrite is present either as a net-
work surrounding the grains or as bands.
Ferrite in this form is very soft and gummy and
causes difficulty in machining. To improve
machinability, the steel is heated to a temperature
well over the Ac3 and then cooled at a fairly rapid
rate. This results in a blocky appearing microstruc-
ture in which the ferrite does not cause undue
machining problems.
Removing Gas
When the steel is in the molten state, it is able to
dissolve gases, some in considerable quantities.
One such gas, hydrogen, has a solubility of about
30 cubic centimeters per 100 grams of molten steel.
When the molten steel solidifies, the solubility of
the hydrogen drops to about 9 cubic centimeters
per 100 grams in the austenitic state and about
3 cubic centimeters in the transformed state.
Therefore, when steel is cast into an ingot mold, its
ability to hold gas in solution decreases abruptly as
Fig. 4-10. Ground and etched cross section of a steel billet sho14'ing {)'pical cooling cracks.
13
the steel solidifies and cools. The gas that is
thrown out of solution can diffuse out of the steel
and be dissipated into the air under certain con-
ditions of time and temperature. If, however, the
conditions for proper diffusion are not met, the
hydrogen will collect in any small void or opening
which might be present and there build up a
tremendous pressure, sufficient to tear the steel
apart. This tearing is evidenced as small cracks,
variously called cooling cracks, snowflakes and
thermal cracks. Their presence is undesirable
because they can act as the starting point of fatigue
failures.
Furthermore, the presence of hydrogen, in
amounts insufficient to cause cooling cracks, can
cause cracking in the hardening operation. Be-
cause of the diffusion characteristics of hydrogen
in steel, the cracks are usually found at a location
roughly halfway between the outer surface and
the center, as in Fig. 4-10.
To avoid such imperfections, hydrogen may be
eliminated by annealing. The annealing procedure
is based on three facts: (a) the cracks do not occur
until the steel reaches a quite low temperature (200
 14
to 300 F or 95 to 1SO C; (b) the rate of diffusion of
hydrogen in austenite is only a tenth of that in the
products of transformation; and (c) the rate of dif-
fusion in the transformation products is at a
maximum just below the Ac! temperature. In
achieving hydrogen removal, the steel is kept
austenitic until ready for annealing. It is then
cooled to some temperature safely above 300 F
(150 C) so that all of the austenite can transform
and conditions of improved hydrogen diffusion
exist. After the steel is thus transformed, it is
reheated to a temperature about 50 to 100 F (26 to
56 C) below the Act and held there a calculated
length of time until the hydrogen content has been
lowered below some safe amount which cannot
result in cooling cracks. The steel is then cooled to
room temperature. Figure 4-11 presents an anneal-
ing cycle for elimination of hydrogen.
Stress Removal
When steel is cooled rapidly, or is severely cold
worked, a high magnitude of internal stress may
develop. This stress can be high enough to cause
the steel to crack or break spontaneously, or at
least to warp or distort. Highly stressed material is
usually annealed to lower the residual stress to
some safe value. For all practical purposes, it may
be said that the higher the annealing temperature
the more stress is eliminated, and at a temperature
slightly below the Ac1most of the stresses will be
eliminated. This process is not instantaneous, and
to achieve the maximum benefit some time at tem-
perature is required. The elimination of stress is
therefore a function of both time and temper-
ature.
Alteration of Grains
When steel is plastically deformed, it becomes
harder and stronger from the work hardening. Un-
der these conditions, the steel exhibits within its
grains a series of lines called slip lines which in-
dicate that some work hardening has taken place.
(See Fig. 4-12.) With increasing amounts of defor-
mation, the grains become distorted and elongated
in the direction of the cold work. The steel also
becomes harder, stronger, less ductile and ·highly
stressed. This condition mayor may ·not be
desirable, depending on the ultimate use for which
the steel is intended. If the condition is un-
desirable, as in the case where further cold work is
to be performed, the steel must be made soft and
ductile.
If steel that has been cold worked is heated in
stages, it will be observed that as the temperature
incr~ases the level of internal stress hardness and
tensile strength decreases. As these properties
decrease, the ductile properties, elongation and
reduction of area increase. Furthermore, as the
temperature increases, the elongated and distorted
cold worked grains will reform, as shown in Fig. 4-
13. If heating is continued beyond the temperature
at which the grains have again assumed their
original equiaxed form, a growth occurs with the
grains becoming fewer and larger as the tem-
perature rises. The method by which the grains
reform and grow is somewhat similar to that which
occurs when metals freeze. In cold worked steel,

AUSTENITE
1i
t t-~.-o:-\~t;----T-RAN-A-;-;i-i-;-I~-~O-N------A3
R:1rf.:rJR!t~t!~·=-··~:··-··:·~'~-~~-~-.fK--AI

TRANSFORMATION ",
~OOOCT ~
- - - - - - - - - - - - - - - - - - - - - - - - ~~.!
TIME---'

Fig. 4-11. Schematic diagram showing method of

eliminating hydrogen in steel.

c d
Fig. 4-13. .PhotomicrograpJ:s showin.g effect on microstructure of cold-worked steel (a) before cold working, (b) after
cold worklng~ (c) afterpartzal annealzng and (d) afterfurther annealing with resulting grain growth.
the nuclei are areas of high energy content located
in the grain boundaries and slip lines where the
deformation actually occurs.
The process just described is usually separated
into three stages; recovery, recrystallization and
grain growth (see Fig. 4-14). Recovery is that
period in which the highly stressed condition of
the cold worked steel is partially eliminated.
Recrystallization is that period in which the
original mechanical properties of the steel are
mainly restored and a new crystal structure
~imilar to that which existed before cold working:
IS formed. The temperature at which the deformed
crystals reform may be well below the Acl (as low
as 860 F or 460 C), as in the case of a very severely
cold worked steel, or above the Ac3' as in the case
of a steel that has received no cold deformation.
Grain growth is an advanced stage of
recrystallization in which the grains grow to a size
in excess of that in the steel prior to cold working.
The processes described are important com-
15
~
• ."
•
., /...."..
" . . ...
,.~ . . ·0··
~
:
• ""
....
':. I·
mercially because through them the effects of cold
working can be eliminated and the steel thus con-
ditioned for additional cold deformation. The
grain size of the steel may also be controlled, a fac-
tor of great importance in many steel processing
operations involving products such as cold formed
wire and rods, cold drawn and formed tubular
products such as pipe, and cold rolled and formed
sheet.
Another factor of importance in annealing cold
worked steel is that of critical strain. If the steel
has received a cold reduction of between 7070 and
15070, it will upon subsequent annealing form very
large grains greatly exceeding those which would
form in steels cold reduced to below 70/0 or above
15070 and annealed in identical fashion. Since these
abnormally large grains tend to make the steel brit-
tle, their presence is usually undesirable where
multiple cold reductions or forming operations are
required. The use of critical strain reductions is
therefore avoided and higher (or lower) per-
centages of reduction employed.
 16
Graphitization
When high shock-resisting properties are
required in steel, a common practice is to add
silicon in amounts ranging to and even exceeding
2.5070. Silicon has an unusual effect on the manner
in which carbon comes out of solution from
austenite. In these amounts graphite is formed that
is quite soft and friable. Its presence causes the
steel to be softer than a steel of similar carbon con-
tent in which the carbon is all in the combined
form. Graphite is a good lubricating agent, as
well. This combination of properties results in
graphitic steels being free machining, since the
graphite particles act as internal chip breakers and,
at the same time, lubricate the cutting edge of the
tool, resulting in longer tool life.
In many cases, deliberate graphitization in the
annealing operation is desirable so that the steel
can be readily machined. Such a procedure must,
however, be performed cautiously, for once the
graphite has formed it is extremely difficult to take
back into· solution. Therefore, if too much of the
carbon in the steel is changed into graphite, the
steel will not respond properly to subsequent hard-
ening and the desired high level of hardness can-
not be secured. While the tendency of the" usual
types of silicon steels to graphitize is not so strong
as .the graphitizing tendency in cast irons, it is

t
en
en
e
.-
C/)

t
CD
.-
N
C/)
I
I I
c: I I
c~
Recovery I Recrystallizo) Grain Growth
(!) t tion 1
1 1

Annealing Temperature •
Fig. 4-14. Schematic diagram showing recovery, recrystallization and grain growth in cold-worked steel.

nevertheless quite important and should not be
overlooked. As a general rule, graphitization in the
silicon steels is accomplished by full annealing or
with long holding times in critical range annealing.
Graphitization can be avoided by use of subcritical
annealing, or by critical range annealing for a
short period of time.
Quench Annealing
Up to this point the types of annealing con-
sidered for obtaining a fully annealed structure in-
volved heating the steel sufficiently high to form
austenite followed by slow cooling, or by isother-
mal transformation.
There are, however, a number of steels for which
the above procedures will not result in a fully an-
nealed structure. For instance, when such slow
cooling practice is applied to the austenitic, or
Hadfield manganese steel, a quite different result
is obtained. When this steel is slow-cooled, carbide
(as manganese carbide) separates out and forms a
network around the grains, which are normally
austenitic at room temperature because of the
manganese content. This network is hard and brit-
tle, making the steel difficult to machine and lack-
ing in ductility. To render the steel soft and duc-
tile, it is heated to about 1900 F (1035 C) and
cooled rapidly (usually quenched in a liquid).
Heating the steel to 1900 F (1035 C) causes the
cementite to be taken into solution in the high tem-
perature austenite. Quenching prevents the .carbide
from precipitating. As a result, the quenched steel
is completely austenitic and is softer and more duc-
tile than when the cementite is precipitated as a
grain boundary network.
Since the Ms temperature of this steel is below
room temperature, the steel cannot be hardened by
normal quenching procedures. The steel is hard-
enable only through cold working and the work
hardening effect. In the as-quenched austenitic
condition, the steel would have a hardness of
about 190 HB. By cold working, the hardness can
be increased to about 500 HB. This characteristic
of strong work hardening effect makes the Had-
field manganese steel very useful where wear com-
bined with shock is a problem. However, since
long life under wear conditions depends on the
work hardening effect, this steel is not suitable in
applications where pure abrasion without impact
17
is encountered. In such applications, the soft
austenite will not, in the absence of impact, work
harden and as a result will rapidly wear away.
The austenitic stainless steels (AlSI series 200
and 300) are also annealed by the quench anneal-
ing procedure. These steels, because of their alloy
content, are also austenitic at room temperature.
Quenching has a further effect in that it keeps the
carbide in solution in the room temperature aus-
tenite and prevents it from precipitating in the
grain boundaries of the steel. Carbide precipitated
in the grain boundaries is chromium-rich and de-
pletes the surrounding metal of the corrosion-
resistant chromium. Thus, the grain boundaries
are made corrosion-susceptible, and intergranular
corrosion can occur. If the cementite is confined
within the austenite grains, this intergranular cor-
rosion is less likely to develop. High chromium
(23.00 to 27.00070) stainless steel is also annealed
by quenching. This steel, which contains up to a
maximum of 0.35070 carbon, is used principally in
applications involving elevated temperatures and
severely corrosive conditions. The stainless steel
types described are not hardenable by quenching.
Rather, they are hardened by cold working, like
the austenitic manganese steels. For more details
the student is referred to a later lesson in this
course - "Constitution and Heat Treatment of
Stainless Steels. "
Decarburization and Scaling
During the annealing process some unplanned
and undesirable changes may also occur. When
steel is exposed to elevated temperature, especially
when the temperature is sufficiently high to
austenitize the steel, one or more reactions may
take place if no form of protection is employed.
For example, the cementite in the hot steel can
react with hot gases as follows:
Fe3 C + O = 3Fe + CO
2 2
Fe3 C + CO = 3Fe + 2CO
2
Fe 3 C + 2H 2 = 3Fe + CH4
2Fe 3 C + 2N = 6Fe + 2CN
In the above reactions, it may be noted that the
carbon in the cementite is being oxidized and
transformed into a gas which is lost into the at-
 18
mosphere. This reaction causes the carbon content
at the surface of the steel to be lower than the car-
bon content of the interior. Decarburization is the
loss of carbon from the surface of a ferrous alloy
as a result of heating in a medium that reacts with
carbon.
Decarburization
Decarburization is undesirable for several rea-
sons. When steel is hardened, the degree to which
it can harden depends on the carbon content.~
If the carbon content at the surface of a piece
of steel is lower than that below the surface, the
specimen will not harden uniformly. After hard-
ening, the surface will be softer than the interior
because of the difference in carbon content. In
parts that must withstand wear, the disadvantage
of the soft skin is obvious. To avoid this soft skin
it would be necessary to machine away the decar-
burized layer prior to hardening. This means the

a
Fig. 4-15. Photomicrographs showing decarburization in a shock resisting tool steel with high silicon content. (a) Free
ferrite depth (white area at surface) and maximum affected depth (darker area): 4% Picral etch:l lOOX. (b~ C:l d, e and
f) Higher magnifications (500X) of the same condition showing the gradual change in carbon content from the car-
bon~depleted surface (b) to an unaffected zone in the interior (f): 4 % Picral etch.

additional expense of the machining operation as
\llell as the loss of metal. The presence of a decar-
burized layer also results in 10\\· fatigue life.
Causes of Decarburization. Decarburization
may be confined to the surface layer or it may ex-
tend to a considerable depth. \\·ith the loss of car-
bon being either partial or complete. The degree to
\\·hich decarburization occurs depends on the (a)
chemical composition of the steel, (b) temperature
to \\;hich the steel is heated, (c) time at tem-
perature, and (d) atmosphere surrounding the
steel. In general, the higher the carbon content of
the steel the more susceptible it \\'ill be to decar-
burization. Other elements also have a bearing on
the tendency to decarburize. For instance, the
high-speed steels containing molybdenum have a
greater tendency to decarburize than molyb-
denum-free high-speed steels. The furnace at-
mosphere has a pronounced bearing on decar-
burization, and the richer the atmosphere in those
gases which promote decarburization, the greater
will be the reaction. Although steel can decar-
burize at lower temperatures, it does not react to
any appreciable extent until a temperature of
about 1300 F (705 C) is reached. From this point
on, the severity of decarburization increases
rapidly with temperature. At anyone constant
temperature, decarburization increases with time.
Detection. Decarburization is not visible on the
steel surface; thus, one of several methods of
testing must be used to detect its presence. One of
the most direct procedures is the file test. If the
surface of the steel is decarburized, it will not hard-
en and a soft layer will be formed over a hard
core. The soft layer can be detected by stroking the
steel surface with a file. The extent of the decar-
burized layer is determined by filing away· the soft
zone until the hardened core is reached. This is, of
course, a relatively rough test depending to a very
great extent on the skill and experience of the in-
s~ector. In competent hands it can, however,
produce surprisingly accurate results.
A second procedure is to cut a· cross sectional
sample. This surface is ground until it is fairly
smooth and etched with a 5070 solution of nitric
acid in alcohol (nital). The acid will attack the
decarburized surface and the normal carbon con-
tent core at different rates, and this will show up as
a light ring surrounding a dark core. Such a
19
procedure is satisfactory for larger sections where
n10re severe decarburization occurs but is less
satisfactory for smaller sections.
Probably the most satisfactory method of deter-
mining decarburization in small sections is to
polish and etch a cross sectional sample and
examine it under the microscope or to make a
microhardness survey. The loss of carbon is
readily detected from its effect upon the
microstructure and can be measured in increments
of thousandths of an inch. In an instance where
the carbon level of the steel has been reduced par-
tially, the depth to which the steel is affected is
measured and reported as the maximum affected
depth. In some instances, loss of carbon might be a
great deal more severe and no carbon may be
present to some depth. The depth to which decar-
burization is complete is measured and reported as
the free ferrite depth. Under some conditions, it is
possible to decarburize completely to a given depth
and partially to some deeper depth. In such in-
stances, both the free ferrite depth and maximum
affected depth are reported (note Fig. 4-15).
If an accurate measurement of degree of·decar-
burization is needed, chemical analysis is used. In
this procedure, the surface of the steel is machined
away in increments of, say, 0.005 inch or 0.010
inch. The chips from each layer are analyzed in-
dividually for carbon, and the degree and depth of
decarburization reported as sbown in Table 4-4.
Chemical analysis is especially useful in rating
decarburization in the high-carbon, high-alloy tool
steels where the effects of decarburization on the
microstructure are obscured by the high alloy con-
tent. It is also useful in medium-carbon and alloy
steels that have already been hardened.
Decarburization can be measured by methods
that depend on loss of weight and changes in the
electrical or magnetic properties. These tests are a
great deal more difficult to perform and are little
used as a means of production testing.
Table 4·4. Decarburization in a 0.400/0 Carbon Steel
Distance from surface Carbon content
oto 0.010 inch 0.22%
0.011 to 0.020 inch 0.29%
0.021 to 0.030 inch 0.37°/0
0.031 to 0.040 inch 0.40%
Beyond 0.040 inch 0.40%
 20
Scaling
A second undesirable reaction that occurs when
steel is heated is the combination of the iron in the
surface layer with oxygen, to form various oxides
of iron, primarily FeO. This reaction is called
scaling and the oxide formed, scale. Scale is very
hard and brittle and gives considerable difficulty
in subsequent machining operations where it
causes prematurely dulled tools and poor surface
finish. The scaling action also results in a relatively
rough surface so that any fine finish would be
ruined. The scale itself represents a loss of usable
steel and is undesirable from this basic economic
consideration.
Formation of Scale. As .in the case of decar-
burization, scaling occurs at the higher tem-
peratures with the severity increasing as the tem-
perature increases. The longer the time at tem-
perature the greater the deg~ee of scaling. The at-
mosphere surrounding the steel also has. a
pronounced effect on the scaling action, as does
the chemical composition of the steel itself. In
general, oxidizing conditions are necessary for the
formation of scale, according to the following
typical scaling reactions:
Fe + H 2 O-FeO + H 2
Fe + CO 2 -FeO + CO
When scale forms, it may have various physical
characteristics. It may be a soft scale, which is not
adherent to the steel surface and tends to peel off
as large flakes or complete sheets, either of its own
accord or on slight agitation. On the other hand,
the scale might be hard in that it forms a con-
tinuous, tight and extremely tenacious layer on the
surface and is removed only with a good deal of
difficulty.
Methods for Removing Scale. Several pro-
cedures are available for removing scale from steel
products.
One way is to work the surface mechanically by
rolling it between rolls or flexing the piece. Since
the scale is more brittle than the steel, it will pop
off the surface when the steel is bent. Generally,
however, methods such as grit blasting or wire
brushing are more practical, provided the scale is
relatively loose. For removing tightly adherent
scale other procedures are required. Pickling is
often used for removal of light scale. By imm(
sian in an inhibited acid, the scale is dissolve\06
without significant attack on the metal.· -
Various salt baths are also used to eliminate
scale. Here the steel is placed in hot molten salts
which react with the scale, chemically reducing it
or oxidizing it to a greater degree, in which state it
is easily removed. The steel may also be heated in
various gaseous atmospheres where the scale is
chemically reduced, leaving a bright and shiny
steel surface.
In any case, removal of scale represents an extra
operation and added expense, and where scale is
objectionable efforts are frequently made to
prevent its formation rather than remove it af-
terward.
Prevention of Decarburization and Scaling
Although the causes of decarburization and
scaling during annealing are not the same, means
of prevention are generally similar or the same.
One of the oldest methods is to pack the steel in
materials such as cast iron chips or spent c~
burizing compound. Another method is to coat t!l\w
steel with one of several proprietary compounds
made for this purpose. However, the best ap-
proach is to use a prepared atmosphere.
Prepared Atmospheres. Electric furnaces used
with air atmospheres, and gas furnaces used with
atmospheres consisting of the products of com-
bustion, cannot be regulated for complete
elimination of oxidation or decarburization of the
steel being treated. Only atmospheres independent
of the fuel are generally considered satisfactory for
clean or bright annealing. Excessive oxidation
during annealing operations is usually prevented
by the use of prepared controlled atmospheres in
conjunction with a suitable furnace that more or
less completely excludes air and combustion gases
from the heati~g chamber. The gases and gas mix-
tures used for controlled atmospheres depend on
the metal being treated, temperature. of the treat-
ment and surface requirements of the parts being
annealed.
Exothermic Gas. The gas most widely used as a
protective atmosphere for annealing is exotherm;'"

gas. This gas is inexpensive, the raw materials for
making it are readily available, and the results ob-
tained with it are generally excellent. Hydrocarbon
gases, such as natural gas, propane or butane, are
commonly burned in an exothermic gas producer,
creating a combustion reaction that is self-
supporting and heat producing. A commonly used
exothermic gas mixture contains 150/0 H 2 , 10070
CO, 5OJo CO 2 , 1% CH 4 ~nd 69010 N 2 • This gas is
used for bright annealing cold rolled low-carbon
steel strip and wire. It will decarburize medium-
carbon and high-carbon steels, however, because
of the carbon dioxide and water vapor it contains.
To reduce its moisture content, exothermic gas
is sometimes refrigerated, partIcularly in geo-
graphical areas where the temperature of water
used for cooling is high. When surface decar-
burization must be prevented, complete removal
of the water vapor and carbon dioxide is
necessary. Purified exothermic gas, with its carbon
dioxide and water vapor removed, has many ap-
plications for heat treatment of steel without
decarburization.
Purified rich exothermic gas, formed by partial
combustion, is used for short-cycle annealing and
process annealing of medium-carbon and high-
carbon steels of the straight-carbon and alloy
types. For long batch anneals, however, this gas is
unsuitable, because its high carbon monoxide con-
tent results in soot deposits on the work, and
because of the possibility of etching of the surfaces
occurring as a result of the relatively long time at
which the work is in -the critical low-temperature
range where gas reactions. can occur. This effect
does not occur in short-cycle annealing, and the
high-carbon-monoxide gas is then desirable
because of its high carbon potential. The fairly
lean purified gas formed by more complete com-
bustion is used for long-cycle annealing of
medium-carbon and high-carbon steels of the
straight-carbon or alloy types, and for batch and
continuous annealing of low-carbon steel strip for
tin plating. The purified rich exothermic gas used
in short-cycle annealing costs more than the fairly
lean gas used in full annealing at higher tem-
peratures.
Dissociated ammonia is sometimes used as a
protective atmosphere for annealing steel. It is ob-
21
tained by cracking anhydrous ammonia (NH 3) at
about 1650 F (900 C) over a catalyst. The gas mix-
ture thus produced consists of 750/0 H 2 and 25070
N 2 , with traces of NH 3 as an impurity. This mix-
ture is very effective for bright annealing or bright
hardening of h"igh-chromium alloys, such as
stainless steels. Dissociated ammonia should not
be used when slight nitriding (which occurs be-
cause of the molecular nitrogen present) is objec-
tionable. A major disadvantage of using dis-
sociated ammonia is its high cost, which is high-
er than for combusted gas atmospheres.
Steam atmospheres are sometimes used for
process annealing up to 1150 F (620 C). Work-
pieces come out of a steam atmosphere with a very
thin coating of bluish-black oxide which is
sometimes useful as a lubricant retainer for sub-
sequent forming operations.
. For more complete information, see Lesson II,
Protection During Heat Treatment Against Decar-
burization and Scaling.
Furnaces for Annealing
Annealing operations have been performed in
almost every size and type of furnace. In size, the
furnace might vary from a small bench model used
for annealing instrument components to a large
car bottom furnace used for annealing several hun-
dred tons of castings or forgings at one time.
For processes such as isothermal annealing,
however, continuous furnaces with zone tem-
perature control are suggested for best results.
Castings and forgings are commonly annealed
in batch-type box or car bottom-type furnaces.
The essentials of a furnace for good annealing,
regardless of size or design include: (a) capacity
for efficient heating, (b) capability of accurate
temperature control, (c) facilities ~or cooling at the
required rate, and (d) facilities for supplying at-
mosphere protection (if required by the products
being annealed).
For mote information on furnaces for annealing
and other· heat treating operations the student is
referred to Lesson 10, Heat Treating Furnaces.
 22

N orma1izing

Normalizing is defined as heating a ferrous alloy ment where a moderate increase in strength is
to a suitable temperature above the transformation gained without-undue increase in stress. Hardness
range <Ac3) then cooling it in air to a temperature developed by quenching, normalizing, and an-
substantially below the transformation range. nealing are compared in Table 4-5.
Regardless of this precise definition, there is of- When normalizing large sections, it is. common
ten some confusion; most often normalizing is practice to follow with a tempering operation, the
confused with annealing. There is some reason for purpose of which is to reduce stress still further
this because, in many instances, the procedur~ for and to control the mechanical properties to within
normalizing and annealing are the same. For closer limits. Because of the variations that exist in
example, a very low-carbon steel can be almost m~ss, shape and air temperature, precise control
fully annealed by heating above the transfor- of hardness is not possible during normalizing,
mation range and cooling in air. At the other a~d tempering must be resorted to if narrow hard-
erid of the spectrum, some steels (notably air- ness limits are required.
hardening tool steels)" can be fully hardened by
heating above the transformation and cooling in Some further degree of control of mechanical
properties during normalizing can be achieved by
air. Therefore, it should be understood that the
cooling the piece with a forced air current or by
term "normalizing" applies more to the product
being treated and the results rather than precisely atomizing some moisture in the air, forced or
to the procedure. otherwise, and so increasing the cooling rate.
While normalizing is primarily a hardening and
. Purposes of Normalizing strengthening operation, it has other uses. which
are generally considered to be preliminary to fur(
When processing large sections (such as heavy ther heat treating procedures. For instance, as-cast
castings and forgings), it"is not always feasible to or as-rolled material tends to be relatively
strengthen the metal by heating and quenching. nonhomogeneous, and in such condition responds"
Because of the mass effect only a thin surface layer poorly to hardening or annealing. The normalizing
would be hardened to martensite, with the interior operation, with its high temperature of austeni-
of the piece being bainite or pearlite depending on tizing and its relatively rapid cooling rate, breaks
the cooling rate. In large sections, very drastic ex- up the coarse nonuniform structure and refines it
tremes in cooling rate would exist during the into a homogeneous one, as shown in Fig. 4-17.
quench. As a result of variations in temperature, It is for this reason that normalizing frequently
hardness and transformation, the work piece would precedes hardening and annealing operations of
be very highly stressed. Because of the relative in-
effectiveness of quenching and the accompanying Table 4·5. Comparative Brinell Hardness Resulting
high risk of breakage, large sections are usually from Annealing, Normalizing, and Quenching
strengthened by oth"er means. Various Steels
Brinell Hardness Number
In normalizing, the cooling -rate is between that
Quenched and
encountered "in quenching; and that used in an- tempered at
Grade of
nealing, and depends largely on the mass of the steel Annealed Normalized 400 F 1000 F
piece being treated (see Fig. 4-16). As a result of 1040 156 192 514 269
this intermediate cooling rate, the treated piece 1340 183 248 578 363
will possess a hardness and strength somewhat 3140 187 285 555 311
greater than if annealed, but somewhat less than if 4140 179 311 578 341
4340 197 363 555 363
quenched. Because· of the slower cooling rates, 4640 207 255 514 341
normalized sections will not be as higbJ.y stressed 8740 207 302 578 363
as quenched sections. Thus, normalizing is a treat- 9440 192 235 578 311
 ~.
Ii.

1600
4-10. dlam - 2~ in,.bQJOJj 5-in. diom ·3 in. below 6-in. dlam
- (Weight. opprox 431b) - surface - .-.-- - - (Weight, opprox 85 Ib) - surface - - 0 - (Weight, opprox 142 Ib)
1035 steel Ii 1- _ 1035 steel _ 2~ 1020 steel
1200 16 - -
lL. I
• 2"
~
:::J

~ 800
eu
a.
E - -...- - - O . - - & - - - - f - I I I
eu
t-
400--·---

Air at 70F Air ot 75F Air at 73F

0. · . . , . , , , , , , ,
o 40 80 120 160 200 o 40 80 120 160 200 o 40 80 120 160 200
Time, min Time. min Time, min
1600 r
4j in. below I . ·····7-r- -r--
sur face ._ B "4 - in. d iom 10 i-in. diom
3~ (Weight, opprox 410 I b) - - _ a_ _ (Weight, opprox 735 tb)'
1200 --2k --.. ---.. 1040 steel 1040 steel
lJ...
.. ~~ -- - 11
o
Q)
\-
:J
a
t 800·-- --
\-
cv
a.
E
.-eu 400 1- f - - I - - - .--- ...---.. --0---0---.. ·-----.1·.·__ ·& _ _ & 1 __ ::::00",,1............: I I

----.--.... I I 1--1--1
Air 01 9'lto7'lF Air ot 72F
o. , , , , , , , , , I I ,

o 60 120 180 240 300 o 60 120 t80 240 300 o 60 120 180 240 300
Time, min Time, min Time, min
Fig. 4-16. E'ffect ofmass and section size on cooling curves obtained in still air.

fj

,,"'.:'1,
 24

a b

Fig. 4-17. Photomicrographs, at 500X, showing effect of normalizing on microstructure of hypereutectoid steel. (a)
Relatively coarse as-forged microstructure with prominent carbide network and other free carbide. (b) Refined struc-
ture after normalizing; note absence ofa prominent carbide network.

cast or forged steels. Normalizing is also used' as The cooling curves in Fig. 4-16 demonstrate the
a grain refining treatment. effect of mass and section size on cooling in still
air; these data relate to carbon steel bars ranging
Normalizing Practice from 4 to 10 ~ in. in diameter.

Good normalizing practice requires that (a) the
piece being treated be uniformly heated to a tem-
perature high enough to cause complete trans-
formation to austenite, (b) it remain at this tem-
perature long enough to achieve uniform tem-
perature throughout its mass, and (c) it be per-
mitted to cool in still air in a uniform manner.
Uniform cooling requires free circulation of air
around each piece so that there is no area in Which
the cooling has been restricted or accelerated.
Restriction of the cooling rate will alter the
operation from a normalizing to an annealing
treatment. In a strict sense, increasing the cooling
rate over that achieved by free circulation of air
will change the operation from a normalizing to a
que~ching tr~tment. In general, air hardening is
not properly called normalizing, because it does
not put the steel in the normal pearlitic condition
that characterizes normalized steel. Sometimes,
however, when the rate of air cooling in the nor-
m~g operation is intentionally accelerated to
cause formation of some martensite, the process is
referred tc;> as normalizing despite the accelerated
cooling and partial hardening.
Furnace Equipment for Normalizing
Normalizing furnaces are usually heated with
gas or oil and seldom employ protective at-
mospheres beyond that degree of protection
provided automatically .by products of com-
bustion.
Design of the furnace depends largely on size
and shape of the workpieces. A variety of furnaces
are used; the primary requirements being to heat
the work thoroughly and then expose the work
load to air so it will cool rapidly. Because of the
latter requirement, furnaces used for annealing
may not be adaptable to normalizing. As exam-
ples, small forgings are usually normalized as
received from the forge shop. They are placed or
piled loosely QD the pans to maximum depth of 3
in. A typical furnace has five pans in each of the
three funiace zones. Heating is adjusted so that the
work reaches normalizing temperature in the last
zone, following which the pans of work are
discharged o~to a cooling conveyor. The work
cools in still air to below 900 F (480 C)- on the pans
and is then discharged into tote boxes, in which it

cools to room temperature. Total furnace time is
approximately 3 ~ hr, during which period the
work has been held at the normalizing temperature
for only 1 hr.
Normalizing large open-die forgings is usually
performed in batch-type furnaces pyrometrically
controlled to close temperature ranges. Forgings
are held at the normalizingtemperature long enough
to allow complete austenitizing and carbide solu-
tion (usually 1 hr per inch of section thickness)
and are then cooled in still air.
Small castings are commonly normalized in the
same manner as forgings. Large castings and
heavy forgings are normalized in various kinds of
furnaces, up to and including car bottom furnaces
wherein the charge is heated, soaked thoroughly
25
then the door is opened and the car is pulled out so
that the charge is exposed to the air.
·As another example, the catenary, or free-loop,
type 9f furnace is designed for continuous nor-
malizing of cold reduced steel unwound from
coils; it does not have rolls or any other type of
conveyor to support material passing through the
heating zone. The heating zones of catenary fur-
naces range in length from 20 to 50 ft. The
preheating and cooling zones are usually shorter
than those in conveyor-type furnaces, and for
some kinds of work may be omitted entirely. At
their exit end, catenary furnaces may have pickling
or other descaling equipment for removing sur-
face oxides formed on the steel during nor-
malizing.
For additional information on heat treating fur-
naces, see Lesson 10 "Heat Treating Furnaces."
 I

J
t. MATERIALS ENGINEERING INSTITUTE·
-----~
I
~
C10L4,C41L5
Test Series F Annealing and Normalizing Steel Page 1

PLEASE PRINT YOUR NAME AND ADDRESS BELOW RETURN TO: ASM Intemations/@
MATERIALS PARK, OH 44073~002
NAME REGISTRAnONNO _

STREET _

CITY STATE ZlP DATE _

MULTIPLE CHOICE

Place the appropriate letter (A, B, C, D) in the box. (Only one answer is correct
and each question has a value of10 points.)

1. When a steel is heated to at least 28°C (50°F) above the upper critical and cooled in
a controlled manner, tbe process is known as: D
(A) critical range annealing.
(B) full annealing.
(C) subcriticaI annealing.
. (D) quench annealing.
I
The practice of annealing so that transformation occurs at constant temperature is
I
I
I
2.
referred to as: D
I (A) quench annealing.
I (B) process annealing.
(C) isothermal annealing.
I
(D) subcritical annealing.
I
I
I
I
I
3. Although normalizing is primarily a strengthening operation, another important
purpose for normalizjng is: o
I (A) to obtain the softest possible structure.
I (B) refming of nonhomogeneous structures.
I (C) elimination of dissolved gases.
I (D) to relieve stresses following abardemng operation.
I
I
I
I
4. Heating of steel to a temperature below, but usually close to the lower limit of the
transformation range and cooling slowly is commonly called:

(A) martempering.
o
I
I (B) quench annealing.
I
(C) process annealing.
(D) isothermal annealing.
 MATERIALS ENGINEERING INSTITUTE

C10L4,C41L5
Test Series F Annealing and Normalizing Steel Page 2

5. Certain stainless steels (notably the AISI 200 and 300 series) and some other special
types of steel are annealed by a process known as: D
(A) quench annealing.
(B) blue annealing.
(C) pot annealing.
. (D) finish annealing.

6. Controlled heating and cooling of steel to produce a globular form of carbide is

called: D
(A) quench annealing.
(B) normalizing.
(C) graphitization anneal.
(D) spheroidizing.

7. In steels which have been severely cold worked the crystal reformation recrystalli-
zation temperature can be as low as:

(A) 355°C (675°F)

(B) 460°C (8600f)
(C) 410°C (775°F)
(0) 385°C (725°F)

8. The most widely used protective atmosphere for bright annealing low carbon steel
is: D
(A) ammonia.
(B) steam.
(C) exothermic gas.
(0) carbon dioxide.

9. Subcritical annealing is often synonymous with:

(A) austempering.
D
(B) process annealing.
(C) full annealing.
(0) quench annealing.

10. When a plain carbonor low alloy steelis heatedto atemperature substantially above
~ and cooled in air, the process is called: '

(A) process annealing. (C) normalizing.

(B) stress relief annealing. (D) full annealing.
 ASM International® is a Society whose
mission is to gather, process, and
disseminate technical information.
ASM fosters the understanding and
application of engineered materials and
The Materials their research, design, reliable manu-
Information Society
facture, use, and economic and social
benefits. This is accomplished via a
unique global information-sharing
network of interaction among members
in forums and meetings, education
programs, and through publications
and electronic media.

Printed on Docutech October 1994

 The Materials
Information Society

Hardening of Steel

Course 10
Lesson, Test 5
Course 11
Lesson, Test 13
Course 40
Lesson, Test 4
Course 41
Lesson, Test 6
 Metric Conversion Factors
To convert from To Multiply by

in. mm 25.4
in. m 2S.4xIO-'
IDIl J.Im 25.4
).lin. J.Im 25.4
• 2
m. m2 6.45 x lQ-4
• 3
m. m3 1.64 x lQ-s
ft. m 3.048 x 10-1
ft. 2 m2 9.29 x 1()-2
ft.' m3 2.831 x 10-2
oz. g 2.834xl()l
lb. kg 4.536 x 10-1
Btu J l.OS4x 10'
Btullb. - "F Jlkg - K 4.18 x 1()3
BtuifL - hr - "F W/m-K 1.730
10inJm.JOF 10" mmlmmJOC 1.8
psi Pa 6.895 x 10'
psi kPa 6.895
ksi kPa 6.895 x 10'
ksi MPa 6.895
ksi - in. 1I2 MPa-m1J2 1.099
ksi~in. MPa-~m 1.099
ozl gf 28.4
Ibf kgf 4.536 x 10-1
lbf N 4.448
Ibf- ft. N-m(orJ) 1356
IbfJin.2 kgfIcm2 14.223
IbfJin. 3 kgflm3 2.768 x 10'
lbJft.3 kg/m3 16.019
lbJin.3 g1cm3 2.768 x 101
IbJm. 3 kglm3 2.768 x 10'
gal (U.5. liquid) L 3.785
gal (U.5. liquid) m' 3.785 x 10'"3
IbJgal gIL 119.826
ft./gal mlIL 748
OF °C (OF-32)/l.8
"F K (OF + 459.67)11.8
"C OF (OC - 1.8) + 32
°C K °C+ 273.15
K °C °C-273.15

Multiple and Submultiple Units

l()ls ..•.._._...•_ exa 10-1 _._ _ •••_ _ .deci

1()l$ ••..._ _ peta 1~ _ centi

1()l2 _ •••••••••••••••••••• tera 10-3 •••_._ _ D1i1li
1C1 ••••••••••••••_ giga 10" _ micro
1f1' _ _mega 1()-9 DaDO
1~ kilo 10-12 ••••••••••••••••••••••••- pico
1()2 _••••_••••• _hecto 10-15 •••••_._••__••••••• femto
1()l _••••••• delca 10-11 atto

Abbreviations
J _Joule m _•.__•.• meter Pa...•••..•.•.•.•••...•pascal
kgf••.••• kilogram force mm.•.__._•. millimeter K kelvin
L _ _ liter N newton W watt

Technical Editor
Howard E. Boyer
Consultant

Editor
Philip D. HarVey
Metals Engineering Instiblte, ~erican Society for Metals

e 1979 American Scciety for Metats

This material m1rf not be repn:duced in whole cr in part in any form whatsoe¥er. Fuly potected by copyright.
 Hardening Of Steel
Ever since man began to utilize machinery to
perform his tasks, he has been confronted with the
problem of making the individual parts lighter,
stronger and more wear resistant. This was
especially true during the development ot the iro~
age. The blacksmith was well acquainted with the
need for increased strength and wearing ability.
He solved the problem by heating a steel part to
some temperature where it was red hot, and then
plunging it into water or other favored medium.
From the experience gained during his apprentice-
ship, he knew that such a procedure would cause
the steel to become hard, and at the same time
more wear resistant. Indeed this method of in-
creasing the hardness, strength and wear resistance
of steel was known long before the Christian era,
as indicated by ancient steel implements. While the
procedure was used for many centuries, the prin-
ciples underlying the process were not understood
until modem times, and the maximum possible
benefit to be derived from hardening was not
realized until more recently. With real knowledge
of hardening, it was possible to take full advantage
of steel and make it more useful, meaning that
components of modem machines, motor vehicles,
airplanes and space vehicles could be made harder,
stronger and more wear resistant.
Commercial Importance of Steel
Basically there are three phenomena that ac-
count for the commercial imponance of steel; all
three relating to the ability of steel to be heat
treated and its properties to be controlled by heat
treatment. The three phenomena are: (a) the
element iron is allotropic - meaning it can exist in
more than one crystalline form, (b) the element car-
bon is also allotropic and (c) the carbon atom is
much smaller than the iron atom.
As an elemental metal, commercially pure iron
has only limited engineering usefulness, regardless
of its allotropy. In abundance in the earth's crust,
iron is equaled and exceeded· by other metallic
elements. Carbon is the key alloy addition which
capitalizes on the allotropic phenomenon and lifts
iron from mediocrity into the position of the
world's unique structural material, broadly known
as steel. Even in the ~ghly alloyed stainless steels,
it is the minor constituent carbon that virtually
controls the engineering properties. Furthermore,
due to the manufacturing processes, carbon in ef-
fective quantities persists in all irons and
J
steels
unless special methods are used to minimize it.
Carbon, above all other elements, is the key
operator in jamming the iron matrix during the
transformation from austenite to ferrite, and
producing the hard transition structure known as
martensite.
As a simple analogy, consider layers of mixed
oranges and peaches in a store window. Con-
ceivably, they would shift among one another in a
somewhat orderly fashion because they are of
nearly the same size. Now imagine a mixture of
oranges and grapes. In this case, the small fruit
 ·2
would occupy the voids between the larger fruit.
By this analogy, carbon is the grape, for its atoms
are so much smaller than the iron atoms, thus the
carbon atoms occupy the interstices.
Iron-Carbon Constitutional Diagram
As is true for other pure metals, the con-
stitutional diagram for pure iron is a straight line
as shown in Fig. 5-1a. Here it is seen that the
structure is ferrite (alpha iron) from 32 F (0 C) to
1674 F (912 C). In this temperature range, the
atom arrangement is a body-centered cubic (bee).
At 1674 F (912 C), the crystal structure changes to
a face-centered cubic (fcc) which is austenite (gam-
ma iron, 2-Fe)...c\t a temperature of2541 F (1394 C),
the structure reverts to a body-centered cubic and
becomes delta iron ( 0 -Fe) in which condition it
remains to the melting point (2800 F or 1538 C). In
hardening, however, the principal interest is in the
temperature range of 32 F (0 C) to approximately
100 to 200 F (55 to 110 C) above the point A 3 •
Effect of Carbon. When carbon is added to pure
iron, the transformation temperatures charac-
teristic of pure iron become altered progressively
as the carbon content increases. This is indicated in
Fig. 5-1 b. For any fixed rate of cooling, the steel
remains liquid longer, transforms from delta
ferrite into austenite earli~r, and hesitatingly reverts
again to ferrite at such lowered ternperat'ures
that a preference of the carbon atoms for the
austenite crystal structure is clearly expressed.
£
• become austenitic (changing to gamma phase) by
OF
-- 0C
2800 . ·1538
FREEZING
POINT ,
LIQUID
FERRITE:
2541 ..··1394 (o-Fe)
AUSTENITE
C""-Fe)
1674- 912- -A3
FERRITE
(a-Fe)
INCREASING CARBON
32 ..·· ·..···0
, , //////
.......................
IRON (PURE Fe)" 0
(a) (JOOCfo Fe) (b)
Fig. 5-1. (aJ Constitutiontll diagramfoT pure iron and (b)
initialeffects ofcarbon on theprincipalthermalpoints of
pure iron.
Again the vertical line of Fig. 5-1 a is the con-
stitutional diagram for l'ure iron (Fe). On th~
right, this diagram is expanded horizontally t
depict the initial effects of carbon on the principal
thermal points of. pure iron. Thus, each vertical
dashed line, like the solid line at the left, is a con-
stitutional diagram, but only for iron containing
that particular percentage of carbon. It should be
noted that carbon lowers the freezing point of
iron, and that it broadens the temperature range of
austenite by raising the temperature AJ. at which
(delta) ferrite changes to austenite and by lowering
the temperature A 3 at which the austenite reverts
to (alpha) ferrite. Hence, carbon is said to be an
austenitizing element. The spread, of arrows at As
covers a two-phase region, which signifies that
austenite is retained fully down to the tem-
peratures of the heavy arrow, and only in part
down through the zone of the lesser arrows.
The fundamentals of hardening of steel are
directly related to the phenomena described above.
Transformation of Austenite
Basically, hardening of steel is quite simple and
consists of achieving a microstructure that wi!'
result in the steel becoming harder and stronger.
The microstructural constituents that are most
desirable in hardened steel are martensite and, to a
lesser degree, bainite. The annealed microstruc-
tures, pearlite and spheroidite, are avoided insofar
as possible. The principles of hardening are also
quite simple and consist of causing the steel to
FREEZING
POINT heating, and then transforming the austenite under
such conditions that martensite or bainite forms and
the soft transformation products are avoided. The
hardening procedure therefore consists of two
essential steps, austenitizing and transforming.
Austenitizing. In the fIrst step, the steel is heated
from room temperature to some elevated tempera-
ture where austenite can form. This should be done
in such a fashion that the steel is uniformly and
evenly heated with a minimum of temperature
differential existing within the piece, or from piece
0.10 0.20 to piece. After the proper temperature has been
C,o;o-
reached, the metal must be held long enough to
ensure that proper diffusion has taken place and all
of the austenite is uniform and essentially
homo~eneous.
 3
Table 5·1. Austenitizing Temperatures for Direct Hardening Carbon and Alloy Steels
Steel, Steel,
AISI, UNS Temperature, AISI. UNS Temperature,
SAE No. F SAE No. F
Carbon Steels
1025 G10250 1575-1650 1059 G10590 1475·1550
1026 G10260 1575-1625 1060 G10600 1475-1550
1029 G10290 1550 -1600 1070 G10700 1475-1550
1030 G10300 1550-1600 1078 G10780 1450-1500
1035 G10350 1525-1575 1080 G10800 1450·1500
1037 G10370 1525·1575 1084 G10840 1450-1500
1038 G10380 1525·1575 1086 G10860 1450·1500
1040 G10400 1525·1575 1090 G10900 1450-1500
1042 G10420 1475-1550 1095 G10950 1450-1500
1043 G10430 1475-1550 1526 G15260 1575-1650
1044 G10440 1475-1550 1527 G15270 1575-1650
1045 G10450 1475-1550 1541 G15410 1475-1550
1046 G10460 1475-1550 1548 G15480 1475-1550
1049 G10490 1475·1550 1551 G15510 1475-1550
1050 G10500 1475-1550 1552 G15520 1475-1550
1053 G10530 1475-1550 1561 G15610 1475-1550
1055 G10550 1475-1550 1566 G15660 1475-1550
Alloy Steels
1330H H13300 1525·1575 4340H H43400 1500 ·1550
1335H H13350 1500 -1550 5130H H51300 1525-1575
1340H H13400 1500-1550 5135H H51350 1500·1550
1345H H13450 1500·1550 5140H H51400 1500-1550
4027H H40270 1525·1550 5150H H51500 1475·1550
4028H H40280 1525·1550 5155H H51550 1475-1550
4032H H40320 1500·1575 5160H H51600 1475-1550
4037H H40370 1500·1575 6l50H H61500 1550-1625
4042H H40420 1550·1600 8627H H86270 1525 ·1600
4047H H40470 1550 ·1600 8630H H86300 1525-1600
4130H H41300 1500 -1600 8637H H86370 1525·1575
4135H H41350 1550 -1600 8640H H86400 1525-1575
4137H H41370 1550 ·1600 8642H H86420 1500·1575
4140H H41400 1550-1600 8645H H86450 1500-1575
4142H H41420 1550 - 1600 8650H H86500 1500·1575
4145H H41450 1500 -1550 8655H H86550 1475·1550
4147H H41470 1500-1550 8660H H86600 1500 ·1575
4150H H41500 1500-1550 8740H H87400 1525-1575
4161H H41610 1500 - 1550 9260H H92600 1575-1625 .
Standard Alloy Boron·H Steels
50B40H H50401 1550·1600 51B60H H51601 1500-1550
50B44H H50441 1550 -1600 81B45H H81451 1500 -1575
50B46H H50461 1525-1575 B6B30H H86301 1525·1600
50BSOH H50501 1525-1575 86845H H86451 1550-1575
50B60H H50601 1500-1550 94830H H94301 1550-1600

For successful hardening, the steel must be thor-
Qughly austenitized. Anything less than complete aus-
tenitizing will result in mixed microstructures and poor
mechanical properties. Under some conditions of in-
sufficientaustenitizing, cracking is likely to occur. When
austenitizing hypoeutectoid and euteetoid steels, it is the
general practice to heat to least SO F (28 C) and more
often nearer 100 F (55 C) above the Ac3 temperature for
the specific steel being heated. The steel is then held at
this temperature for complete solution of carbides and
homogeneous diffusion of carbon in the austenite.
Carbon steels with large carbides in the microstIUeture
or alloy steels require either the higher austenitizing
temperatures or soaking for several minutes after the
steel has reached the proper austenitizing temperature
throughout its cross section. Very coarse cementite or
alloy carbides require longer times to dissolve and
diffuse into the austenite. Total heating time for a
specific thickness is a function of the heating medium.
With hypereutectoid steels, the austenitizing
conditions are somewhat different. If the hyper-
 4
eutectoid steels were heated above the Acm, Table 5-2. Reheating (Austenitizing) Tempera-
they would in the presence of air and at the very tures for Hardening Carburized Steel and Alloy
high temperature experience severe decar- Steels (SAE)
burization, scaling and grain growth. Fortunately, steel Temperature. F Steel Temperature. F
in the hypereuteetoid steels" with their excess of
carbide, enough carbon solid solution occurs at carbon Steels 4626 ..• 1500-1550
lower temperatures to assure proper hardening, so 1010 ..• 1400-1450 4718 1500-1550
1012 1400-1450 4720 1500-1550
that for these steels it is the general practice to heat 1015 1400-1450 4815 ..• 1475-1525
about 100 F (55 C) above the Acl for oil or polymer 1016 •.. 1400-1450 4817 1475-1525
1011 ... 1400-1450 4820 1475-1525
quenching and about SO F (10 C) above for water or 1018 •.. 1400-1450
1019 1400-1450 8115 1550-1600
brine.
in some of the very high alloy tool steels, the 1020 1400-1450 8615 1550-1600
1021 1400-1450 8617 1550-1600
proper degree of solid solution is not achieved un- 1022 1400-1450 8620 1550-1600
til temperatures greatly in excess of the Ac3 are 1023 1400-1450 8622 1550-1600
8625 1550-1600
reached. In steels of this type, proper solution of Free-Cutting 8627 1550-1600
carbon Steels
the carbon and alloy is a very important con- 8720 1550-1600
1108 1400-1450 8822 1550-1600
sideration. In such cases, decarburization, scaling 1109 1400-1450
9310 .•. 1450-1525
and other ~onditions make it mandatory that very 1115 1400-1450
1117 .. _ 1400-1450 Alloy B Steels
short holding times be employed, or special 1118 •.. 1400-1450 331011•.•... 1550
precautions be taken. These cases are, however, 1119 1400-1450 4027!1..•... 1600
1120 1400-1450
confined to special steels in the tool steel group. 402811 1600
For plain carbon steel, constructional alloy steel Alloy Steels 4032!1 •..... 1600
3310 1450-1525 481511•..... 1550
and many tool steels, the austenitizing practices 4320 1525-1550 481711 1550
outlined are quite satisfactory. In the lower car- 4615 1500-1550· 482011 1550
4617 1500-1550 862511 1600
bon, lower- alloy steels the extent of solution may 862711..•... 1600
4620 1500-1550
be less critical. 4621 1500-1550 9310li••.... 1550
Best hardening practice dictates that the max- not known, C!1'1d its origin is lost in antiquity.
imum temperature needed to achieve the nec- Descriptions of hardening by quenching are found
essary extent of carbide solution should not be in early writings, such as Homer's Odyssey.
exceeded. Recommended austenitizing tem-
peratures for most of the AISI carbon and alloy The quenching medium used depends on the
steels are .presented in Table 5-1. Recommended cooling rate required for the transformation
austenitizing temperatures (either for reheating or characteristics of the specific steel being hardened.
lowered from the carburizing temperature) are For example, in certain high alloy tool steels, the
given in Table 5-2 for carburizing grades. When rate of transformation provided by quenching in
fully hardened core properties are specified, it still air is sufficient to transform the austenite to a
is necessary to rai~e the temperature slightly above completely martensitic structure. For most steels,
the Ae3 for that specific grade of steel. however, water or another suitable liquid is used.
Quenching mediums are discussed later in this
Quenching lesson.
After the steel has been austenitized, the second Regardless of the type of quenching medium,
step in the hardening operation is to transform the the primary purpose of the quenching medium is
soft austenite into one of the harder microstruc- to remove heat from the steel and cool it to the
tures; usually martensite or lower bainite. This ·tempe~ature region where martensite is formed,
normally is accomplished by quenching, which is and at such a rate that transformation to softer
the rapid cooling of steel from an elevated tem- microstructures is avoided. The rate of cooling
perature. Usually, but not-always, this is done by necessary to ensure hardening is called the critical
immersion into a liquid such as water or oil, where cooling rate; that is, the minimum rate of cooling-
the steel is rapidly cooled to approximately the which will produce the maximum hardness.
temperature of the quenching medium. Exactly Cooling at rates faster than the critical cooling rate
when and where quenching was first developed is will not result in any increase in hardness. Slower

rates, howe\'er, will result in some of the austenite
transforming to one of the soft microstructures,
thereby lowering the overall hardness of the steel.
In quenching, both the Isothermal Trans-
formation Diagram and the Continuous Cooling
Diagram may be used as a guide· to proper
procedure, although the latter diagram has its
greates~ effectiveness in this field. When using the
Isothermal Transformation Diagram (also known
as the TTr or Time, Temperature, Trans-
formation Diagram or S-curVe), to plan a hard-
enin~ pro~d~e, i~ should be realized that the
pearlite nose of the diagram represents the critical
zone in the quenching procedure insofar as
avoidance of th~ softer transformation products is
concerned. It is in the region of the pearlite nose
(sometimes called the gap) that the .minimum
amount of time is required for transformation to
the soft produCts. Therefore, if a steel is hardened
successfully, the critical cooling rate must be such
..........-.
-.
.....
...
-
Critical Cooling.......
Rate
Product Of Transformation
100 % Martensite
Time
Fig. 5-2. Schematic isothermal transformation diagram showing the critical cooling rate necessaryfor complete trans-
formation to martensite.
5-
that the pearlite nose of the Isothermal Trans-
formation Diagram is avoided, as indicated in
Fig. 5-2.
For certain types of steel, the arrangement of the
isothermal diagram is such that the position of the
pearlite nose is far to the right, and relatively slow
rates of cool may be employed without any danger
of the austenite transforming to pearlite. This con-
dition is illustrated in Fig. 5-3a where, for that
particular grade of steel, the first transformation
to pearlite takes place at about 1400 F (760 C) after
approximately 15 minutes. With steels of this type,
rapid cooling in liquid quenchants to avoid the
pearlite nose is not necessary. Much slower rates
of cooling, such as would be encountered in air
cooling, can be employed and proper hardening
achieved. Steels of this type are usually designated
as air hardening steels.
On the other hand, some plain carbon steels may
have an Isothermal Transformation Diagram of
•
 ·6
such a nature that the pearlite nose cannot be
avoided, as shown in Fig 5-3b. In such instances,
pearlite begins to form as soon as the transforma-
tion temperature is reached, and unless very small
sections are being processed, it is impossible to
avoid at least partial transformation to pearlite.
Such steels cannot transform to 1000'/0 martensite
and thus never reach the full hardness potential
indicated by the carbon content.
If very small sections of this type of steel are
quenched, the resultant cooling rate will be such
that relatively little time is spent in the region of
the pearlite nose. Thus, not much transformation
to pearlite will occur 'and near maximum hardness
can be achieved.
When using the isothermal transformation
diagrams to predict quenched results, remember
that the diagrams were developed under conditions
approaching equilibrium and do not take into con-
sideration the effect of increased cooling rate on
the suppression of the ferrite, pearlite and bainite
stages. Therefore, a. certain degree of inaccuracy
exists when translating Isothermal Transformation
D~agram data to conditions of accelerated cooling,
as encountered in quenching, and they must be
judiciously interpreted.
In the absence of other data, the use ofisothermal
transformation diagrams for predicting quenched
results is acceptable, if the possibility of error
is recognized and some compensation made for it.
Continuous cooling diagrams are more useful,
because they are truly indicative of what may be
expected in a quenched product with a given
cooling rate. Working with continuous cooling
diagrams to plan a quenching procedure, or
predict the results of quenching, is not very dif-
ficult. As seen in Fig. 5-4, the reactions that occur
at a given temperature with a given cooling rate
have already been determined experimentally. All
that is necessary is to cool at whatever rate will
result in the desired microstructure. For instance,
from Fig. 5-4 it is seen that if a completely mar- ,
tensitic steel is to be obtained, the austenite must
be cooled at a rate of X degree F per hour, or
faster. This is the critical cooling rate. If some
lesser overall hardness is acceptable, as would
result from a mixture of martensite, bainite and
some ferrite. cooling rate in the region between Y
degree F to X degree F per hour may be used. If
a soft structure is desired for optimum machin-
ability or other purpose, a very slow cooling rate,
Z degree F per hour, may be used.
Quenching Systems
In instances where only a few small parts are
being quenched at one time and the quenching
medium is allowed to cool before any more quench-
ing is done, the quenching system can be as sim-
ple as a barrel that contains the quenching
medium, regardless of the medium used. In such a
simple operation, the necessary agitation of the
workpiece can be obtained by manual movement.
However, in production quenching, to retain
control of the quenching medium temperature a
much more sophisticated system is required, even
to the degree of that illustrated in Fig. 5-5. This
system includes all the features of a high-
production quenching system - quenching tank,
pumps, heat exchanger, storage tank, filters and
other. All of these components are required for
proper control of temperature and agitation.
There are, of course, many systems which are
somewhere in between the simple tan~ and the
complex system shown in Fig. 5-5. Many modern
heat treating furnaces include an integral quench-
ing system as shown in Fig. 5-6.
Quenching Mediums
Quenching has one serious disadvantage.
Because of the extreme variation in rates of
cooling which exists throughout the cross section
of the piece being quenched, severe and
nonuniform volume changes occur. These are due
to unevenness in both temperature and trans-
formation. The net result is that the quenched
piece may be very highly stressed and may warp
badly or even fracture. This is particularly true of
specimens containing large variations in cross-
sectional dimensions, sharp corners and other in-
tricate details, such as bolt holes, keyways and
splines.
Because of the practical considerations
discussed above, it is purely a matter of good
judgment to select a quenching medium that will
cool the part only as fast as required to harden the
particular steel involved. The critical cooling rate
 7

140 JCr-Mo-W
C 0.32
Mn 0.35
1200 Si 0.95
I.L Cr 486
0 W 1.29
I 1000 Mo lAS
C1)
~
::7 (From Payson 8 Klein)
~

e
CD
80 Trons. ASM 31,1943,p.231

c.
E
~

200

(0) 2 461020 60 600

Time - Minutes

1500

1400
.t4l1' ...

u..
0
J

...... I0.45 %
G)

:J
Carbon Steel(
e C 0.45
CD Mn 0.67
c. 8 Mo 0.009
E
~ (After Blanchard, Parke 8 HerziG)
Trans.ASM 29,2,1941, p.321

100 1000 10,000

(b) Time - Seconds
Fig. 5-3. Two extremes in isothermal transformation diagrams. fa) A deep hardening (or air hardening) steel; (b) a
shallow hardening steel.
 8
depends on the section thickness and the trans-
formation characteristics of the steel.
Quenching mediums commo~yused are listed as
follows in order or their decreasing cooling power.
Molten salt and molten metals -are also used for
quenching, but usually for special applicatioDS
which are discussed in subsequent sections of th:
lesson.

1. An aqueous solution of 5 to 10070 brine or Water and water-base solutions are the least ex-
caustic.soda.
pensive quenching mediums, and approach the
2. Water. maximum cooling rate obtainable in a liquid. One
3. Synthetic quenches. disadvantage of plain. water is that its rapid
4. Fast oil. cooling persists through the lower temperature
range, in which distortion or cracking is 'most
S. Conventional oil. likely to occur. COI;lSequently, water quenching is
6. Forced air. usually restricted to simple, symmetrical parts
7. Still air. made from shallow-hardening grades of steel.

.~.

I
i:_.
Austenite
~~.;

Unstable
0\
.:~

.~:\
':~
-:.
::~
t::
VA
!IE Austenite .... Bainite
~:t:·~ t:~
.-
~
~ ~
~ ~
..---M· - - -............--------~~..~----------I
~
• ~
* ~
Austenite ..... Martensit~£
~. 1j.
.....--Mf----~·:...
~ --------~I.~l.------ ..
oo:~r XOFIHoul :::[f.:: YOFIHour

Time •
Fig. 54. Schematic continuous coolin.g transformation diagram for a hypoeuteetoid steel. The uiagrllm shows the
cooling rates needed for hardening to 100% martensite:l XO Flhr:/· to a mixture of martensite and bainite plus a sma'l
amount offerrite. yo FIhr:/· and to ferrite pluspearlite:/ ZO FIJ:zr.

Self-contained
oil cooter
I
II
VI Removable~ ~ ;
I I
::
II
~
I:
.
••
Fig. 5-5. Typical oil quenching system consisting ofquench tank, pumps, cooler, storage tank and other components
of a complete system.
Another disadvantage of using plain water is
that its vapor blanket stage may be prolonged
(pockets of steam around the workpiece) which
can result in nonuniform hardness and un-
favorable distribution of stresses. Also, plain
water is temperature sensitive; that is, cooling
power decreases rapidly when the temperature of
the quenching water exceeds about 90 F (32 C).
Brine and Caustic. Aqueous solutions con-
taining 5 to 10070 sodium chloride, calcium
chloride, or sodium hydroxide have cooling rates
faster than plain water. Their operating tem-
perature is less critical compared with plain water;
in general, the brine and caustic solutions can
operate up to 125 F (SO C) without significant loss
of cooling power. Also, there is less likelihood of
steam pockets that result in nonuniform quenching
and possible cracking.
Fig. 5-6. Typical shaker-hearth typefurnace thaI features an inte~rQI quenching SJ'SleI11.
9
Distributor Quench tonk-~~-"'~
Positive-displacement
. "" I
priming pump
suctIon I I
I stub I I
I I I
I Footvolve~ I
I I I
-~
Solutions of sodium and calcium chloride have
the disadvantage of being, corrosive to metal
equipment. Sodium hydroxide (caustic) solutions
do not corrode most metals used in quenching
equipment, but splash can cause severe skin
irritation to operators.
Synthetic quenching mediums were developed
for the purpose of having a quenching medium
that possessed cooling rates somewhere between
water or other aqueous solutions, and the
relatively slow cooling characteristics of oil. Many
different proprietary compounds are available
having cooling rates ranging from as fast as water
to as slow as oil are now on the market. This gives
the heat treater a chance to select the synthetic
quench that best fits the requirements. A disad-
vantage of the synthetic quenches is that the
ingredients which are added to water tend to cling
 ·10
Table 5-3. Quenching Severity Evaluations for
Three Common Quenchants
AI1tatkm on Water
None •••••••• 0.25 to 0.30 1.0
Mild ••••••••• 0.3 to 0.35 1.0 to 1.1 2.0 to 2.2
Moderate ••••• 0.35 to 0.40 Uta 1.3
Good •••••••• 0.4 to 0.5 1.4 to 1.5
strong ••••••• 0.5 toO.s 1.6 to 2.0
Violent •••••• 0.8 to 1.1 4.0
to the work thus reducing their concentration.
Therefore, it is mandatory that a synthetic quench-
ing bath be analyzed frequently and replenished
as required.
Quenching oiJs.-are of two general types; fast and
conventional. Water-oil emulsions represent a
third type of oil quenchant, unclassifiable as either
conventional or fast. Characteristics of these
emulsions can vary a great deal depending on the
percentages of water and oil.
A conventional quenching oil is an oil that con-
tains no additives that alter its cooling charac-
teristics. Conventional quenching oils are fractions
produced by distilling crude oils and blended to
produce viscosities ofabout 100 SUS at 100 F (38 C).
Fast quenching oils are blends, usually of lower
viscosity than are conventional oils. They contain
specially developed proprietary additives 'that'
BriDe
provide a faster quenching rate.
2.0
Quenching Severity. The ability of a quenching
medium to extract heat from a hot workpiece c~
be expressed in terms of H-Value (the severity of
5.0
quench). I f the H- Value of still water is assumed as
1.0, comparative values for oil, water, and brine
are as shown in Table 5-3. Agitation is a major
factor in any quenching operation and is also con-
sidered in Table 5-3.
For more information on quenching and the
variables that affect quenching, see Lesson 7
"Factors Affecting Hardness and Hardenability" ,
and Lesson 15 of MEl Course 6, "Quenching
Media and Equipment".
Interrupted Quenching Techniques
At least three techniques have been developed
for interrupted quenching. Objectives in the use of
these techni,ques include development of specific
microstructures, minimizing distortion and warp-

IAustempering I
Austenite
Stable
Ael-~~~;--------==========~=-----
~~. '~:.

I
~,
':!:.
~. Ferrite
?~
~~ and
COOling--'. Cementite
Stable
~~
.~:
~
~,
~~.

·~~·:.Equalizing.
~:.:~"~",•.. :,....":::.-::,,,!,,_ _ ,,,:o,: .••:"'~:~:":=',~~'~~~"::<':....; .:•..•

To Bainite

~-M f -------------~P-ro-d~u-c~t-O~f-=T=-r-a-ns~f:-o-rm-a~tion:
Untempered Bainite

Time---'"
Fig. 5-7. Schematic diagram showing meth"od ofaustempering.

ing, and mlDlII11Z1ng susceptibility to cracking.
Actually, the procedures were developed at an
early date through trail and error. Their benefits in
relation to lower internal stress were recognized as
early as 1750. Their full utilization was not
realized, however, until about 1940, when un-
derstanding of the austenite transformation was
more fully developed. Procedures such as austem-
pering and manempering were then removed from
the realm of art and placed in that of an exact
science. Three different techniqll~s are discussed
separately in the following sections.
Austempcring
Austempering .consists of austenitizing the steel
and then quenching into some suitable medium
which is maintained at a temperature in the r:inge
of bainite transformation. This procedure is
illustrated in Fig. 5-7. The object is to isothermally
transform austenite to bainite under which con-
ditions no subsequent tempering is needed. In
commercial practice, most steels are isothermally
transformed to bainite in a temperature range of
about 400 to 700 F (205 to ,370 C). The range is not
fixed and will vary somewhat, depending on the
hardness or properties desired and the trans-
formation characteristics of the steel being
processed. The bainite formed is lower bainite,
and its hardness approac.hes the maximum hard-
ness which is possessed by martensite. Since the
temperatures involved in the isothermal trans-
formation of austenite to bainite exceed the
boiling point of water, and the fire point of most
oils, molten salts are generally used as the quench-
ing medium. Low melting-point metals have
sometimes been used.
Molten salt at 400 to 700 F (205 to 370 C) has a
low rate of heat extraction as compared with water
or brine. As a result, the use of salt quenching
should be restricted to small pieces, unless the
transformation characteristics of the steel are such
that relatively slow cooling to the bainite formation
temperature would not produce pearlite.
It should also be noted that for steel of the hot
work type shown in Fig. 5-3a, the bay of the
isothermal transformation diagram (where the
bainite is formed) is such that exceedingly long
times would be necessary for the transformation of
11
austenite to bainite. Austempering of such steels
would, therefore, be impractical.
When it is desired to austemper thicker sections,
and the position of the pearlite nose of the isother-
mal transformation diagram would appear to
make such procedure impossible, several
procedures are available which frequently make it
feasible to harden by this method. It is established
that increased amounts of alloy will force the
pearlite nose to the right, thus allowing more time
in which to cool the piece below the region of
pearlite formation. In other words, the steel will
now have a slower critical cooling rate. In this
procedure, a minor disadvantage is incurred, since
the time required for bainite formation may be in-
creased considerably. Another procedure for
austempering thick sections is to quench (by con-
ventional means) rapidly past the pearlite nose to
some predetermined temperature above Ms and
then proceed with bainite transformation in the
salt bath or a furnace held at the temperature of
bainite formation. Other possibilities are to in-
crease the cooling ability of the salt by increasing
the agitation of the bath, to introduce external
cooling of the salt, and to add small amounts of
moisture to the salt. It must be considered,
however, that moisture in molten salt can result in
violent explosions. Therefore, any addition of
water to the salt should be made only by ex-
perienced and qualified personn.el, using proper
techniques.
Hardening by austempering requires more careful
control than hardening by conventional quenching.
In general, i~ is a more expensive operation and
requires closer supervision. Because of the size
limitations and increased expense, austempering
does not hav~ a large field ofusefulness. Comparing
tempered martensite and bainite of approximately
equal hardness, it is found that the bainite has
greater ductility and toughness than the tempered
martensite in the hardness range of47 to 55 HRC, as
illustrated in Fig. 5-8. Furthermore, since trans-
formation during austempering occurs in a tempera-
ture range where the steel is relatively plastic, and
when the piece is of uniform temperature, the
stresses developed are a great deal lower than in a
conventionally quenched piece ofsimilar chemistry,
size, and shape. As a result, warpage and breakage
by austempering may be considerably lesse~ed
 12
. when compared with conventional quenching. Also,
it is possible to sucessfully and safely harden
intricate sections that are almost impossible to
harden without breakage with conventional quench-
ing. Thus, in many cases, the advantages of aus-
tempering more than outweigh the disadvantages
and make it a very useful hardening procedure.
As indicated in the foregoing, the field of
usefulness of austempering is limited, but it is
widely used in the spring industry and in the manu-
facture of lawn mower blades.
Martempering
Martempering (sometimes called marquenching)
is another interrupted quenching technique, but is
applicable to a much broader field than is austem-
pering.
Manempering of steel consists of: (a) quenching
from the austenitizing temperature into hot oil or
molten salt held at a temperature in the uppermost
.part of or slightly above the martensite formation
range, (b) holding at this temperature until the
temperature throughout the workpiece is sub-
stantially uniform, and then (c) cooling (usually in
air) at a rate slow enough to prevent any great tem-
perature differential among various sections of the
workpiece. The formation of martensite occurs at
a reasonably uniform rate throughout the work-
piece during cooling to room temperature, thereby
avoiding excessiv~ amounts of residual stress com-
monly encountered in conventional quenching.
Martempering definitely is not a replacement
for tempering. The workpieces should always be
Conventional quenching
Time
Fig. 5-9. Time-temperature transformation diagrams with superimposed curves showing quenching and tempering by
(left) conventional process, (center) martempering and (right) modified martempering.
OL-.--------"-....,;;;;;;;;;::c===__iiiaIIII_....
~o 45 50 55 60 65
Hardness· Rockwe! I C
Fig. 5-8. Relationship between impact strength and hard-
ness oj 0.74% C steel that has been conventionally
quenched and tempered and one that has been austem-
pered.
tempered in the conventional manner just as if
they had been directly quenched. Time-
temperature relationships for martempering and
conventional quenching and tempering pre- are
sented schematically in Fig. 5-9 left -and c~nter.
Because the final phase of cooling during mar-
tempering is relatively slow, heavy and light sec-
tions transform from surface to center in about the
same time.
The primary purpose of martempering is to
minimize distortion in the workpieces. When the
hardenability of the steel is adequate, mar-
tempering provides better dimensional control
than does conventional quenching. Also, fewer
harmful residual suesses are developed in mar-
tempering than in conventional quenching,
because the greatest thermal variations occur while
the steel is still austenitic and in a relatively plastic
Modified mortempering
Martensite
Time Time

condition, and because final transformation and
thermal changes occur throughout the workpiece
at approximately the same time. In many in-
stances, martempering has eliminated the need for
quenching fixtures that were required to control
distortion in conventional quenching. Also, in case
of certain unwieldy workpieces, if they are
quenched at a temperature just above Ms (Fig. 5-9,
center), there is time to perform a straightening
operation before any substantial amount of hard-
ening takes place.
Modified martempering differs from the con-
ventional technique only in that the workpiece is
quenched to a lower temperature as shown in Fig.
5-9, right. Faster cooling rates are thus obtained,
permitting application of this technique to steels of
lower hardenability-
Martempering Mediums. Both molten salt and
hot oil may be used for martempering. Oils are
used for martempering up to 400 F (205 C) and
molten salt may be used for the temperature range
of 350 to 700 F (165 to 370 C). Each has some ad-
vantages and disadvantages. As a rule, the molten
{Cool From The Austenitizing Temperature
tEqualize
A 3 - - - - - - - - - -......~-----------
t----AI---------~o4P------------
CD
'-
....
:J
E!
CD
Q.
E
~ .....--M I
.....--M , - - - - - - - - - - - . , - - - - - - - - - . . - - - - -
Zero Time,:
Time---"·
Fig. 5-10. Schematic diagram showing method ofminimizing thermal gradients during quench.
13
salt has more cooling power at temperatures which
are workable for either. For more information on
austempering and martempering see ASM Metals
Handbook, 8th Edition, Vol. 2.
Other Interrupted Quenching Techniqoes
In addition to austempering and martempering,
there are several modifications of these processes
that have been developed for specific hardening
applications. The various techniques for surface
and other localized hardening are covered in
Lesson 9, "Surface Hardening".
Another interrupted quenching technique in-
tended to minimize stresses in the quenched work-
piece is shown schematically in Fig. 5-10. In this
method of hardening, the steel is austenitized in
the normal fashion. Just prior to quenching, the
steel temperature is lowered to and equalized at a
temperature immediately above the upper cooling
critical, A 3 - After the steel has been made uniform
in temperature, it is quenched from the equalizing
temperature by conventional quenching methods.
This procedure has the advantage over a quench
,Quench To Room
Temperature
po- ..-...i_ _ - ~ - - - -
r
I
I
I
I
I
 14
direct from the higher austenitizing temperature of
less drastic temperature differential and volume
change during transformation. As a result, less
stress and breakage are encountered.
This procedure of delay quenching is a com-
promise between conventional quenching and mar-
tempering. It is used where the transformation
characteristics of the steel are such that it cannot
be cooled to the martempering temperature
without some transformation to pearlite, and yet a
level of stress lower than that experienced in con-
ventional quenching is necessary. This technique,
sometimes called the delayed quench procedure, is
sometimes used in conjunction with martempering
or austempering; its advantage is that in quenching
from the lower temperature, the quenching bath
has less heat to extract from the steel and more
rapid cooling is realized. This means that with the
more rapid rates of cool it is possible to austemper
or martemper sections which would othe~se not
harden by these procedures.
~n using this technique, caution must be exer-
cised so that the equalizing temperature does not
fall below A 3 (Fig. 5-10). If the temperature does
fall below this point, some upper transformation
product (ferrite, cementite or pearlite) will result
and the steel will not fully harden. Because of the
tendency of small pieces to cool very rapidly,
especially at the surface, this danger is very real
and can occur within the time the piece is trans-
ferred from the equalizing furnace to the quench.
It is for this reason that the d~layed quench
procedure is used mainly in the processing of large
workpieces where the danger of undercooling is
not so prevalent.
In all of the hardening procedures discussed, the
ideal goal is to form 1000]0 of either martensite or
lower bainite and at the same time avoid other
transformation products such as ferrite, cementite,
pearlite and upper bainite. In conventional hard-
ening', these upper transformation products do,
in many instances, form at some point in the cross
section of large pieces of steel. This is due to
inability to remove heat fast enough to exceed the
critical cooling rate at all depths. When the steel
does not harden to a fully martensitic or bainitic
microstructure because of slow coaling, it is
sometimes described as slack quenched. Formerly,
slack quenching was performed deliberately in the
o 400 800 1200 1600
Temperina Temperafure- OF
Fig. 5-11. Effect of tempering temperature on
elimination ofresidualstresses.
hope that its slower cooling rates would result in
less stress· and consequently less warpage and
breakage. Over a period of time it was found,
however, that the presence of the other trans-
formation products in a hardened piece resulted
-in inferior mechanical properties, particularly
where .plastic deformation and notch toughness
were concerned. The presence of these constituents
in a martensitic matrix is -therefore considered
undesirable and to be held to a minimum wherever
possible. The presence of some lower bainite is not
usually objectionable when mixed with martensite,
and in some cases may even be desirable.
Tempering
When steel is hardened, varying amounts of
residual stress are induced, depending on the
method used to harden. The level of residual stress
can become so high that it overcomes the inherent
strength of the steel and causes spontaneous
cracking without the application of any external
load. In cases where the level of residual stress is
below the point where cracking would occur,
distortion o;r warpage can result. Regardless of the
manner in which the workpiece was hardened,
some residual stress will exist. To eliminate this
stress, or to bring it to some safe level, tempering
is used. Tempering is a process of reheating a hard-
ened steel to a temperature below the Acl and
then cooling at any rate desired. As a result of
reheating, the level of residual stress in the hard-
ened steel is reduced, as shown. in Fig. 5-11. The
higher the tempering temperature, and the longer
the· time at temperature, the greater will be the
elimination of stress.
In addition to the elimination of residuai stress,
tempering serves another useful and equally im-
ponant function. When martensite first forms, it
is white, and quite hard and brittle. For most
applications, other than those involving pure abra-
sion, this condition is undesirable because of
the danger of an abrupt brittle type of failure.
When etched, newly-formed martensite develops a
light straw-colored microstructure; when tem-
pered, the etched structure changes to the darker
colored martensite which is slightly softer and at
the same time appreciably more ductile. In this
tempered condition, the martensite is much more
useful. It may therefore be said that the purposes
of tempering are to reduce the level of residual
stress and to improve the ductility of the steel. The
fact that steel when tempered loses some of its ;hard-
ness and strength is often misconstrued to be the
purpose of tempering. Actually, if hardened steel
were sufficiently ductile and tough, it could be
used in the hardest possible condition, and, in
many cases, this combination of high hardness and
high ductility would be a most desirable one.
Figure 5-12 illustrates the effect of tempering tem-
perature upon hardness and toughness.
In consideration of the benefits realized from
tempering, it is a.general rule that all quench hard-
ened steels should be tempered after hardening,
almost regardless of the hardening method used.
As stated earlier, austempering is the exception.
Because of the temperature range in which bainite
is formed (400 to 700 F, or 20S to 370 C) it is tem-
pered automatically as it transforms from the
austenite.
The preferred practice is to temper quench hard-
ened steels at the highest temperature that is com-
patible with retaining the degree of hardness and
strength required for the service application. There
are applications where essentially no reduction of
the initial hardness can be tolerated; examples in-
clude hardened tools made from carbon or low-

700 56
en
600 48~
::s
e.
~
CD
.c
z
§ 500 40 ...
I
o
o
=400
CD
c:
32 LL
I
-0 CD
a-
e 300
J:
24 .2
- ~
G)
c:
.t: 200 16 (:)
c
CD c.
E
100 8 \:J
o
N

0 -----------------.1--------......-.---------...-------------'0
0 400 800 1200
Tempering Temper~ture - OF

Fig. 5-12. Effect oj tempering temperature on hardness and toughness. Based on 0.50% plain carbon steel:l water
quenchedfrom 1525 F (830 C).
 16
alloy steels. This requirement does not alter the fact
that they should still be tempered at least into the
first tempering stage (180 to 400 F, or 82 to 205 C).
This tempers the white martensite and greatly
reduces brittleness without significant loss of hard-
ness. Common practice is to temper such steels at
300 F (150 C), at which tempering temperature
hardness decreases little if any.
In all cases, tempering not only affects stress
and ductility but' also hardness and strength. In
selec1;ing a tempering temperature, it is therefore
often necessary to effect a compromise that will
result in an acceptable amount of stress removal,
accompanied by a similar increase in ductility,
without lowering the hardness and strength of the
steel below a useful level.
Tempering operations are discussed in Lesson 8,
"Tempering"
Difficulties Associated With
Hardening of Steel
. Decarburization, scaling, excessive residual
stresses and quench cracks are all very real and
must be dealt with in quench hardening of steel.
Decarbnrization and Scaling
Depending on their composition, steels are
heated to 1400 F (760 C) or higher for austenitizing
(often much-higher). When steels are heated to the
austenitizing range without some sort of protec-
tion, they are susceptible to both decarburizing
and scaling (oxidizing). These two conditions may
occur separately, but more often they occur
simultaneously. As temperature increases, the
rates of both decarburizing and scaling increase
exponentially. There can be a number of reactions
between hot steel and the surrounding atmosphere.
A common reaction which results in decarburizing
is:
In the above reaction, the carbon in the steel sur-
face reacts with water vapor in the atmosphere and
is actually leached out of the steel, forming carbon
monoxide.
The following reaction is one of those that
results in scaling or oxidizing.
Fe + CO2 ....peO + CO
In the above reaction, carbon dioxide react'
with the iron forming iron oxide (scale) and carbcm
monoxide. Thus, decarburizing and scaling are the
result of entirely different reactions. However, the
means of preventing either are similar, if not the
same.
Preventing Decarburization and Scale
There are several means of preventing, or other-
wise dealing with decarburization and scaling, the
most common are as follows:
1. Allow stock on the parts for removing the
scale and decarburized layers by machining
or grinding after hardening.
2. Heat parts in boxes packed with a protective
solid material.
3. Wrap the parts with stainless steel foil.
4. Coat with a suitable material.
5. Heat in a protective environment such as
molten salt, molten lead or gas.
In many instances, decarburization and scaling
proceed simultaneously so that, although there is ~
loss of metal from formation of oxide, the carbol..
content at the surface under the oxide is not decar-
burized so that it can be hardened without having a
soft surface. Under these conditions, it is possible
that acceptable surfaces can be obtained after
quenching by mechanically re;moving the scale,
such as by grit blasting.
In other instances, heating in fuel-fired furnaces
where the workpieces are directly exposed to
products of combustion, scale may be almost
nonexistent while decarburization might be severe
and the parts would fail to show the expected hard-
ness on their surfaces. Therefore, it is essential to
consider the conditions of the heating environment
.before deciding on procedures for dealing with
decarburization and scaling. The five methods
listed above are discussed briefly in the following
paragraphs.
Allowance of stock for removal by machining or
grinding is common practice and is a positive
method of removing with scale and decar-
burization. The economics must be considered to
determine whether it is cheaper to prevent the UI

Fig. 5-13. Large f20-in. diQl17eter) hardened steel roll that broke SpoJ1/(J/1eou.f.;Jy becuuse (~f hil!.h internal residual
stress. IVo external STress had been applied to the roll.
desirable conditions or remove them. One distinct
disadvantage of this procedure is that heavy
coatings of oxide may inhibit quenching to the ex-
tent that full hardness is not achieved.
Packing the parts in boxes with cast iron chips or
spent carburizing compound is an effective
procedure for preventing both scale and decar-
burization. However, it is usually inconvenient to
remove parts for quenching, and in addition, it is
very inefficient because of having to heat the boxes
and the protective compound.
Coating the parts with an aqueous solution of
borax, or with anyone of several proprietary com-
pounds is a reasonably effective method of prevent-
ing scale and decarburization. The main disad-
vantages are the cost of applying the material, and
the cost and difficulty of removing the coating af-
ter quenching.
Wrapping in Type 310 stainless steel foil has
proved to be an effective method for protecting the
workpieces during austenitizing. However, this
method would generally be too expensive for
production hardening. Single tools are often wrap-
ped for hardening when an atmosphere controlled
furnace is not available.
Heating in molten salt is widely used for
austenitizing. A properly controlled salt bath is
capable of heating parts without decarburization
or scaling. Not only does the molten salt provide
full protection during heating,. but the thin film of
salt that adheres to the workpieces provides
protection while being transferred from
austenitizing to quenching. Molten lead baths have
been used to provide protection during
austenitizing, but their use has gradually decreased
in favor of molten salt.
17
Gaseous atmospheres (including vacuum) are
also widely used for protection of workpieces
during austenitizing. Purified exothermic at-
mospheres, lean endothermic atmospheres and
nitrogen-base atmospheres are those most often
used in prevention of decarburization and scaling
of carbon and alloy steels. However, not all have
maximum suitability for all grades of steel. For
example, exothermic atmospheres are not suited
for heating high-carbon steels because exothermic
atmospheres will decarburize high-carbon steels.
The use of vacuum furnaces for austenitizing
various grades of steel is increasing in popularity.
It is an effective means of preventing both decar-
burization and scaling.
For additional discussion on protecting the steel
during austenitizing see Lesson 11 , ~ 'Protection
During Heat Treatment Against Decarburization
and Scaling" .
Residual Stress
Probably the most serious and no doubt the
most spectacular defects which can occur in hard-
ening are those due to the build-up of residual
stress: breakage and warpage. This generally oc-
curs spontaneously, without the introduction of
any external applied stress, during the hardening
operation or shonly thereafter. Figure 5-13
illustrates a large hardened steel roll that broke
because of high internal stress. Internal residual
stress in a hardened object is the result of (a) non-
uniform thermal contraction during cooling, (b)
expansion or contraction during transformation,
or (c) a combination of the two. The simplest
mechanism of residual stress build-up is that
resulting from thermal contraction alone.
 18

11
.--Center (6 From Surface)

~ 1400
I
e
~ 130
c
~
CD
~ 1200
~
1100

100

900'L..---~--"'--","",,------'--_----I'-----~-~------
a 2 4 6 8 10 12 14 16
Time - Minutes
Fig. 5-14. Variation in cooling rate oj different points in the cross section of a 12-in. diameter bar quenched from
i550 F (854 C) into brine. . .

The transfer· of heat through steel is relatively
slow, as may be demonstrated by a simple home
experiment. For example, hold a needle or a pin by
one end and apply the heat of a lighted match to
the other end; observe the length of time required
for the heat to travel from the hot end of the needle
to the end held in the fingers. (Repeating the ex-
periment with a piece of copper wire will show a
surprising difference.) Sluggishness of heat trans-
fer in steel occurs whether one is heating or
cooling. Figure 5-14 illustrates the variation in
cooling rate experienced when a 12-inch round slug
is brine-quenched from 1550 F (845 C). It should
be noted that at the center of the slug the cooling
rate is so slow that at about 1330 F (705 C), the
eutectoid reaction takes place with the formation
of pearlite. The latent heat of transformation
causes a plateau in the cooling curve. The rates of
cool 1 inch and 3 inches from the surface are ap-
parently rapid enough to prevent the formation of
pearlite, at least in quantities sufficient to affect
the slope of the cooling curve. This appreciable
difference in cooling rate is experienced whether
the steel is quenched from above or below the
critical temperature and is due to the relatively low
thermal conductivity. Of course, when cooling
from below the Acl no transformation occurs, and
the cooling curve representing the center of the
piece would not be interrupted. It should also be
noted that as steel is heated, it expands at a
predictable rate proponional to the increase in
temperature, and as it cools it contracts in similar
fashion.
The effect that uneven distribution of tem-
perature and volume change may have upon the in-
ternal stress condition of quenched steel is best
illustrated by considering what occurs when a steel
cylinder is heated to 1200 F (650 C), which is below
the critical temperature, and then quenched. The
.microstructure of such a piece will consist of
ferrite and cementite with no austenite being in-
volved; therefore, no transformation need be con-
sidered. In such a steel, the only changes which oc-

cur wiIl be thermal changes due to heating and
quenchIng (no heat of transformation), and
volume changes due to heating and cooling (no
transformation volume changes). If the cylinder is
slowly and uniformly heated to 1200 F (650 C), so
that there is no thermal lag, all parts of the piece
will expand uniformly and in unison so no stress or
distortion will be incurred.
When such a piece is quenched from 1200 F
(650 C), a non-uniform temperature distribution
will be set up throughout the cross section of the
cylinder. In the first instants of the quench, the
outer layers of the cylinder will cool and try to con-
tract. Being attached rigidly to the core of the cyl-
inder, which is still hot and in expanded condition,
the other shell is mechanically prevented from con-
tracting. As a result, the outer shell is forced to re-
main elongated and is in tension. The core Q.f the cyl-
inder is placed in compression since it is opposing
the shrinkage in the outer shell. Thus, the outer
shell is elongated slightly and the core compressed
slightly; since the shell is cooled to a low tempera-
ture, this slight elongation is permanent in nature
on account of the rigidity of the steel at the lower
temperature. As the core subsequently cools, it
wants to contract and become shoner, but at this
stage the shell is no longer subject to thermal con-
traction and has to remain the length it has already
assumed. The shell will then resist·the contraction
of the core and will be placed in compression,
while the core will be in a state of tension. The net
L ·~:':':" .1
Fig.5-15. Effect of repeatedly quenching Q steel cylin-
der from be/ow the critical temperature; (a) before
quench and (b) after quench. (Sketch has been greatly
exaggerated.)
19
result of this opposing stress distribution will be a
distortion of the cylinder as shown in Fig. 5-15.
This stress (and accompanying distonion) is due
entirely to temperature and volume differential
and does not involve transformation. Stresses
beyond 80,000 psi have been reported resulting
from quenching steel cylinders from below the
critical temperature, so the condition described is
obviously both real and appreciable.
The second source of internal residual stress lies
in the expansion which occurs when -austenite
transforms. Iron, the basic constituent of steel, is
allotropic in nature; that is, it can possess more
than one crystalline form. The classic example of
allotropy is carbon, which can exist as a diamond,
as graphite, or as lampblack. At room tem-
perature, the crystalline form of iron is body-
centered cubic, with the atoms of iron arranged as
in Fig. 5-16a. This form is called alpha iron.
When the iron is heated and the ~l is reached,
the crystal structure is rearranged into the face-
centered cubic structure, with the iron atoms
arranged as in Fig. 5-16c. This structure is called
gamma iron and forms the basis for austenite,
which is a solid solution of carbon in gamma iron.
The face-centered cubic arrangement is more dense~
ly packed. This explains why contraction is noted
in the dilatometer curve when austenite begins' to
form. When the austenite is cooled, the reverse oc-
curs with a slight expansion noted at the stage
where transformation begins.
If the gamma iron contains carbon in solution,
and the austenite is rapidly quenched so that mar-
tensite forms, an entirely different crystal struc-
ture will be observed. The crystal form of mar-
tensite is body-centered tetragonal, in as-quenched
condition, with the atoms arranged as in Fig. 5-16b.
This crystal structure creates a real and measurable
volume increase, depending upon the carbon con-
tent of the steel, as shown in Fig. 5-17. When mar-
tensite is tempered, the crystal structure again
changes to body-centered cubic which explains
why the tempered martensitic structure is more
stress-free than the untempered. The volume
change caused by the formation of manensite (and
other transformation products) is therefore the
second source of internal residual stress developed
during hardening.
The total residual stress set up in a quenched
piece is the sum of stresses from thermal con-
 ·20

I : ~
C,..,_.I e.';··•.•
...

p"-;..-
" 6
bee bet fcc
a b c
Fig. 5-16. Three possible crystal structures ofsteel. (aj Bcc, body-centered structure, characteristic offerrite,· (b) bet,
body-centered tetragonal structure, characteristic of martensite; and (c) fcc, face-centered cubic, characteristic of
austenite.

traction and stresses from transformation expan- ening. The first consideration is the chemical
sion. The nature and distribution of the stress can composition of the steel. Other conditions being
become very complicated even in a simple shape, equal, it is true that as the carbon content of the
.because of the widely varying conditions of tem- steel increases, the tendency to crack also in-
perature distribution and transformation creases. This relationship is probably due to the ef-
throughout the cross section during quenching. In fect of carbon on the axial ratio of the tetragonal
fact, during the quench itself, the nature of the martensite, which causes more drastic volume
stress at anyone particular spot may reverse itself. changes when the higher carbon steels are hard-
When a piece has finally cooled to room tem- ened. In other words, martensite appears to be
perature and all transformation is complete, it is more voluminous as the carbon content goes up.
generally the case that for a piece which has hard- Carbon also has a strong effect upon depressing the
ened throughout its cross section, the surface Ms and Mf temperatures. If the Ms temperature is
fibres will be in tension and those in the core in
compression. In a piece which has not hardened
throughout the cross section, the surface fibres !
will be in compression and those in the core in ten- i
sion. While this is true of solid cylinders, it may I.°ar
not be for bored cylinders or for pieces of non- i
LOsr
uniform cross section or co'mplicated design.
Hence, each case must be considered on its own I
.~ J.04~
merits; the stress pattern, or level of stress, may be c I
0:: i
very complicated and not predictable or calcul-
010 I
able by any ordinary means. The only reliable 1.02,
method of determining stress in a hardened piece is ,
I

to measure it by the various experimental tech-

!
niques of stress analysis. i
o.98'--.......-.--------..I..-........ -'O'-"-------"""---.J

Quench Cracking o 0.4 0.8 1.2 1.6

Corbon Content-%
Numerous factors have an influence on crack- Fil!. 5-17. £.[fecl (~f carbo1l Oil the letrago/lulir... · (~r l1urr-
ing, which is an ever-present danger during hard- tensile.

lowered sufficiently, the transformation from
austenite to martensite will occur in a colder steel
where relatively little plasticity exists. Therefore,
transformation volume change will take place
when the steel is less able to adjust plastically, and
cracking results.
The relationship between Ms and cracking is
quite real in steels which harden through the for-
mation of martensite, and the lower the Ms , the
greater the tendency to crack. Therefore, any fac-
tor which results in a lowering of the Ms will result
in greater tendency to crack when using the same
quenching technique. In addition to carbon,
numerous other elements have a depressing effect
upon M s ' as does increasing the austenitizing tem-
perature. The influence of carbon content and
austenitizing temperature upon depressi.on of the
Ms is illustrated in Fig. 5-18.
Increased austenitizing temperature is a source
of quench-cracking, not only because of its effect
10. 95% cl
~ 400
I
CD
~
~
+-
~
Q)
~ 300
~
200
IOO~---------------""-~-----.J
1300 1500 1700 0.5
Austenitizing Temperature
Fig. 5-18. Effect ojaustenitizing 'temperature and carbon content on the M s temperature. Curves are/rom A1S1 8695
type szeel:l· composition: 0.95 C, 0.82 Mn~ 0.23 Si, 0.56 Ni, 0.52 Cr and O. 19 Mo.
21
upon Ms but also because at the higher
austenitizing temperatures greater thermal
gradients exist in the cross section of the piece
during the quench. Furthermore, at the higher
temperature the austenite has a greater solubility
for carbon; more of the carbon is taken into
solution in the austenite and less remains as
mechanically entrained cementite. This has a direct
effect upon the volume changes which occur when
austenite transforms to martensite.
Another factor influencing the susceptibility of
steel to quench-cracking is grain size. When steels
of similar chemical composition but varying grain
size are heat-treated under identical conditions, the
steel with the coarser grain size has an appreciably
greater amount of residual stress than the fine-
grained steel, as shown in Fig. 5-19. With the finer-
grained steel, it is believed that the larger number
of differently oriented grains per unit volume is
more favorable for the mutual balancing of any
Austenitized at
1600°F
0.6 0.7 0.8 0.9 1.0
Carbon Content- %
 22
_.... -..
.. - --_-._"
fi~~\~~%~~~~-_/- '1
~

r~~II~II~i~~J~fir~1- -
I .
16
12
I
I
I
f
I
I
I
,
J
J
: ! I f
,
~
.I 8 I
I
I
f

~
I I
!
I I ~ J :
I J
I r
I
o :
I
Center I
~ I·
...--.--- Surface ------. I
8 - - Surface _~ t

(0) (b)
Grain Size 8 Grain Size 2 - 3
(128 Grains /Sq. In. -IOOX) (2-4 Grains /Sq.ln. -IOOX)
Fig. 5-19. Comparison of the level of stress distribution in two steels of similar chemical composilion but different
grain size, after identical heal treatn7ents. Internal stresses are considerably greater in the coarse grain steel.

stress developed. It is also believed that the greater
number of boundary layers of grains in the fine
grained steel is more favorable for plastic defor-
mation. Grain size has another influence of a more
direct nature. It has been demonstrated that in
martensite there exist very minute cracks, and as
the grain size is increased the size of the micro-
cracks also increases, as in Fig. 5-20. It is, there-
fore, reasonable to expect that if such cracks
existed, they would act as points of weakness
where the stress would concentrate and so create a
condition favorable to cracking. The larger the
grain size and the larger the microcrack, the
greater would be the tendency of a piece to crack
with any given concentration of stress.
The method of hardening may have a very pro-
nounced effect upon the tendency of a hard-
ened piece to crack. In general, conventional
quenching procedures, where the hot steel is
plunged directly into a cool quenching medium,
are more liable to result in cracking than
at a very uniform temperature throughout the
cross section when transformation takes place, so
that stresses resulting from non-uniform tem-
perature distribution and accompanying trans-
formation are minimized. Furthermore, since
transformation occurs at a slightly elevated tem-
perature, the steel retains a ce~tain degree of
plasticity and can more readily· adjust itself to any
stress which is developed. The austempered or
martempered steel can also temper itself to a
limited but effective degree.
GrOIns Per SQ Inch -IOOX
6~ 32 16 S 4 2
;

j
procedures utilizing isothermal transformation
types of hardening. Quenching into the more ef-
O......_.....I.._ _"--_....r-_ _L . - _......- - l I
fective cooling media, such as water and brine, Po 7 6 5 .: 2
also results in more cracking than quenching into a Grein S:ze
less effective medium such as oil. In austempeFing.. Fig. 5-20. Relationship between size of microcrllcks in
and to a lesser extent in martempering, the steel is martensite and grain size.
 .23
Table 5-4. Effect of Various Quenching Conditions on Quench-Crack Sensitivity of Alloy Steels

Quenched from - 1850°F 1650 1475 1850 1650 1475 1850

Quenching Medium - Oil Oil Oil Salt Salt Salt Salt
Bath Temperature - 80~ 80 80 300 300 300 400
4320 a 0 0 000 0
8730 00 0 0
4032 00 0
5130 00 0
1030
1330 i
00
xo*
, 0
0
4130 xo 0
4037 x I 00 00 0 0

I
I
5140
4053 x
I

II x
xx 00
0
I
0
0
0

8742 I xx xxoooo 00 0
I
4140 x .xxx0 xxxxo 0000 00 0
1050 xxxx x.xxx 0000 00 j
5150
- x xx 00 00 0 0
I
1090 x xx 000 00
4063 x xx 00 0
1060 x xx xoo oc
I 4340 x I

I
xx xx xxx 0 xoo 00
1345 xx xx xx 00 000
5160 I x x xx ~ 00 0 0
4150 x x :xx xxo xooo 0000
8750 x x x xx xxo 00000

o = Not cracked. x = Cracked.

·xo indicates that two tests were made; one specimen cracked, the other did not.

The effect of quenching practice and trans-
formation temperature upon the tendency to
crack is illustrated in Table 5-4. Here, a number of
different steels were tested for cracking tendency,
using· ~ specially designed test specimen and
varioushardening procedures. The effects of tem-
perature, carbon content, alloy content and quench-
ing conditions are clearly illustrated.
Precipitation Hardening
In all of the hardening procedures considered
thus far, the principles have revolved around the
transformation of austenite. It was noted that the
degree of maximum attainable hardness depended
on the carbon content of the steel, and the lower the
carbon content the lower the hardness obtained.
Transformation characteristics of the simple low-
carbon steels are such that the formation of mar-
tensite is possible only with extremely rapid rates
of cool and in very thin sections: Low carbon steels
are therefore relatively difficult to harden by
quenching, especially in larger sections.
Cenain low-carbon steels can, however, be
hardened to some degree by the mechanism known
as precipitation hardening.
If a very low-carbon steel is quenched from
about 1325 F (720 C) instead of from above the
~, and then tempered for 10 hours at 140 F
(60 C), it will be found that the hardness has in-
creased from about 70 HRB in the quenched condi-
tion to 81 HRB after tempering. If, instead oftem-
pering, the quenched steel is allowed to rest at room
temperature for a period of time longer than 10
hours, the hardness will also show a similar in-
crease. This increase is called precipitation harden-
 ·24
ing, age hardening, or simply aging. As steel be-
comes harder and stronger, the ductile properties
generally decrease. In higher carbon steels, a simi-
lar change is noted, although to a lesser degree. The
effects of aging may still be plainly detected in
steels of less than 0.15070 carbon content but are
less noticeable, if at all, in steels of higher carbon
content.
This characteristic may be desirable, or grossly
undesirable, depending on processing and end
requirements. For example, in one shop 8-in. wide
1008 steel coil stock was being deep drawn into
cup-shaped parts in a mechanical press. The steel
showed a hardness of about 6S HRB as it was
being processed. No difficulty was encountered
in making the cups in a single draw. For lack of or-
ders, the job was temporarily stopped and the steel
was stored for over six months, after which the job
was resumed. Using the same press and the same
dies the drawn parts fractured during drawing,
resulting in 100070 scrap so that the job was stop-
ped. Investigation showed that the steel had in-
creased in hardness during storage from 65 HRB
to 75 ·HRB and therefore had to be annealed be-
fore it could be used for deep drawing.
Effect of Carbon
The ability of ferrite to dissolve carbon varies
with temperature. At a temperature of 1341 F (727
C), (AI)' ferrite can hold about 0.022070 carbon in
solution, but only about 0.008070 carbon can be
retained in solution at room temperature. This is
shown in Fig. 5-21. If a low carbon steel is heated
to just below the Ael' the ferrite will take into
solution about 0.022070 carbon and will retain it in
solution if the steel is quenched rapidly. This
results in the ferrite remaining relatively soft.
Since ferrite at room temperature would normally
hold only 0.0080'}'0 carbon in solution, the presence
of 0.022070 carbon means that the quenched ferrite
is super saturated and therefore in an unstable
condition at room temperature. (This condition is
brought about by the quench, which caused the
excess of carbon to be trapped in solution by the
sudden cooling.)

~---Ae,---------------...-f
LL
o
I
CD

...
'-
::s
o
'-
Q)
Q.
Alpha + Cementite (Fe3C)
E
~

0·.02 004 006 008 0.10

Carbon Content - %
Fig. 5-21. Variation in solubility of cementite (or ofcarbon) in iron below the A e 1 temperature.
 al 84
--~ 80
~
(,)
o
a: 76
I
U)
U)
CD 72
C
~
c
::I:
5 10 30 I 5 10 50100 500
t~Minutes 'I Hours------....
Time At Aging Temperature- (Logarithmic Time Scale)
Fig. 5-22. Effect of time and temperature on precipitation hardening oj 0.06% C steel after quenching from 1325 F
(720 C).
I f the steel is heated slightly, it will go to the aries, or certain crystallographic planes of the
equilibrium condition and the excess carbon will grains. This condition is not so effective for in-
precipitate out as cementite. When this occurs, the creasing hardness, and further, it tends to form
cementite particles come out of solution as fine, planes of brittleness and weakness.
submicroscopic particles. They cause lattice distor-
04 CU
tion. This, in turn, results in numerous localized
3 4 5 6
strains interfering with slip, which induces an in-
crease in hardness. At room temperature the tiny
precipitate remains as well-distributed and in-
1000 t----+--~---+---__+_-__#_+---~
dividual panicles. At slightly elevated tem-
peratures, they tend to coalesce to form spheroids. IY-Fe
I
They do this in exactly the same manner as the
larger carbide particles during spheroidize an-
nealing treatments in tne vicinity of 1200 to 1300 F
900 ....--.~--....--~--~------....------I
(650 to 705 C) and.a drop in hardness is also ex- U
o
perienced. Because of their extremely small size
850C
and limited quantity, the carbide particles spheroi-
dize and soften 'the steel at much lower tem-
peratures than do the larger carbides encountered
800 ....----+-....-.#--~----+---------+----...
in the conventional spheroidize anneal. The effect 1
of temperature upon precipitation and hardness is a-Fe+Cu
shown in Fig. 5-22. I

If the low carbon steel is not quenched from the

region just below the Acl but is slow cooled in- 2 :3 4 5 6
stead, cementite particles will precipitate out. Un- w4 Cu
der these conditions, they wiU not be finely Fig. 5-23. Partial iron-copper phase diagram showing
distributed throughout the ferrit grains but will the solubilizy of copper in a.-Fe (body cenzered cubic)
tend to form and concentrate in the grain bound- at temperatures below 1670 F (910 C).
 -26
Effeq of Other Elements
In addition to carbon, several other elements
can be·taken- into solution and precipitated out in
similar fashion. Copper is a good example of the
limited solubility of elements other than carbon in
alpha iron. At the eutectoid temperature (1534 F
or 851 C for the eu-Fe system), alpha iron can
In addition to copper, numerous other elements
.can cause precipitation hardening. These include
aluminum, cobalt, molybdenum, nickel, and
tungsten. The elements may be used individually
or in combination. A typical precipitation hard-
ening steel is one containing 0.25070 C, 0.60070
Mn, 0.30070 Si, 3.5010 Ni, 1.2O'J0 Cr, 0.25070 Mo and
~ .1010 AI. This grade shows strong precipitation

dissolve about 2.1 0'/0 copper, but only about 0.3070
at room temperature. The solubility of copper in
alpha iron, and the effect of temperature is shown
in Fig: 5-23. When a copper-iron alloy is quenched
from just below the euteetoid temperature, a
supersaturated solid solution results, just as with
plain carbon steel. Tempering the solution results
in some of the copper precipitating out and in-
creasing the hardness and strength. Furthermore,
if iron is alloyed with carbon and copper, a cop-
per-bearing steel results which can be dual hard-
ening; that is, from bainite or martensite for-
mation, as well as from the precipitation hard-
ening effects of the copper-rich phase. Figure 5-
24 shows the influence of precipitation on the
jlardness of copper steel.
hardening characteristics in the unhardened state.
The annealed hardness can be readily increased
from about 207 HB to 300 to 400 HB by tempering
several hours in the vicinity of 1100 F (595 C).
The precipitation hardening alloys are very
numerous and in some cases quite complex, so that
complete and detailed consideration would be out
of the question here. Their increase in hardness
and strength through precipitation can be extreme-
ly great, as indicated in Fig. 5-25 which shows
precipitation hardening in an alloy made up of
75070 iron and 250/0 tungsten.
The precipitation hardening (600 series) grades
of stainless steels can be precipitation hardened to
values approaching 50 HRC. The hardening

200....--------,.---...----....------..-------

180

CD
c:
di 14

120'-----......-----~---J.------L.-----Jl....------I
o 025 I 4 I~ 64 4096
Time At Aging Temperature - Hours
Fig.5-24. Precipitation hardening oj a 1.6 Cu-O.06 C steel after quenching from 1500 F (815 C) and aging at times
and temperatures indicated. Maximum hardness is achieved sooner at higher temperatures.

procedures for these steels are covered in a sub-
sequent Lesson in this course; "Constitution and
Heat Treatment of Stainless Steels" .
Advantages ·of Precipitation Hardening
When hardening heavy sections by quenching, a
large difference in cooling rate exists between the
outside and the center, the center having the slower
rate of cool. As a result, the austenite in the center
of the piece may not always transform to mar-
tensite but may transform to one of the softer and
weaker microstructures. The center of large hard-
ened sections is therefore softer and weaker than
the surface in many instances.
When a precipitation hardening steel is quenched
from below the eutectoid temperature, the super-
saturated condition of the ferrite can be achieved
with relatively slow rates of cooling by reason of
the sluggishness of the reaction. Subsequent heat-
ing and holding will cause precipitation and hard-
ening. It is frequently possible to harden and
27
strengthen the center of large forgings by
precipitation hardening whereas such would not be
possible by the transformation mechanism. A fur-
ther advantage of precipitation hardening is the
fact that it occurs uniformly and gradually, thus
minimizing residual stresses within the workpiece.
This is due to the fact that there are no drastic ther-
mal or transformation volume changes. However,
the hardness that can be achieved through
precipitation in steels of medium or high carbon
content is not as high as that which can be attained
with transformation hardening, but in DliUlY in-
stances where transformation hardening· is not
possible, precipitation hardening is the only way.
Normalizing
Normalizing, which is defined as heating to
some temperature (usually about 100 F, or 55 C)
above the upper critical <.t\:3) and cooling in air, it
is often used as a means of increasing strength of
large carbon or alloy steel forgings and castings. In

65 1110 0 F

Q)
~
~
(.)
o
a::
U)

~
c: 45
-C
~
C
:r:

a 20 40 60 80 100
Time At Aging Temperature - Hours
Fig. 5-25. Precipitatio.~ hardening of a 75 Fe-25 Walloy a.fler quenchinp, from 2730 F (1500 C) and rehearin?, for
'Vllrying time intervills between 1070 and 1470 F (575 and 800 C).
 28
such instances the normalizing operation is usually
followed by tempering. In the instance of air hard-
ening steels, normalizing and hardening by trans-
formation are achieved by the same procedure.
Procedures for normalizing are covered in detail
in a previous Lesson in this course; "Annealing
and Normalizing of Steel"
Furnaces for Hardening
Selection of a furnace. for hardening depends
upon a long list of factors which include the
following:
1. Size and shape of the workpiece.
2. Composition of the workpiece.
3. Required operating temperature.
4. Number of identical workpieces to be
processed.
s. Type of protection needed during hardening.
Because of the many factors involved almost
every conceivable type and size, of furnace has
been used for hardening of steel. Details of fur-
naces for hardening ate covered in a subsequent
lesson in this course; "Heat Treating Furnaces"
 ASM IntemationaP
MATERIALS ENGINEERING INSTITUTE
C10LS,C11L13,C40L4,C41L6
Test Series G Hardening of Steel
Page 1 of2

PLEASE PRINT YOUR NAME AND ADDRESS BELOW RETURN TO: ASM Intsmationafi)
MATERIALS PARK, OH 44D73~02
NAME REGISTRAnONNO. _

STREET _

ClTY STATE ZlP DATE _

MULTIPLE CHOICE

C, D) in the box. (Only one answer is correct and

Place the appropriate letter (A" B,
each question has a value of10 points.)

1. The crystal structure of as-quenched martensite is,known as:

(A)
(B)
body-centered tetragonal.
face-centered cubic.
o
(C) mixed structure.
(D) body-centered cubic.

2 If the severity (H-value) ofquenching using mildly agitated plain water is 1.0-1.1,
what severity can be expected for brine, also mildly agitated?

(A) 2.0-2.2
o
(B) 4.0-5.0
(C) 1.4-1.5
(D) 0.8-1.1

3. As the carbon content of a steel increases, the Ms temperature:

(A) remains about the same.

o
(B) increases.
(C) decreases.
(D) decreases to a carbon content of 0.80%, then increases.

4. The isothermal heat treating process used for transforming austenite to bainite is
called:

(A) martempering.
o
(B) austempering.
(C) equalizing.
(D) equilibrium hardening.
 ASM Intemationalf}
MATERIALS ENGINEERING INSTITUTE
C10LS,C11L13,C40L4,C41L6
Test Series G Hardening of Steel
Page2of2 TIle Materials
Inf.aratioa Society

5. If a very-Iow-carbon steelis quenched from 720°C (132S°F), then tempered at 60°C

(1400f) for 10 hours, its hardness will: D
(A) increase by 10 or 11 HRB.
(B) decrease slightly.
(C) remain the same.
(0) increase by at least 25 HRB.

6. As a rule, the primary purpose of using manempering is to:

(A) eljminate the need for tempering.

D
(B) conserve on consumption of the quenching medium.
(C) minimize distortion ofthe workpiece.
(D) increase the maximum attainable hardness.

7. In hardening hypoeuteetoidcarbon steels,the austenitizing temperature decrease as:

(A) carbon content decreases.

D
(B) grain size decreases.
(C) carbon content increases.
(D) section thickness increases.

is caIled:
8. The rate of cooling necessary to ensure fun hardening of a specific steel

(A) the time required for it to reach the temperature of the quenching medium.
D
(B) the time required for it to reach the Ms temperature.
(C) the rate required to cool a 25.4 mm (1 in.) diameter to room temperature.
(D) its critical cooling rate.

~ 9. Whenasteel does notharden to a fully martensitic orbainitic structure because it was

cooled too slowly, it is said to have been: D
(A) tempered
(B) slack quenched..
(C) subjected to modified martempering.
(0) austempered

10. Reheating of a hardened steel below AC1 and cooling at a desired rate is known as:

(A) full annealing.

D
(B) modified martempering
(C) tempering.
(D) equalizing.
 ASM Intemational~ is a Society whose
mission is to gather, process, and
disseminate technical information.
ASM fosters the understanding and
application of engineered materials and
The Materials their research, design, reliable manu-
Intonnatlon Society
facture, use, and economic and social
benefits. This is accomplished via a
unique global information-sharing
network of interaction among members
in forums and meetings, education
programs, and through publications
and electronic media.

Printed on DocuTech February 1994

 The Materials
Information Society

Tempering of Steel
by
Howard E. Boyer

Course 10
Lesson, Test 7
Course 40
Lesson, Test 5
Course 41
Lesson, Test 9
 Metric Conversion Factors
To convert from To Multiply by

in. mm 25.4
in. m 25.4 x 100s
mil ~ 25.4
~. J.lDl 25.4 X 100s
• 2
m. m2 6.45 x 1()-4
• 3
m. m3 1.64 x 10-S
ft. m 3.048 x 10-1
ft.2 m2 9.29 x 10-2
ft.3 m3 2.831 x 10-2
oz. g 2.834 x 101
lb. koc 4.536 x 10-1
Btu J 1.054 x l(Y
Btullb. - of Jlkg· K 4.18 x 103
Btu/ft. • hr • of W/m-K 1.730
10inJinJoF 1()"'6 mmlmrnJOC 1.8
psi Pa 6.895 x l(Y
psi kPa 6.895
ksi kPa 6.895 x 103
ksi MPa 6.895
ksi - in. 1I2 MPa e m1l2 1.099
ksi;/in. MPa·;/m 1.099
oz.f gf 28.4
lbf kgf 4.536 x 10-1
lbf N 4.448
lbf· ft. N· m (or]) 1.356
lbf/in. 2 kgflcm2 14.223
lbflin. s kgflm3 2.768 x lO'
Ib.lft. 3 kg/ms 16.019
IbJm. 3 glcms 2.768 x 101
IbJin. 3 kg/ms 2.768 x lO'
gal (U.S. liquid) L 3.785
gal (U.S. liquid) m3 3.785 x 10-3
lb.lgal gIL 119.826
ft./gal mlIL 748
of °C (OF - 32)/1.8
of K (OF + 459.67)/1.8
°C of (OC - 1.8) + 32
°C K °C + 273.15
K °C °C-273.15

Multiple and Submultiple Units

1018 •••••••••••••••••••••• exa (E) 10-1 ••••••••••••••••••••• deci (d)
1015 ••••••••••••••••••••• peta (P) 10-2 centi (C)
•••••••••••••••••••••
1012 tera (T) 10-3 ••••••••••••••••••• milli (m)

109 ••••••••••••••••••••• giga (G) 1Q-6 •.•...••••..••••••• micro (J.L)

1()6 ••••••••••••••••••• mega (M) lQ-9 •••••••••••••••••••• nano (n)
1()3 ••••••••••••••••••••••• ki1o (Ie) 10-12 ••••••••••••••••••••pico (P)
102 hecto (h) 10-1s femto (f)
101 ••••••••••••••••••• deka (da) 10-18 ••••••••••••••••••••• atto (a)

Abbreviations
J ..........••.............joule m meter Pa ••••••••••••••••••••• pascal
kgf kilogram force mm millimeter K •.••...•...........•.. kelvin
L liter N newton W watt

Editor
Philip D. Harvey
Metals Engineering Institute, American Society for Metals

e 1980American Society for Metals

This material may not be reproduced in whole ex in part in any form whatsoever. FuDy protected by copyright.
 Tempering of Steel
Tempering of steel consists of heating previously
hardened or normalized steel to a temperature
below the lower transformation range (AI)' and
cooling at a suitable rate, primarily to obtain the
desired mechanical and metallurgical properties.
As a rule, the cooling rate is not critical.
Tempering is always done at some temperature
between ambient and AI' but heating of steel within
that. range does not always produce "tempering."
This same temperature range often is used to re-
lieve stresses imparted to steel as a result of cold
working, process (subcritical) annealing, or pre-
cipitation hardening of some grades of steels; most
notably in certain types of stainless steels. Again,
.it is emphasized that the term tempering applies
only to steels that have been hardened or normal-
ized. The term drawing sometimes is referred to as
the process of heating hardened steels below AI.
However, drawing as applied to this process is
incorrect and its use should be discontinued.
For plain carbon or very low-alloy steels, as the
tempering temperature is increased the hardness
(and likewise the strength) after tempering de-
creases, and the toughness, in terms of impact
energy, is increased as illustrated in the curves of
Fig. 8-1. Thus, selection of the tempering tempera-
ture for a given steel is based mainly on the desired
combination of mechanical properties (see Selec-
tion ofTempering Temperature later in this lesson).
As alloying elements are added, however, the
property curves do not follow the simple forms
illustrated in Fig. 8-1, as will be discussed in subse-
quent sections of this lesson.
Proceaure for Gathering Data. Hardness versus
tempering temperature data are developed by
following one of two procedures and employing
hardened steel specimens, usually Vi to 1 in. (13
to 25 mm) in diameter:
1. A number of these specimens are placed in a
tempering furnace, and they are all heated at
some predetermined temperature, usually
300 F (150 C), for one hour. One specimen is
then removed from the furnace, cooled to
room temperature, tested for hardness, and
the data recorded. The temperature of the
furnace is then increased to some pre-estab-
lished increment, usually increments of 50 to
100 F (26 to 55 C). Heating is then continued
for another hour at which time another spec-
imen is tested and discarded; and so on for
the entire tempering range. Data are thus
obtained to plot a curve such as shown in
Fig. 8-1.
2. A second and similar procedure is done with
a single specimen; that is, the same specimen
is withdrawn after each hour at the specific
temperature, cooled, tested and then heated
at a higher temperature; thus providing the
data for a hardness-tempering temperature
curve.
 2

700 56
en

'-
600 48 -g::s
CD
.c C£
E 500
::3
40 I
+-
Z 0
en 0
en
CD
400 32 LL
c:
'"'C
I
Q)
'-
c
:::c 300
::J
24
~
CD +-
c:
.i:: 200 16 c(.)
al ~
E
100 8 ~
0
N
O~--~-_---.l._--~-----'--_--I_-_-L- __ 0 .-.I

o 400 800 1200

Tempering Temperature - of
Fig. 8-1. Effect of tempering temperature on hardness and toughness for an 0.50% plain carbon steel, water
quenched from 1525 F (830 C).

The difference in the curves developed by the
two procedures probably is not great, but because
of the time variable in using one specimen or many
specimens, the curves will not be precisely the
same. Likewise, data secured by either procedure
will not be identical compared· with .tempering a
load of workpieces to one piece at one specific
temperature, even though section thicknesses are
similar to those of test specimens.
Metallurgical Changes Caused
by Tempering
When steel is in the fully hardened as-quenched
condition, it generally has limited usefulness be-
cause it is brittle, it is under a condition of high
stress, and also, it is dimensionally unstable. All
hardened steels (medium to high carbon) should be
tempered before they are placed in service. Low
carbon steels sometimes are used in the as-quenched
condition for high yield strength applications.
Case hardened parts often are used in the untem-
pered condition to take advantage of compressive
surface stresses for improved resistance to fatigue
failure.
The importance of tempering to minimize brittle
fracture in service increases as the carbon and/or
alloy content increases, especially the carbon con-
tent. The reason most steels are brittle in the as-
quenched condition relates to the crystal structure.
Three possible crystal structures of steel are
shown in Fig. 8-2. The body-centered cubic crys-
tal, which characterizes ferrite, is shown in Fig.
8-2a. When this crystal is heated to a temperature
above the transformation range, it changes to the
face-centered cubic structure, which characterizes
the austenitic phase (see Fig. 8-2b). The carbon
that was present in the steel in the form of iron
carbide (Fe3 C) dissolves interstitially in the auste-
nite. Normally, it is considered that when the face-
centered crystal is cooled rapidly it reverts to the
 ·3
body-centered type (Fig. 8-2a). Although this is
true for iron, it is not true for iron containing
carbon (steel). Instead, the face-centered cubic
crystal transforms to a body-centered tetragonal
crystal as shown in Fig. 8-2c. Because there is not
sufficient time for diffusion to occur, the atoms
are trapped in the interstices. The body-centered
tetragonal crystal is the structure of martensite and
is referred to as untempered martensite. The de-
gree of tetragonality is a function of the carbon
content, as indicated in Fig. 8-3.
This body-centered tetragonal structure for
medium and high carbon steels is not only brittle;
that is, can be fractured, but it is also metallurgi-
cally unstable and will readily change to the body-
centered cubic ferrite and iron carbide (Fe3 C).
Actually, some precipitation may occur at or near
room temperature over a long period of time, such
as months or even years. However, no appreciable
tempering will result. Initial tempering is begun by
heating to a temperature of approximately 180 F
(85 C). This is demonstrated by the fact that years
before heat treating was considered a science, tool-
makers knew that tools made of high carbon steel
were unacceptably brittle in the as-hardened condi-
tion, and that the degree of brittleness could be
greatly reduced by placing the as-hardened tools in
a pot of boiling water for an hour or two. Hot oil

1.10------,--.,...-------------------.. .
1.08

1.06

~ I.
e
a::
010 T
1.02
j
c

0.98-----------------0010.-.--------------
Fig. 8-2. Three possible crystal structures of steel. (a) o 0.4 0.8 1.2 1.6
Bcc, body-centered structure, characteristic of ferrite; Carbon Content-%
(b) fcc, face-centered cubic, characteristic of austenite,·
and (c) bet, body-centered tetragonal structure, char- Fi.~. 8-3. E.ffeci of carbon on the tetragonality oj mar-
acteristic of martensite. tensile.
 4
baths also were used, but boiling water provided a
·simple means of controlling the temperature.
(Generally, -the use of boiling water to temper tools
is no longer practiced.)
Stages of Tempering
Hardened steels, containing martensite and
retained austenite, pass through three stages as
they are tempered. These stages are identified by
changes in physical properties associated with
changes In structure. Phase changes, however, do
not occur discretely. Because the reactions of the
three stages partly overlap, the temperature limits
assigned to each stage of tempering are approxi-
mate. Figure 8-4 illustrates the phase changes
taking place in the tempering of a high carbon steel.
When steel is hardened it consists of as-quenched
martensite and possibly some retained austenite
which, for various reasons (mainly chemical com-
Quenched to
Room Temperature
Firs! Stage of
Tempering (20 - 200°C)
Second S toge of
Tempering (200-300°C)
Third Stage of
Tempering (:>300°C)
.. Cementite Formed from tbe Reaction of E-tron Carbide
wit h Ferrite and / or Low-Corbon Martensite
Fig. 8-4. Phase changes taking place in the tempering
of Q high carbon steel.
position), has not completely transformed. Marten-
site, the hard acicular microstructure, light etching
and tetragonal crystal structure (Fig. 8-2c) consti-
tutes a supersaturated solid solution of carbon in
ferrite. Since tetragonal martensite has formed from
cubic austenite when the steel is relatively cold, the
increase in volume due to the lattice expansion re-
sults in a highly stressed condition.
First Stage. The IIrst stage of tempering involves
decomposition of martensite and proceeds at mea-
surable rates below 200°C (390°F); decomposition
has been observed even at room temperature.
When the steel enters the first stage of tempering,
the supersaturated condition of carbon in the fer-
ritic phase (alpha iron or a-Fe) will be partially
relieved. A transition phase, different from cemen-
tite, forms on tempering martensite at 265 to 570 F
(130 to 300 C). The transition precipitate is the
hexagonal close-packed carbide of iron and is re-
ferred to as £-carbide.
Precipitation of £-carbide is accompanied by
formation of a low carbon martensite. Its carbon
content varies from 0.17 to 0.280/0, depending on
the temperature of tempering, but is independent
of the carbon content of the primary martensite.
Since the finely divided carbide gives a dark-
etching effect, the martensite itself will now appear
darker under the microscope, in contrast to the
light as-quenched (white) martensite. When the
precipitation occurs, there is a slight hardening
effect. This effect may not be permanent, for with
the passage of time,. the martensite is further tem-
pered and becomes softer, so that the measured
hardness becomes lower, thus offsetting any hard-
ening effect contributed by the precipitation
mechanism.
At the same time that the precipitation is occur-
ring, a change occurs in lattice structure of the
tetragonal martensite. It changes from tetragonal
to cubic with an accompanying reduction in vol-
ume of the steel. Dimensional changes will be
discussed later in this lesson.
As a result of this dimensional change, which is
in the form of contraction, the highly stressed con-
dition brought about by the formation of tetrago-
nal martensite when the steel is cold and not plastic
is partly relieved.

Second Stage. During the second stage of tem-
pering, the retained austenite undergoes transfor-
mation at 300 to 570 F (150 to 300 C). This second
stage of tempering is characterized by an increase
in volume and a large increase in magnetization.
The product, bainite, is an aggregate of ferrite and
a carbide. In this stage, carbon diffuses from the
austenite and forms iron carbide. This lower car-
bon austenite, on cooling to room temperature,
has a different martensite transformation range
and additional martensite forms. This requires a
second temper for the newly formed martensite.
The transformation of retained austenite during
the second stage of tempering is a time-dependent
reaction, so that the complete transformation is
not accomplished unless the steel is held at the
tempering temperature for a sufficient length of
time. As the retained austenite transforms it may
add to the hardness of steel. This additional hard-
ness may offset the loss of hardness due to temper-
ing of the previously-existing martensite, and in
some cases may contribute to an overall increase in
hardness. This is particularly true in the case of
high-alloy steels, which will be discussed later.
During the second stage of tempering, the con-
traction of the steel noted in the first stage contin-
ues to occur with the passage of time. As a result,
the stressed condition is relieved still further, al-
though contraction may be countered to a large
extent by the increase in volume due to the trans-
formation of the retained austenite, if a significant
amount of retained austenite was present.
Third Stage. The third stage of tempering ex-
tends ftom approximately 550 F (290 C) on up to a
maximum temperature which generally does not
exceed 1250 F (675 C). In this stage, more pro-
nounced diffusion occurs, and the very fine car-
bide particles (£-carbide) formed during the two
previous stages coalesce to form larger carbide
particles which are easily visible under the optical
microscope.
A rOd-shaped carbide has been observed to form
between 395 and 480 F (200 and 250 C) in steels
containing less than 0.57070 C. At 750 F (400 C),
the rod-shaped carbides dissolve and are replaced
by spheroidal Fe3C precipitate. These tend to form
at the martensite and prior austenite grain bound-
aries.
5
Reactions also occur in the a matrix. The rIrst
occurs between 930 and 1110 F (500 and 600 C),
and involves the recovery of the dislocations in the
lath (martensite) boundaries. A low-dislocation
acicular ferrite is produced. Heating to 1110 to
1290 F (600 to 700 C) causes the acicular ferrite to
recrystallize and form equiaxed ferrite. The recrys-
tallization is more difficult as the carbon content
increases (due to pinning of the ferrite boundaries
by carbide).
Tempering Temperatures
Tempering temperatures most commonly used
in industry range from 300 to 11 SO F (150 to 620 C),
although on occasions steels are tempered at tem-
peratures as high as 1300 F (705 C). Although in
most instances a steel temperec;i at 1300 F (705 C)
will be but little harder than it was in its annealed
state, there are certain applications where the tem-
pered martensite structure is preferred over the
annealed structure. An example is the preparation
of certain alloy steels for nitriding. For this pur-
pose they are fully hardened by quenching from
about 1700 F (925 C) after which they are tempered
at 1300 F (705 C) to achieve a machinable hardness
of approximately 250 HB. The tempered marten-
site structure obtained by this procedure is super-
ior for nitriding compared with the annealed struc-
ture. Mechanical properties of tempered martensite
generally are superior to those of the annealed
counterpart at similar hardness levels. See also the
section at the end of this lesson: Selection of Tem-
pering Temperature.
Equipment for Tempering
Steel usually is tempered in an air atmosphere, a
molten salt bath or a hot oil bath. Flame and in-
duction tempering units also are used for special-
ized applications.
Tempering in Air Furnaces
The most widely used furnace is the forced-air
recirculating type. For most of the tempering tem-
perature range the convection mode of heating is
adequate. As a rule, convection is more efficient
than radiation up to about 900 or 1000 F (480 to
540 C). At temperatures above about 1000 F (540
C), .efficiency of the radiation mode of heating
gradually increases and efficiency of the convec-
tion mode decreases. In the pit-type batch furnaces
and the continuous type of tempering furnace, the
air is heated in a separate compartment and circu-
lated over the work. Therefore, in these types the
work is heated almost entirely by convection in the
lower temperature range.
Convection furnaces can accommodate a tem-
pering temperature range of 300 to 1400 F (150 to
760 C). Forced convection is essential for best
results in the tempering range of 300 to 600 F (150
to 315 C). The temperature range of 900 to 1400 F
(480 to 760 C) can be attained with either forced-air
convection or radiation, or a combination of both.
The electrically heated box furnace is well suited to
heating parts for tempering·at the upper end of the
common temperatlU'e range. This furnace is heated
primarily by the radiation mode, but the circulat-
ing fan provides a convection assist.
Efficient use of a continuous furnace is not at-
tained when production quantities are small and
parts vary in size, shape and mechanical require-
ments; the batch furnace is better suited for work
of this type. When a continuous furnace is used for
such applications, production time is lost in raising
or lowering furnace temperature. Sometimes a
dummy load must be placed in the furnace to ac-
celerate a desired reduction of temperature, or
production must be stopped until the temperature
is adjusted to suit. To conserve energyi~ batch-type
tempering, small lots are tempered on an ascend-
ing temperature sequence to minimize frequent
furnace cooling.
Tempering in Liquid Baths
Tempering by immersion in molten salt, oil or
molten metal is common practice. With this type
of heating, radiation, convection and conduction
can playa part in transferring heat. First, heat is
conducted directly into the work by contact with
the liquid, and second, the automatic stirring
which takes place in most liquid baths provides
convection heating, and finally, in the higher range
of temperature, the radiation mode plays an im-
portant role.
·Salt bath furnaces, either the type heated by im-
mersed electrodes, or the externally heated type,
"may be employed for tempering at 325 F (165 C)
and above. Natural convection in the bath pro-
motes uniformity of workpiece temperature.
All moisture must be removed from parts before
they are immersed in the molten salt, because hot
salt reacts violently with moisture. If dirty or oily
parts are immersed in the bath, the salt will become
contaminated and require more frequent rectifi-
cation.
All parts tempered in salt must be cleaned soon
after being removed from the bath, because aily·
salt that clings to them readily absorbs and retains
moisture and may cause severe corrosion. Parts
with small or blind holes from which salt is diffi-
cult to clean should not be tempered in salt.
One disadvantage of using a salt bath for tem-
pering is that no one salt is compatible with the
entire temperature range that may be used for
tempering. Some typical salt compositions and
their operating temperature ranges are listed in
Table 8-1. When contemplating the use of molten
salt for tempering, the salt supplier should know
the temperature range that will be used so that the
most appropriate salt compositions can be fur-
nished. Also, this gives the supplier an opportunity
to properly instruct the user regarding safety pre-
cautions that relate to the various salt compo-
sitions.
Oil baths for tempering may be similar in design
to that used for salt baths, or they may be steel
tanks with large hotplate-type burners. Submerged
electric heating elements also may be employed.
Stirring is essential for temperature uniformity
and longer oil life. Simple, oven-type temperature
controls may be employed, but overheating must
be avoided because of fIfe hazards. A standard
thermometer of the proper range may be used to
check the oil temperature.
Low-temperature tempering in a hot oil bath is a
simple and inexpensive method that is especially
useful for holding work at temperature for long
periods of time. The upper temperature limit is
about 375 F (190 C) for practical operation with-
out fume hoods. With proper ventilation, temper-
atures up to 450 F (230 C) are practical. When tem-

Table 8·1. Compositions and Operating Temperatures for Salt Baths Used in Tempering (a)
Composition of bath, 0/0
Sodium Sodium Potassium Sodium
Class nitrite nitrate nitrate carbonate
1........ .... 37 to 50 0 to 10 50 to 60
2. . . . . . . . . . . . 45 to 57 45 to 57
3. _. . . . . . . . . . 45 to 55
4. . . . . . . . . . . .
4A
(a) Military Specification MIL-5-10699A, Ordnance.
pering at a temperature above 400 F (205 C), a salt
bath usually is preferable to an oil bath.
Oils for tempering must resist oxidation and
have a flash point well above the operating temper-
ature. The most comm.o~yused oil is 600W steam-
cylinder oil.
Molten metal baths are used to some extent for
tempering, but they have been mostly replaced by
molten salt baths. When employed, a molten metal
bath usually is composed of commercially pure
lead, which melts at about 620 F (325 C). Gener-
ally, it is the m.ost suitable of all metals and alloys.
For special applications, however, lead-base alloys
having lower melting points are used.
Lead oxidizes readily. Although lead will not
adhere to clean steel, adherence of the oxide to
steel surfaces is a problem, especially at higher
tempering temperatures. W~thin the range of the
temperature usually employed, a film of molten
salt will protect the surface of the lead bath, and
the work will be easily cleaned. Above 900 F (480
C), a granulated carbonaceous material may be
used as a protective cover.
Specific applications exist where the use of
molten metal for tempering is distinctly advanta-
geous. Because of its high thermal conductivity,
lead heats the work rapidly and is thus useful for
selective tempering. A typical example is temper-
ing a ball joint. This part is carburized and
quenched to a minimum case hardness, of 59 HRC
and a core hardness of 30 to 40 HRC. The thread
and taper are then tempered in lead to produce a
maximum case hardness of 40 HRC. Molten lead
also is used in patenting of wire.
7
Operating.. Fuming·
Sodium Potassium Barium calcium temperature, temper-
chloride chloride chloride chloride F ature, F
325 to 1100 1175
550 to 1100 1200
45 to 55 11 50 to 1700 1720
15 to 25 20 to 32 50 to 60 11 00 to 1650 '1725
10_t_o_1_5_2_5_to_3_0_4_0_t_o_45_ _
15_ t_o_2_0_1_02_5_t_o_1_40_0__1_4_50_
Because of the high specific gravity of lead,
parts tempered in molten lead will float unless held
down by fixtures. All parts and fixtures must be
dry when immersed in the bath, to prevent the for-
mation of steam and explosion in the molten lead.
Precautions also must be taken to protect person-
nel from industrial lead poisoning; hoods and
ventilating equipment are necessary.
Temperature Control. For either gas or electric
heat, properly adjusted potentiometers of the on-
off type will control the tempering temperature to
within ±10 F (S C) at the thermocouple location.
With proportioning controls, temperature at the
thermocouple location can be held within about ±2
F (1 C).
Induction Tempering
There are two fields of application where induc-
tion tempering is especially adaptable: selective
tempering, and progressive tempering of bar stock
previously hardened by induction. Usually, hard-
ening and tempering is done in one integrated in-
duction unit. Because of the need for minimizing
temperature gradients in the workpiece, low fre-
quencies (sometimes as low as line frequency)
usually are used for induction tempering.
A disadvantage of induction tempering is that,
just as in induction hardening, some development
work must precede induction tempering of any
given part in order to establish a time-temperature
cycle that will provide the desired results. Once this
cycle is established, induction tempering can be
applied for mass production of identical work-
pieces.
 8
Principal Tempering Variables

Composition of the metal being tempered, prior

microstructure, hardenability, tempering tempera-
ture and tempering time are the major variables
which can act individually or in combination to
affect results obtained in tempering.

Effect of Steel Composition

In plain-carbon steels, the as-quenched hardness
gradually decreases as tempering temperature in-
creases. This is demonstrate~ by an 0.50070 carbon
steel in Fig. 8-1. The relationships between carbon
content, hardness and tempering temperature are
shown in Fig. 8-5 for six carbon ranges. All of the
curves decrease in a uniform manner, which is
characteristic of plain carbon steels.
Similar characteristics prevail for carburized
cases of plain carbon steels as shown in Fig. 8-6.
These data incorporate both tempering tempera-
ture and distance below the surface, showing their
effect on hardness.
Effect of Alloying. When alloying elements are
added to steel the conditions change, depending on
the alloying elements added and the total alloy
content. In the first and second stages of temper-
ing and below 700 F (370 C) in the third'stage of
tempering, the addition of alloying elements has a
relatively small effect. However, at tempering tem-
peratures above 700 F (370 C), the carbide-forming
elements such as manganese, chromium, molyb-
denum, vanadium and" tungsten, when added to
steels, produce higher hardness and strength for a
given tempering temperature. than can be obtained·
I
I!
10 200 600 1000 1400
Tempering Temperature-OF
Fig. 8-5. Hardness oj plain carbon steels of various
carbon contents, quench hardened and then tempered
at a series of temperatures. Tempering temperatures
were uniformly held for one hour.
in plain carbon steels having the same carbon con-
tent. The effect of adding several different alloy-
ing elements on hardness of steel after ·tempering
at several different temperatures is shown in
Table 8-2.
In general, for a given carbon content additions
of the carbide-forming elements retard softening
from tempering, and most notably for tempering
temperatures above 700 F (370 C).
For most of the conventional low-alloy steels,
the tempering temperature versus hardening curve
is higher than for plain carbon steel and drops
gradually. However, the drop is not as steep as for

Carburized at 1700 F, 4.5 hr, oil quenched and tempered

80 I
1024
70 +--+----+-- I-in. diem bar -

: 60....~~-==~~~--+--t----;--;----t
1 ~I:
~........~~~--+--~---+---+--~---+--01
I
c:
~

~ 50 t--~----~~----~~---+--;---+-~....-r-Io_-+----+--"...~~___;_-I__~__+___1
U
~ 40 t--~.....--¥--~--..iII!~d__--.::._...t.+__-!--~~___+--+----7f-___;---~:.___;_~~~~---I
3
.:tt:

~ :30 I----!----+--~--!-------

20 ~~--+---I----!-~--+---+--;---+-~~---L.._-+----+-----1--+---+-_f--~-+---I
o As Quenched Tempering temperature: .400 F ~600 F .800 F
, , I I
10
o 10 20 30 40 50 0 10 20 30 40 50
Distance below surface, 0.001 in. Distance below surface, 0.001 in.

Fig. 8-6. Effect of tempering on hardness 0.( carburi~ed cases in carbon -steels.

Table 8·2. Percentage of Alloying Element
~equired to Retard Softening by One Rockwell
"e" Unit at Various Tempering Temperatures
Temp~ring temperature, F (C)
400 600 SOD 1000 1100
Element added (205) (315) (425) (540) (595)
Manganese 1.33 0.65 0.40 0.40 0.40
Silicon (a) 0.22 0.22 0.22 0.22
Chromium (a) 1.33 0.45 0.19 0.19 0.19
Nickel 3.33 1.67
Molybdenum (a) 0.32 0.13 0.06 0.06
(a) Up to 1 % silicon, 1.5% chromium, or 0.5% molybdenum.
the plain carbon steel. This condition is illustrated
in Fig. 8-7, in which tempering characteristics of
an 0.40% carbon steel (AISI 1040) are compared
with those of an 0.40% carbon alloy steel (AISI
4140). Data were obtained by measuring the sur-
face hardness of I-in. (25-mm)-diameter speci-
mens. The AISI 1040 steel was water quenched
from 1550 F (845 C) and the AISI 4140 steel was
oil quenched from 1575 F (855 C).
As the amount of a carbide-forming element is
markedly increased, or substantial amounts of
more than one carbide-forming element are added
to the steel, it not only resists softening by temper-
ing to the ~xtent that the curve drops less rapidly,
but higher tempering temperatures result in the
curve taking an upward turn beginning at about
700 F (370 C). This condition results because as
alloy content is increased the amount of retained
austenite is increased in the as-quenched structure.
During tempering this retained austenite is condi-
tioned by carbon diffusing from the austenite and,
.,
tained austenite is transformed and the structure
of the steel then is predominantly tempered mar-
tensite; undissolved carbide and submicroscopic
alloy carbic;ie. Yet, when the double tempered steel,
1200 which no longer contains appreciable austenite, is
(650) reheated for long times at temperatures up to about
0.40 1000 F (5~ C), it does not soften appreciably.
0.22
0.19 Secondary Hardening. The effects of various
o.n amounts of a carbide-forming element (molybde-
0.06
num) on the tempering characteristics of an 0.35010
carbon steel are shown in Fig. 8-8. Here it is seen
that a molybdenum content of 0.07070 did not ef-
fect the gradual slope of the tempering tempera-
ture-hardness curve. However, as molybdenum
content was increased to 5010 the change in the
curve increased until the hardness after tempering
at over 1000 F (540 C) was substantially greater
than after tempering at approximately 800 F (425
C). This is typical secondary hardening resulting
from transformation of retained austenite to mar-
tensite during tempering.
In the more highly alloyed steels such as many
tool steels, the secondary hardening characteristics
are even more pronounced as shown in Fig. 8-9 for
two grades of high speed steels that are used for
metal cutting. Because high speed steels are used
Tempering Temperature, C
205 425 650
60
.....--........
-
o 40
~

'"~
~
~
4140

~
oil cooling to room temperature, transforms to c: "-
:c: 1040

"
martensite. Also, the diffused carbon forms com- en
CJ)

"~\
Q)
plex carbides with the alloying elements resulting c:
in precipitation of metal carbides and a condition "E
ca
known as secondary hardening. The persistence of :r: 20
the high level of hardness over a long period of \
time at high temperatures is referred to as red
hardness or resistance to softening.
o
The effect of red hardness of the alloy austenite 400 800 1200
caused by the transformation of retained austenite Tempering Temperature, F
generally is not important. This is clear in the case
Fig. 8-7. Effect of tempering temperature on hardness
of hardened high speed steel. As quenched, this of plain carbon versus an alloy steel of the same car-
steel may contain up to about 20 to 25070 untrans- bon content; all specimens were I-in. (25-mm)-diame-
formed austenite. When th.e quenched steel is tem- ter. The A/Sl 4140 steel was oil quenched from 1575
pered for 2 hours at 1050 F (565 C), and again at F (855 C); the AISI 1040 steel was water quenched
the same time and temperature, most of the re- from 1550 F (845 C).
 10
60 ~-.....------------
k'o~--r-i
I'

::l
~
,! ~ I \lT
Q) -§ 60 -_.-:.. ·--t-·----·;
. '
·-----·T-·-·-t .__ .
I
-T\
~40
\0" i I I : I

u ~ I I 1 : i\
o ~ 55 I.: I

cr:: ~~ \
:nen 3 0 t-t---+-----+------+-.....;Jip.....;..--~_..Io___I ~ 50 .....---_ _--O-_---I~--_;__-~-~:_+
Q)
c: -0- Ty~ 630 (MI) 2200 F 6
.-0 --0-- Typt! 623 (TIS) 2275 F
:;:.!

~20
I I
45 I--------!------r-·-·-T ----~-
I
---t--
! i i i
; I I
i
!
I
I a........-.
1
~_...._....__""O'___---'- _ _J

o 400 600 800 1000 12001400 AsO 200 400 600 800 1000 1200 1400

Tempering Temperature- 0 F Tempering Temperature, F

Fig. 8-8. Effect of molybdenum content on the resist-
ance of an 0.35% carbon steel to tempering. Note the
strong secondary hardening effect in the curves repre-
senting grades containing 2 % and 5 % molybdenum.
for cutting other steels, they must possess the max-
imum possible hardness and yet retain a certain
degree of toughness to withstand the heat and
pressure of the machining operation. To possess
the necessary degree of red (hot) hardness; that is,
hardness at the operating temperature, they must
contain appreciable amounts of elements such as
tungsten, chromium and vanadium. As a result of
their high alloy content, the steels contain consid-
erable retained austenite in the as-quenched con-
dition.
When such steels are tempered after the quench-
ing operation, an initial softening takes place as a
result of the decomposition of the hard tetragonal
martensite to the softer cubic form. As the temper-
ing temperature is raised, the softening continues
rapidly until a temperature of about 600 F (315 C)
is reached. At temperatures above 600 F (315 C) a
secondary hardening occurs that is the result of
precipitation of complex carbides from retained
austenite, and transformation of the remaining
austenite to martensite. At these upper tempering
temperatures, precipitation of a complex carbide
occurs directly from the retained austenite. When
this happens, the chemical composition and the
characteristics of the retained austenite change. It
becomes less stable and transforms to martensite
when the steel is cooled below the new M s from the
tempering temperature. In commercial practice, at
the temperatures usually used for tempering tool
Fig. 8-9. Hardness versus tempering temperature for
Ml and TIS grades of high speed tool steels (Teledyne-
Vasco).
steels (about 1025 F or 550 C) and with the short
tempering times employed, very little, if any, re-
tained austenite transforms at the tempering tem-
perature, and thus the steel must be cooled below
the Ms again before the austenite can transform.
Longer isothermal holds may, however, result in
transformation of retained austenite, the steel
undergoes secondary hardening as shown in Fig.
8-9. The peak usually occurs at approximately
1000 F (540 C). When the tempering temperature
exceeds this peak the hardness decrease is rapid.
For this reason, tempering of high-alloy steels into
the secondary range requires precise process con-
trol; otherwise loss of hardness will result.
When the retained austenite transforms during
cooling, it does so at temperatures very close to
room temperature. This means that the expansion
accompanying the change from austenite to mar-
tensite occurs when the steel is quite rigid and not
in a good condition to adjust itself to the volume
change. As a result, the steel is placed in a highly
stressed condition which must be relieved if break-
age is to be avoided. Therefore, it is recommended
that the steel be subjected to a second tempering
operation, using the same time and temperature
cycle, or about 50° F (28 0 C) less, as was used for
the initial tempering operation. For further details
see the section on Multiple Tempering, later in this
lesson, or Lesson 15, Heat Treatment of Tool
Stct»l~.
 ·11
Effect of Prior Microstructure 65 340F AQ
60 44~ --+-~~--+---+--+

Carbon and low alloy steel parts as quenched in 55 S4?~~~--+---+---+~~~+--'t"--'1

650--*-~~~~--+---+---l_I--+--"i
production are seldom fully martensitic. Fre- en 50 -750
quently, only 50 to 90070 martensite is present at ~
~
45 -850

the center of the section, and sometimes the cen- ~ 40 - 950;...;;'~~~ ~-+--;---t--+--+--+--t
U 1050
tral structure may be mostly bainite, ferrite and ~ 35 I ...;;;:p.~~~~~---;~I--;---;

pearlite. For rational design of such parts and as ~o 30 1I50L~~3;~*~~~~

I
a: 25 1200=-;..-==t~~--.,.....-j~1==~
an aid in heat treating them, it is important to I
2 0 ~~- 1275 ;;::=::;:=--;'-~~-+---+--+---t
know the temperability of structures other than
I
15 ""-+--+---+-,~---;----;--;-r__;___-r-_t_1
martensite. Ideally, this information should be
I 0 '---J..-- ...I.-""""-........ .........~-----~
available for all structures found in the end- o 4 8 12 f6 20 24
quenched hardenability bars of the standard steels Distance from Quenched end, sixteenths of an inch

and is available from some steel suppliers. Fig. 8-10. Hardness along end-quenched hardenability
specimens after tempering.
Figure 8-10 shows the results obtained after tem-
pering hardenability bars from one heat of 4063
steel for 1 hour at·ll different temperatures.
Time Versus Temperature
Figure 8-11a compares the tempering character-
Relationships
istics, at 1050 F (565 C), of martensite, isother-
mally transformed pearlite and bainite in AlSI Temperature and time are interdependent vari-
1095 steel. Figure 8-11b shows data obtained on ables in the tempering process. Within limits, low-
AlSI 4320 end-quenched hardenability specimens ering temperature and increasing time usually can
before and after being tempered. produce the same result as raising temperature and
decreasing time. The relationship between time
The data presented in Fig. 8-11a indicate that
and temperature can be closely approximated from.
there is no great difference in the rate of softening
empirical formulas developed for this purpose.
for three basic microstructures, and that tempering
response is a function of hardness and is essentially Although, as previously demonstrated in this
independent of microstructure. lesson, temperature is the major controlling factor

70 I ,
60
65
I ~O95
Tempered at 1050F .. 55
I
1 43 20
Martensite. 0/0
J J 1 I i 1 1 i
50 100 99 90 7S 50 25 10 S
.......
'

~ 55 en 451--~--+---+--+----+--+---+-t---+--t

~
tit en
4D CP
c:
-Eo 50 ~ Martensite "E
o
401---+--~
~:-
.I:
u45 " r"~~ ~ V Pearlite - .&:.
(.) 35 t--JIlII-+----+-~
a:
CD ~~~ ~ 301---+-~~---'~-+­
~40
r-;;:r:::: i""..
u ~"""""- u
~
o ~
o
a: 35 a: 25~~--+---+-~~~r~~~-t---+---i
~~~
~
30 '-- """'" 20J--4--+-~-~4--+-"::::+~~=+=--f
- BainiteJ41 ..... ,,~
25 I Ill' , 15
20
o J 2 10
II100 1000
I 10,000
(a) IOa.-.....--..........
o 2 4
"""--
8 6
~-.(-b~)

10
Tempering time, sec Distance from quenched end. 0.1 in.

Fig. 8-11. Effect of prior microstructure on hardness after tempering. (a) AISI 1095 steel tempered at 1050 F (565
C) for various periods of time. (b) Hardness before and after tempering and amount of martensite present before
tempering in AISl4320 steel end-quenched hardenabi/ity specimens tempered for two hours.
 12
in the tempering process, both the stress relief and
structural change reactions also are influenced by
the holding time at the tempering temperature.
Essentially, the effect of increasing the holding
time is equivalent to raising the temperature a cer-
tain number of degrees. For example, a tempering
treatment of 5 hours at 1050 F (565 C) will produce
the same hardness and properties as Y2 hour at a
temperature of 1125 F (605 C), or 75 F (40 C)
higher. This relationship between tempering tem-
perature and time bas been investigated extensively
for many steels and has been found to fit the fol-
lowing formula quite consistently:
T2 + 460 log t 1 + 20
=----
T} + 460 log t 2 + 20
where: T} is temperature in OF and· t 1 is time in
hours for one tempering treatment. T2 is tempera-
ture in OP and ~ is time in hours for another treat-
ment giving identical hardness and properties.
Note that the common logarithm of the time is
used in this formula, just as it is used in all pub-
lished TTT curves. As before, this means that the
first hour is more important than the second, the
second is more important than the third, and so
on. Thus, when a short tempering time such as ~
hour is used, it is necessary to control the time at
temperature very accurately to obtain consistent
hardness and properties. A longer time like 5 hours,
on the other hand, will not produce as much varia-
tion if an error of a few minutes is made in esti-
mating when the piece reached the tempering tem-
perature. Where multiple. temperi:ng is employed,
the total time at temperature determines the final
hardness, regardless of. the number of cycles. Ac-
cordingly, a treatment of 3 cycles of 2 hours at
temperature each time should give about the same
hardness as a steady 6-hour hold at the same tem-
perature. However, the strength and toughness
properties will not be the same due to the improved
structural transformations resulting from the mul-
tiple tempering operations.
To illustrate briefly the use of the formula, con-
sider a steel which is known to give a hardness of
60 HRC after tempering 1 hour at 550 F (290 C).
With a large die block of this steel, it might be
desirable to use a to-hour tempering treatment to
ensure a thorough job. What temperature should
be used with the lo-hour holding time to still ob-
tain 60 HRC? The known values are T 1 = 550, t}
= 1, ~ = 10; and we can solve for T2 as follows:
T 2 + 460 log 1 + 20 0 + 20 20
= = =~
550 + 460 log 10 + 20 1 + 20 21
T2 + 460 = (20/21) (550 + 460) = 962
T2 = 962 - 460 = 502 F (261 C)
This result indicates that 10 hours at about 500 F
(260 C) will give the same hardness and properties
as 1 hour at 550 F (290 C).
A number of graphical methods have been de-
veloped to show this relationship between hardness,
tempering temperature and time for various steels.
One method is demonstrated in Fig. 8-12. These
data were plotted from careful hardness readings
on 120 specimens of a medium alloy die steel which
had been hardened under specified conditions and
then given tempering treatments ranging from 200
to 1300 F (95 to 705 C) and 0.1 to 100 hours.
In the curves shown in Fig. 8-12 tempering tem-
perature is plotted against holding time (logarith-
mic scale) with lines of constant hardness- drawn
across the diagram. From this plot, the heat treater
can select the combinations of temperature and
time that will give a specified hardness.
Again it should be emphasized that tempering
temperature is the major factor to be considered in
response to tempering as shown in Figs. 8-1, 8-5,
8-6 and 8-7. In considering results shown in Fig.
8-7, the hardness values shown. are for tempering
for a period of one hour at the specified tempera-
tures. .If shorter holding times are utilized the
hardness will not be reduced quite as much, and
likewise longer tempering times will show the same
further reductions in hardness. Figure 8-12 indi-
cates that the same degree of te:qtpenng (as far as
mechanical properties are concerned) may be at-
tained with a shorter time at a higher temperature,
or with a longer time at some lower temperature. It
is for this reason that a commercial tempering
procedure is often designed for short time periods
at higher temperatures because this practice is the
most desirable from an economic viewpoint at no
sacrifice in properties and with a saving of energy.
In large sections, or where heavy furnace charges
are tempered, considerable lag may exist between
 13

30

~
o
I 1000
!::t
~

2
J ax> t-55--+----F~IJI_W~-__+___+___+___+__;._+_+_4_+__-+___..::;:;:::=.~~
~
.,.c:
i 600~---+-......-.~~
E
~

O~-
a .......----..............................
--~..L- ........................L...I..__......L._£._..JI......J.~...u.J
0.1 LO 10 100
Tme-Hcus (Log Scale)
Fig. 8-12. Relationship of hardness, tempering temper-
ature, and tempering time jor a medium alloy die steel.

the time when the outside and the interior of the
steel charge reaches the tempering temperature.
Under sw;h conditions it is safer practice to utilize
longer time periods at lower temperatures so that
more uniformity is obtained. However, once the
part has been heated uniformly throughout its
thickest section, little if any benefit can be obtained
by longer holding times.
For alloy steels, particularly the higher alloys in
which substantial amounts of austenite are re-
tained, somewhat more time should be allowed in
tempering compared with plain carbon steels.
However, for high-alloy steels, the practice of
double tempering is preferred (see the section on
Multiple Tempering).
maximum amount of stress can be removed, but
this maximum is reached only after the lapse of
considerable time. The relationship of time and
temperature, as it pertains to the removal of resid-
ual stress, is shown quantitatively in Fig. 8-13. The
removal of equivalent amounts of stress can be
accomplished by shorter time periods at higher
251-------+----04----p...-----t---I
800°F

Tempering Versus Stress Relief JOOQOF

5~..::==::t===:±===t=-~12~O~O~o~F=-=
The effect of time and temperature on the re-
moval of stress follows the same pattern as for the 4 8 12 16
lowering of hardness. The higher the tempering Time at Temperature-Hours
temperature, the more complete is the removal of Fig. 8-13. Influence of temperature and time at tem-
stress. Furthermore, at anyone temperature, a perature on relief of stress in an 0.27% steel.
 ·14
.~
temperatures, or much longer time periods at o
lower temperatures. Note that the reduction to a )( 0
constant level is substantially complete after two
hours.
-11-1 -4
<l t-----+----~+----+_-____f
c
o
In other words, the removal of stress closely :; -8
parallels lowering of hardness in tempering. This is
logical because the hardness is brought about by
stress within the lattice; that is, by transformation 0.1 1.0 10
of austenite to martensite at a relatively low tem- Tempering Time, Hrs.
perature. Fig. 8-14. Effect of tempering time on dimensional
changesjor an 0.94% carbon steel quenchedfrom 1500
Effect of Tempering on Dimensions F (815 C) and tempered at 200 F (95 C).
v
Virtually all heat treating operations are accom- o
panied by dimensional changes. All metal expands
on heating and contracts on cooling. However,
when fcc austenite is cooled slowly and converts to
pearlite (bec ferrite + Fe3C), the metal expands to
some extent during transformation. When auste-
nite is cooled rapidly enough (quenched), it trans- 0.1 1.0 10
forms to tetragonal martensite; the degree of Tempering Ti me - Hrs.
expansion during this transformation is signifi- Fig. 8-15. Effect of the second stage of tempering on
cantly greater compared with expansion during the contraction of an 0.94% carbon steel:t water
transformation of austenite to pearlite. Tetragonal quenched from 1500 F (815 C) and tempered at 450 F
(untempered or as quenched) martensite is the least (230 C).
dense (occupies the greatest volume) of the steel
structures. As explained earlier in this lesson, te- ()
tragonal martensite will transform over long peri-
ods of time to the body-centered cubic structure, ~ 56 r---~..,...,.,..,~~.,..,...,....,.,.,.,..---,.""""""",.,.,..-~~
~
even at room temperature, and some contraction U

will take place. This contraction will take place

~ 52 t------+----~---..;.---~
I
more rapidly when the tetragonal structure is heated ~ 4 8 t-----~--+~----+----~-____f
Q)
to about 200 F (95 C) or higher. Figur.e 8-14 illus- C
trates the rate of contraction when O.94OJo carbon ~ 44 "'----..-..........................~. . . . . ~........--""-......................... ......... ..J

steel is tempered at 200 F (95 C) for periods of 0.01

to 100 hours.
:E t5% Ni-l%C Steel!
~ 6 0 .--...-......~.......- .........~"t""'P"'!""~......-..,......""""P"""'.......,...--.,.."""P"-r .....
The effect of tempering the same steel at 450 F .....
Q)

(230 C) is shown in Fig. 8-15. The metal experi- .....~40

enced a slight expansion during the first one-half CJ)

hour, then slowly contracted during the remainder ~20

-0
of the test. Q.)
c: 0
The above is based on the assumption that there .....o
is essentially no retained austenite. Even in plain ~ 0.01 0.1 1.0 10
carbon steels, there usually is some retained auste- Tempering Time - H rs.
nite, and in alloy steels there may be significant Fig. 8-16. Effect of time on the second stage of tem-
amounts. Therefore, when the quenched structure pering:1 on the transformation of retained austenite
contains both tetragonal martensite and retained and hardness in a 1.0% carbon:1 4.82% nickel steel,
austenite and is subjected to tempering, first there oil quenched from 1700 F (925 C) and tempered at
is contraction due to changing the tetragonal struc- 500 F (260 C).

ture to the body-centered cubic structure; then,
when a temperature of about 500 F (260 C) is
reached the austenite is conditioned and, on cool-
ing to room temperature, transforms to martensite
and expansion takes place. The net result is often
difficult to predict.
The effect of tempering time at 500 F (260 C) on
the transformation of retained austenite and the
. resultant effect on hardness of a nickel steel is
illustrated in Fig. 8-16. Mter the austenite had
transformed, the hardness was greater than the as-
quenched hardness. In this instance, a large amount
of expansion would be expected.
When tempering the plain carbon steels at tem-
peratures above 550 F (290 C), further contraction
may occur provided there is little or no retained
austenite. However, for the higher carbon steels
which are prone to retain austenite, considerable
expansion can be expected as the retained austenite
transforms to martensite.
Effect of Cooling From
Tempering Temperature
Since tempering is carried out below the lower
critical temperature, no austenite is formed during
the operation. Therefore, it would be assumed that
the method of cooling from the tempering temper-
ature to room temperature would have no bearing
upon $e microstructure of the steel and hence the
mechanical properties. For carbon steels and low
alloy steels this is correct, and with most of these
metals the method of cooling from the tempering
temperature usually is not critical. In some cases,
where time is a factor, such steels might be water-
quenched after tempering. However, in the more
highly alloyed tool steels, particularly those tem-
pered at low temperatures and therefore having a
high degree of residual stress, such practice is not
desirable. In highly alloyed tool steels, transfor-
mation of retained austenite during cooling from
the tempering temperature also may be a factor
which must be considered because this can affect
both the microstructure and the mechanical prop-
erties. With rapid rates of cooling, the normal
volume change occurring when the steel passes
from the tempering temperature to room tempera-
ture may create an additional stress which, imposed
on the existing and only partially relieved residual
stress, may result in breakage. Therefore, if there
15
is any question, it is best to cool from the temper-
ing temperature at a moderate rate and to avoid
such drastic cooling as water quenching.
In most commercial applications, the tempered
parts are removed from the furnace and allowed to
cool in still air to room temperature. When parts
have been tempered in molten salt, they should be
placed into a water rinse while they are still warm;
otherwise the salt may be very difficult to remove.
For some of the highly alloyed tool steels, how-
ever, for reasons stated in the preceding para-
graphs, it is essential to furnace cool from the fIrst
tempering operation; otherwise, cracking is likely.
High speed steels such as M6 and M44 are typical
examples of steels which must be cooled slowly
from the tempering temperature.
The level of stress that can be set up in a piece of
steel, when quenched from tempering tempera-
tures, is entirely due to the normal thermal volume
changes. These stresses can be very high in inten-
sity. Figure 8-17 shows the magnitude of the stress
developed when a plain 0.45070 carbon steel round
bar was quenched from various subcritical temper-
atures. This stress may be undesirable in that it
contributes to breakage. In some cases, however,
the stress may be used to advantage. For example,
a compressive stress set up on quenching from the
tempering operation can be useful in the preven-
tion of fatigue failure, provided that cracking is
not a problem.
Temper Colors
When a bright piece of steel, such as results
from machining or polishing, is heated in air, it
will oxidize, forming a thin surface f"Ilm oxide. The
thickness of the film will vary with the temperature
to which the steel is heated and, to a lesser extent,
with the time held at that temperature. The oxides
thus formed display a wide range of color known
as temper colors. Since the color seen depends on
the thickness of the oxide formed, which is a func-
tion of the tempering temperature and time at that
temperature, the temper colors can be used as a
rough guide to indicate extent of tempering for
temperatures up to near 600 F (315 C). The rela-
tionship of temper color and tempering tempera-
ture of an 0.95070 carbon steel is presented in Table
8-3. The range of temper colors shown in this table
 16
Table 8-3. Relationship of Temper Color and Tem- straw, brown, purple, and finally blue or black as
pering Temperature, 0.95% Plain Carbon Steel the oxide layer thickens. When the oxide layer
Approximate tempering thickens further the color cycle repeats, but is less
Temper colors (first order) temperature, F (C) intense. In the case of a relatively thick section
Yellow 380 (195) where considerably more time was required to
Straw 420 (215) bring the center of the piece to temperature, as
Brown 460 (237)
Purple 500 (260) compared to the surface, the color observed does
Blue 580 (305) not always reflect the true temperature.

is not complete, because many in-between hues can

exist.
These gradations of shade can be extremely fine
and beyond the ability of the average person to
differentiate. Furthermore, the colors will vary
somewhat with the alloy content since different
steels will form oxides of different composition.
This in turn affects the color for a given thickness.
The use of temper colors generally is limited to
application to thin sections or small pieces, and in
shops not having precision temperature indicating
equipment. The limitation of size is brought about
by the fact that the temper colors progress along
the color range not only because of increasing tem-
perature but also with the passage of time, which is
needed for through heating of thick sections.
Thus, if a piece of steel formed the yellow oxide at
380 F (195 C), continued holding at this tempera-
ture .would result in a progressive color change to
Embrittlement Resulting
From Tempering
Many forms of embrittlement can lead to brittle
fracture of steel parts. At least four forms can
occur during a tempering heat treatment. These
are blue embrittlement, temper embrittlement in
alloy steels, 500 F or 350 C embrittlement in high-
strength, low alloy steels, and 750 to 930 F or 400
to 500 C embrittlement in ferritic stainless steels.
Embrittlement is sometimes detected during
fabrication, where it frequently appears in the
form of unusually high numbers of parts that
crack or break in regions of moderate to severe
deformation. If embrittled parts are not detected
during fabrication and are placed into service,
there is an increased danger that they will undergo
brittle fracture, especially if subjected to impact

Longitudino·1 Stress Tangential Stress Radial Stre ss

!
en
CL

o 40 "'--'--+--I-I-~+---+---t
-0
o

-80
38.5 20 o 20 38.5 38.5 20 0 20 38.5 38.5 20 0 20 38.5
Cross Sectional Area-Sq. Inches
Fig. 8-17. Stress patterns detected in four 7-in. (IB-em) round SAE-I045 steel cylinders water quenched from vari-
ous subcritical temperatures. ~ represents center-line of bar,· edges represent surface of bar. Positive stress values
are tension; negative values are compression.

loads. Consequently, the possibility of embrittle-
ment should be considered during the design of a
steel part, and provision made for countermeas-
ures should embrittled material be detected at any
time during manufacture or subsequent service.
It is not always possible to detect embrittlement
by tension testing or hardness testing. In most
instances, embrittled material has higher hardness,
higher strength and lower ductility than material
that is not embrittled. Unfortunately, the differ-
ences are not always great enough t6 be positive
indicators of embrittlement. Impact properties, on
the other hand, are reliable indicators of embrittle-
mente Often, embrittled steel exhibits a complete
loss of toughness at room temperature, or a very
significant rise in the ductile-to-brittle transition
temperature when compared to the same steel that
is not embrittled, as indicated in Fig. 8-18. Some-
times, particularly for steel sheet, embrittlement
can be detected in a bend test.
Blue Brittleness. When plain carbon steels and
some alloy steels are heated between 450 and 700 F
(230 and 370 C), there is an increase in strength
and a ,marked decrease in ductility and impact
strength. This embrittling phenomenon is known
as blue brittleness, because it occurs in the blue-
heat range. The increase in strength and decrease
in ductility' are caused by precipitation hardening
within the critical-temperature range. Deforma-
tion while the steel is heated in the blue-heat range
results in even higher hardness and tensile strength .
after cooling to room temperature. If the strain
rate is increased, the blue-brittle temperature range
increases.
The use of susceptible steels that have been
heated in the blue-brittleness range should be
avoided, especially if the steels are subjected to
impact loads, because the toughness of this mate-
rial will be considerably less than optimum.
Temper Embrittlement. Many of the common
low-alloy steels exhibit an increase in their ductile-
to-brittle transition temperatures after being heated
in the range of 700 to 1070 F (375 to 575 C) or
slowly cooled through this temperature range.
Plain-carbon steels containing less than 0.30070
manganese are not susceptible to temper brittle-
ness, although any steel containing appreciable
amounts of manganese, nickel or chromium will
be susceptible provided the steel also contains one
17
-148 32 212 392 OF
140 .---..---~~--.----.--...--........---~-..
:J 120 r--_T_ra_n..,.s..,it_io_n~Te~m~pe=r==a==tu_r.,..e~~::sP--f
1 I I
Li: I00 t---~___#_-+----+-.......,..--..-..----I
I
~ 80 t------I-~~-----f---.p-:.f.---I----J
~
Q)
~60~--"""""'--"""'---"""""--~---I
-to-
(.) 40 ....----.---
c
~
E 20 ...---~_.p._.:_-__+_-_#__~---+-----f
""-_""""'....-.Io ---'
O~ I.-oo-.
-100 0 100 200
Testing Temperature - °c
Fig. 8-18. Determination oj the transition temperatures
and the effect of temper embrittlement on the transi-
tion temperature.
or more of the impurities antimony, phosphorus,
tin or arsenic. Thus, a combination of an alloying
element and an impurity is required to cause tem-
per embrittlement. Nickel-chromium steels are
more susceptible than steels containing a single
alloying element, but even nickel-chromium steels
are not susceptible if made from sufficiently high-
purity materials. Carbon-free alloys are suscepti-
ble to a much lesser degree than steels containing
0.30 to 0.40070 carbon. Molybdenum (0.5 to 1.0070)
appears to delay, but not eliminate, embrittlement.
Silicon may increase susceptibility; steels made
using carbon-vacuum deoxidation appear to be
less susceptible than those made using silicon deox-
idation.
. Embrittlement occurs most rapidly around 840
to 885 F (450 to 475 C). Embrittlement is reversible
and may be completely eliminated by heating for a
short time above the embrittling range followed by
rapid cooling to below 570 F (300 C). Following
such a de-embrittlement treatment, heating in the
embrittling range again produces embrittlement.
SOO F Embrittlement. 500 F (350 C) embrittle-
ment of high-strength, low-alloy steels occurs over
a temperature range of approximately 400 to 700 F
(200 to 370 C). It occurs mainly in steels that have
been heat treated to a microstructure of tempered
martensite; thus, the more descriptive term tem-
pered-martensite embrittlement is sometimes used.
Steels with microstructures of tempered lower
 18
bainite also are susceptible to 500 F embrittlement,
but steels with pearlitic microstructures and other
bainitic steels are not.
500 F embrittlement is evaluated by measuring
the effect of tempepng temperature on room-tem-
perature impact energy; this is in contrast to tem-
per embrittlement, which is evaluated by measuring
the effect of tempering temperature on the ductile-
to-brittle transition temperature.
500 F embrittlement is believed to be caused by
ferrite networks resulting from precipitation of
cementite platelets along prior austenite grain
boundaries. However, some investigators believe
that the precipitation of grain-boundary cementite
platelets as such is responsible for 500 F embrit-
tlement.
Steels containing substantial amounts of chro-
mium or manganese are highly susceptible to 500 F
embrittlement. Aluminum contents above 0.04070
reduce embrittlement, and additions of 0.1 070 alu-
minum usually eliminate the problem. Some de-
gree of embrittlement has been observed when
phosphorus, antimony, arsenic, tin, silicon, man-
ganese or nitr9gen additions were made to high-
purity steels. Additions of nitrogen produced
intergranular fractures, but the other embrittling
agents did not. Commercial grades of steel that are
subjected to 500 F embrittlement treatments frac-
ture intergranularly.
400 to 500 C Embrittlement. Fine-grained, high-
chromium stainless steels normally possess good
ductility. However, if they are held for long peri-
ods of time at temperatures in the range of 400 to
500 C'(750 to 930 F), they will become harder, but
embrittled. Embrittled high-chromium ferritic
stainless steels contain two ferrites, one rich in iron
and one rich in chromium.
Susceptibility to 400 to 500 C embrittlement
increases with increasing chromium content, with
the highest degree of embrittlement occurring with
chromium contents greater than 19070. At least
150]0 chromium is necessary for embrittlement to
occur. The -effect of carbon content on embrittle-
ment is minimal. High-chromium steels that con-
tain at least 1070 titanium are more susceptible to
embrittlement than are similar steels with lower
titanium contents. High-chromium steels that
contain 2.4 to 4.15070 niobium are more susceptible
to embrittlement than are similar steels with lower
niobium contents. 400 to 500 C embrittlement
occurs more rapidly in steels with high levels of
silicon.
The embrittlement caused by prolonged soaking
within the 400 to 500 C range can be removed by
soaking at somewhat higher temperatures for sev-
eral hours.
Multiple Tempering
Multiple tempering is principally used (a) to re-
lieve the quenching and straightening stresses in
irregularly shaped carbon and alloy steel parts and
thereby lessen distortion, (b) to eliminate retained
austenite and improve dimensional stability in
such parts as bearing components and gage blocks,
and (c) to improve yield and impact strengths with-
out decreasing hardness.
Stress Relieving. Tempering and relieving of
stresses induced by cold work are closely related.
To achieve specific properties and to solve specific
problems. associated with heat treated parts, a
secQnd tempering operation, such as described in
the following case history, is commonly employed.
A six-throw, seven-bearing diesel engine crank-
shaft, weighing 175 lb (79.4 kg), distorted in heat
treating and rough machining to such a degree that
cold straightening was required. The straightening
operation induced additional stresses, which
caused distortion in final machining. The problem
was solved by first tempering the AISI 1046 steel
shaft at 850 F (455 C) to a hardness of about 321
HB, which allowed hot straightening. The shaft
was then retempered at 900 to 1000 F (480 to 540
C) (depending on the composition of the particular
heat) to produce a hardness of 269 to 302 HB and
relieve residual stresses.
Elimination of Retained Austenite. The struc-
ture of an as-hardened steel is a heterogeneous
mixture of retained austenite, untempered marten-
site and carbides. Retained austenite is present in
small quantities in the as-quenched plain carbon
and low alloy steels, and in the highly alloyed
grades it is present in appreciable amounts. The
untempered tetragonal martensite is transformed
to a tempered body-centered cubic martensite dur-

ing the tempering cycle. Also, in the highly alloyed
gradeS, a further amount of untempered tetragonal
martensite is formed from the retained austenite
still remaining, during the cool-down from the
tempering cycle. It is, therefore, good practice to
double or triple tempe~ to ensure more complete
transformation of austenite and to temper the
freshly formed martensite. It is important to note
that this retained austenite is relatively stable, and
therefore is difficult to transform; also, the mar-
tensite only form~ during the cooling cycle. Pro-
longed heating on a single treatment is therefore
not nearly so effective as introducing multi-tem-
pering. Studies on the progression of tempering
have shown that the process follows a logarithmic
pattern; that is, the frrst hour is more important
than the second ho~r, and so on.
Three or more tempering cycles are sometimes
used for some of the highly-alloyed high speed
steels such as M46 or M47. Three complete cycles
(triple tempering) is common, and as many fiveas
complete cycles are sometimes used in an endeavor
to obtain the ultimate in austenite transformation,
and in turn, the highest possible hardness. Al-
though use of additional tempering cycles usually
is a matter of gradually diminishing returns, in
terms of transfOrming retained austenite, there are
applications where the extra energy and time con-
sumption is justified, by increased life obtained
from slightly harder tools. By use of multiple tem-
pering it is possible to attain hardness values as
high as 70 HRC on some of the highly-alloyed high
speed steels.
Some of the lower alloy tool steels are sometimes
subjected to three or more complete tempering
cycles when the ultimate in dimensional stability is
important. Gage blocks are notable examples.
Protective-Atlnosphere Tempering
Although protective atmospheres are not often
used for tempering of hardened steel parts, there
are applications where it is desirable.
In some specific instances where parts must be
tempered at relatively high temperatures, and pro-
duction of scale-free work is essential, the parts
are tempered in a closed furnace wherein the prin-
cipal mode of heating is radiation, and the protec-
19
tive atmosphere is argon or nitrogen, both of which
are inert gases. Tempering with atmospheres of
inert gas is, however, expensive because a great
deal of gas is used. For complete protection, the
furnace must be purged with the inert gas until it is
virtually free of air before the temperature is ele:-
vated. In many instances, more of the inert gas is
used for purging than is used for protection during
tempering. Thus, the amount of gas used for purg-
ing and continued protection during tempering
results in a high cost for tempering.
Exothermic Gas for Protection. There are in-
stances where a rich exothermic gas supplied to
tempering furnaces was used as a means of pre-
venting scaling. However, the use of such an atmo-
sphere is absolutely restricted for use above about
1175 F (635 C), and even at this temperature, and
with close control of gas composition the practice
is dangerous and is not recommended because of
the danger of developing an explosive mixture.
Leaner exothermic gas compositions may be used
as an alternate to provide some protection.
Tempering in Molten Salt. When it is essential
that parts be tempered in the temperature range
where scaling occurs, the use of a molten salt bath
often is a practical and economical approach.
Molten salt provides complete protection against
scaling, and salt compositions are available which
can accommodate a wide temperin~ temperature
range (Table 8-1).
Steam Treating and Tempering. A s~eam atmo-
sphere during tempering cannot be properly called
a protective atmosphere; at least not by the normal
considerations for a protective atmosphere. How-
ever a steam atmosphere can provide a form of
protection, and at the same time develop a useful
oxide' finish on parts that are amenable to being
tempered in the range of 550 to 1150 F (290 to
620 C).
While the parts are being tempered by heating at
any temperature within the above range, the pro-
cess imparts a blue oxide finish of good appear-
ance. This oxide coating often is useful as a means
of preventing scoring of moving parts, because it
acts as a wick and holds oil tenaciously. It also
provides a measure of resistance to atmospheric
corrosion; especially with a fIlm of oil. In many
instances the oxide treatment, which can be ob-
 20
tained simultaneously with tempering, has success-
fully replaced the oxide treatments normally
obtained by chemical means, but at lower cost.
A typical procedure for simultaneous steam
treating and tempering is given as follows:
Carefully cleaned parts are loaded into the fur-
nace (batch type). The furnace is sealed and the
temperature controller is set at 700 F (370 C).
When the furnace temperature reaches 650 to 675
F (345 to 355 C), steam at full line pressure is in-
troduced into the furnace. When the furnace
reaches 700 F (370 C) it is held at this temperature
and purged for approximately 15 minutes. Fur-
nace pressure is then adjusted to slightly positive,
and steam purging is continued for another 15
minutes. The load is then heated to 1000 F (540 C)
and held at that temperature for 1 ~ hours. With
the steam the load is then cooled to 550 F (290 C).
The steam is then shut off, and the workpieces are
removed from the furnace and quenched in soluble
oil. Other means of cooling can be used, but
quenching in soluble oil improves the lubricity of
the parts, and also enhances corrosion resistance~
The procedure described above was developed
for parts that were to be tempered at 1000 F (540
C), but other procedures can be developed for
temperatures as low as 550 F (290 C).
Care should be exercised when processing large
quantities of powder metal parts. Pores in the
sintered parts contain combustible gases which
may be evolved and form an explosive mixture.
Selective Tempering
Selective or localized tempering is applied to
parts that are required to have a significant varia-
tion in the hardness of adjacent areas. Chisels,
punches, the upset ends of cold formed rivets, and
the threaded portions of carburized parts are typi..:
cal examples. Localized tempering also is employed
in preheating and postheating weld areas when it is
desired to lower the hardness in the heat-affected
zone of the weld.
Selective tempering entails heating a local area
to the required tempering temperature without
heating the remainder of the part to this tempera-
ture. Induction heating coils, flame heads and
immersion in lead or salt baths are employed to
achieve this selective heating. Induction and flame
techniques generally are used in high-volume pro-
duction applications and are easiest to control. To
obtain rapid heating for selective tempering by
immersion in lead or salt baths, it usually is neces-
sary that the bath temperature be considerably
higher than the desired tempering temperature.
Consequently, the immersion time becomes the
controlling factor in obtaining the desired temper-
ing results. Lead, because of its higher rate of heat
transfer, is more effective than salt.
Tools of a variety of types and sizes are exam-
ples of parts that require two or more hardness
ranges, and must be selectively tempered to meet
these requirements. For instance, a forged pipe
wrench handle made of AISI 4053 steel was fully
hardened and then tempered for 1 hour at 675 F
(355 C) to produce an overall hardness of 47 to 52
HRC. This hardness proved ideal for the wrench
teeth, but did not permit sufficient toughness for
the handle. Therefore, the handle was tempered
further, selectively, by induction heating for 1
minute at 900 F (480 C) to produce a hardness of
4Ot046HRC.
Prompt TelDpering
Because of their carbon and/or alloy contents,
some steels are susceptible to cracking if they are
permitted to cool to room temperature (or even
worse, if allowed to cool below room temperature,
as on a cold winter morning), immediately follow-
ing the quenching operation. Susceptibility to
cracking will be accentuated by certain design fea-
tures such as abrupt changes in section, sharp cor-
ners, poorly located holes and recesses. Accord-
ingly, steels that are known to be crack sensitive
should be removed from the quench while they are
still warm, but have been allowed to nearly com-
pletely transform. It is essential that transforma-
tion be nearly complete before placing the part in
the tempering fwnace, or cracking will result.
Differences of opinion exist regarding the mini-
mum and maximum temperatures for parts to
reach before tempering, but it is reasonably well
agreed that the ideal minimum is about 125 F (50
C), or just above that temperature at which a part
can be held in bare hands without burning. The

safe maximum temperature that should exist be-
fore tempering is about 200 F (95 C). At tempera-
tures higher than 200 F (95 C) there is always dan-
ger of insufficient transformation. Martempering
(marquenching) is an exception to the above.
Steels that are known to be sensitive to cracking
due to becoming too cold before tempering include
AlSI 1060, 1070, 1080, 1095, 1340, 4150, 4340,
52100, 6150, 8650, 9850, type 440C stainless steel,
and most of the high-carbon tool steels, and high-
carbon high-alloy tool steels.
Other carbon and alloy steels are generally less
sensitive to this type of delayed quench cracking
but may crack as a result of part configuration or
surface defects. These include AISI 1040, 1050,
1137, 1144, 4047, 4132, 4640, 8632, 8740 and
9840. Some steels such as AISI 1020, 1038, 1132,
4130, 5130 and 8630 are not sensitive; in general,
this is true of carbon steels containing not more
than 0.40070 C and of alloy steels containing not
more than 0.35070 C.
The following note (or one very similar) is com-
monly found on heat treating specifications.
Remove from quenching medium at 120 to 180 F
(50 to 80 C) and. traDSfer to the tempering furnace
as soon as possible; at least within 1 hour.
Selection of TeJDpering
Temperature
In many instances, temp,eratures used for tem-
pering are established through specifi~tions, but
when only mechanical properties are specified, the
correct temperature must be developed.
When there is some flexibility in selection of the
tempering temperature, some guidelines or impor-
tant rules exist which are listed and described as
follows:
1. Nearly all quench-hardened parts should be
tempered; at least at 250 F (120 C), and pref-
erably at 300 F (150 C) or higher.
2. It usua,Ny is good judgment to use a temper-
ing temperature as high as permitted without
sacrificing required hardness. For example,
there is seldom any significant loss of hard-
ness in tempering at 300 F (150 C), which is
the reason this temperature is so widely used
21
for tempering of parts to retain the as-
quenched hardness, or nearly so. However,
if a few Rockwell hardness points can be
sacrificed it usually is good practice to tem-
per at a slightly higher temperature. For
example, 350 F (175 C) instead of 300 F
(150 C).
3. For steels.. that have marked secondary hard-
ening characteristics, the tempering tempera-
ture that should be selected is based on the
peak of secondary hardening. For tempering
characteristics of specific steels, see Metals
Handbook, Vol. 2, eighth edition or data
from the steel manufacturers.
4. Service temperatures to which parts will be
exposed should be known; then a tempering
temperature should always be selected that is
at least as high and preferably higher than
the maximum service temperature.
Experimental Selection
of Temperature
Although much technical knowledge and many
formulas have been developed for selecting the
correct tempering temperature, there are instances
in heat treating shops where the tempering temper-
ature must be determined by experimentation.
Otherwise, parts may be tempered outside the
specified hardness range.
For example, parts made of AlSI 4140 steel
arrive for heat treating to a hardness of 35' to 40
HRC. Section sizes of the parts do not exceed 1 in.
(25 mm). In a conventional manner the parts are
oil quenched from 1550 F (845 C). Now, what tem-
pering temperature should be used to meet the
specificaton of 35 to 40 HRC? From the curve in
Fig. 8-7, it is seen that tempering at 950 F (510 C)
would be about right for developing a hardness of
35 to 40 HRC. However, the heat treater knows
there are variables so that values from curves can-
not always be depended on completely. Therefore,
one approach frequently used is to take two or
three of the parts and temper them at 875 to 900 F
(470 to 480 C) for about ~ to ~ hour, then cool
them and check their hardness. If the hardness is
high, then increase the furnace temperature and
try again until the optimum temperature is devel-
oped. Another approach, and sometimes simpler,
 22
would be to heat the entire load at, say, 850 F (455
C); that is, well below the estimated temperature,
and after ~ hour remove two or three parts and
check their hardness. If hardness is high, then
increase temperature of the fwnace in small incre-
ments' until the proper hardness is achieved, then
-shut off the furnace, because it must be remem-
bered that time at temperature will continue to
decrease hardness, but to a lesser degree compared
with temperature.
This experimental approach can be tedious, but
often it is preferable to taking a chance on over-
tempering. For example, if a batch of parts is tem-
pered at 975 F (525 C) without prior testing, they
could come out below the minimum specified
hardness. Under these conditions they must be
annealed and completely reprocessed, all of which
is costly, entails a time delay, wastes energy and
may result in an excessive amount of scrap.
Normal Variation. The hardness ~ ... tempered
parts is not always the same, although they were
made from the same mill heat of steel and heat
treated at the same time. To begin with, variation
in testing the same part will be at least plus or minus
one point on the Rockwell C Scale (testing machine
variable). In addition to testing variables, there are
other variables, all of which add up to a significant
total.
Normal variations which can be expected in
commercial practice are demonstrated in Fig. 8-19.
In Fig. 8-19a, data are presented for variation in
tempered parts from a single mill heat, -and like-
wise for AISI 4140 at a high~: hardness level in
Fig. 8-19b~ In Fig. 8-19c, all the variables, includ-
ing heat-to-heat, comprise a spread of about four
points on the Rockwell C Scale. Therefore, in se-
lecting tempering temperatures, some allowances
should be made for normal process variation.

~ 400 70
~ 4140 4140 G) ES0100
~ 350t--- Tempered 2 hr _ J-ti:if~~~-tempered 2 hr ~
en at 1125 F atl075 F ~ 65

i 300t---~·_··~···.~
...-~._..~._...~
.....~-..~.- - - - I I---~-------t~
J
=250
~ S~cffi~tl~~~ j~o~~~~~~~~~~~~~~~~-~
~
] I '-Specification ~ Specification range
~200 ~~e (o) (b) ~~~~~~~~_~_~_~_~~~_(c_)
5 10 5 10 5 10 15 20 25 30 35 40
Batch number Batch number Batch number

(a) Valve bonnets, 3 in. (76 mm) in diameter, made of 4140 steel from one mill heat. Parts were heated at 1600 F (870 C),
oil quenched, and tempered at 1125 F (60S C) to a hardness specification of 255 to 302 HB. (b) Valve segments made of
4140 steel from one mill heat. Section thickness varied from ~ to 1 in. (13 to 26 mm). Parts were heated at 1600 F (870
C), oil quenched, and tempered at 1075 F (580 C) to a hardness of 321 to 363 HB. (c) Needle rollers made of E50100 steel
wire varying in diameter from 0.086 to 0.103 in. Data represent 36 batches of parts (nine mill heats of steel) tempered to a
hardness specification of 60 to 64 HRC. Each batch represents a minimum furnace load of 600 lb.

Fig. 8-19. Variations in hardness after production tempering two a/loy steels.
 MATERIALS ENGINEERING INSTITUTE
C10L7
C40LS
C41L9
Test Series F Tempering of Steel Page 1

PLEASE PRINT YOUR NAME AND ADDRESS BELOW RETURN TO: ASM Intemationafi;
MATERIALS PARK, OH 44073-0002
NAME REGISTRATJON NO. _

STREET _

CITY STATE ZlP DATE _

MULTIPLE CHOICE

Place the appropriate letter (A, B, C, D) in the box. (Only one answer is correct
and each question has a value of10 points.)

1. As tempering of quench-hardened steel begins, the first metallurgical change in

crystal structure is: D
(A) bee to fec.
(B) bet to bec.
(C) fcc to bec.
(D) bec to hcp.

2. A carbon steel measures 28 HRC afterhardening, and tempering at 540°C (1OOOOf).

The amount of chromium addition (approximately) required to increase tempered
hardness by five points HRC - all otherconditions being the same as for the carbon
D
steel - would be:

(A) 0.38%.
(B) 0.19%.
(C) 0.60%.
(D) 0.95%.

3. In tempering of high-alloy steels, the precipitation of complex carbides and the

transformation of retained austenite result in: D
(A) secondary hardening.
(B) rapid softening.
(C) fonnation of bainite.
(D) formation of fine pearlite.

4. Considering the most widely used tempering temperature range of 150 to 620°C
(300 to 1150°F), the principal mode of heating in air is: D
(A) conduction.
(B) conduction with a radiation assist.
(C) radiation.
(D) convection.
 MATERIALS ENGINEERING INSTITUTE ASH
C10L7
C40LS
\
C41L9
Test Series F Tempering of Steel Page 2

A certain low-alloy steel is known to result in a hardness of50 HRC when tempered
5.
for 1 hour at 315°C (600°F); the calculated temperature (approximately) which will
result in the same hardness when tempered for 10 hours is:
D
(A) 270°C (515°F). (C) 305°C (580°F).
(B) 290°C (550°F). (0) 300°C (570°F).

6. For some applications, the use ofa molten lead as a tempering medium has a distinct
advantage because: D
(A) steel floats in lead.
(B) lead has very high thermal conductivity.
(C) lead can be used at any temperature.
(D) lead does not oxidize.

7. After carburizing, the as-quenched hardness of 1024 steel is approximately 63 HRC

at 0.25 mm (0.010 in.) below the surface; the hardness expected in the same location
after tempering at 425°C (800°F) would b~ approximately:
D
(A) 4OHRC. (C) 47HRC.
(B) 56HRC. (0) 42HRC.

8. In tempering at 482°C (900°F) ofa quench-hardened 0.27% carbon steel, the major
amount of stress relief is substantially complete after a time at temperature of: D
(A) 2 hours. (C) 1/2 hour.
(B) 8 hours. (D) 12 hours.

9. For a 0.40% carbon steel which shows an as-quenched hardness of 53 HRC,

tempering at 425°C (800°F) will result in a hardness of approximately: D
(A) 22HRC. (C) 38 HRC.
(B) 29HRC. (0) 48 HRC.

10. When a steel is subjected to double tempering, it means that:

(A) a second complete cycle identical to the first tempering cycle was used,
D
sometimes less 28°C (50°F).
(B) tempering time was increased by 100%.
(C) the normal tempering temperature was doubled.
(D) the previously tempered part was tempered at 150°C (300°F).
 ASM Intemational~ is a Society whose
mission is to gather, process, and
disseminate technical information.
ASM fosters the understanding and
application of engineered materials and
The Materials their research, design, reliable manu-
Information Society
facture, use, and economic and social
benefits. This is accomplished via a
unique global information-sharing
network of interaction among members
in forums and meetings, education
programs, and through publications
and electronic media.

Printed on DocuTech February 1994

 The Materials
Information Society

Heat Treatment
of Tool Steels

Course 10
Lesson, Test 14
Course 41
Lesson, Test 13
 Metric Conversion Factors
To convert from To Multiply by

in. DUn 25.4

in. m 25.4 x 10-'
mil J.Im 25.4
)Jin. J.Im 25.4
in.2 m2 6.45x1~
in.) m) 1.64 x 10-5
it. m 3.048 x 10-1
ft. 2 m2 9.29 x 10-2
ft.3 m3 2.831 x 10-2
oz. g 2.834 x 101
lb. kg 4.536 x 10-1
Btu J 1.054 x 103
Btullb. - of Jlkg - K 4.18 X 103
Btulft • hr • of W/m·K 1.730
10 inJin.JOF 1~ mml111I11?C 1.8
psi Pa 6.895 x 1()3
psi kPa 6.895
ksi kPa 6.895 x 1()3
lcsi MPa 6.895
ksi - in. l12 MPa-ml12 1.099
ksi-.Jin. MPa-vm 1.099
o1-f gf 28.4
lbf kg{ 4.536 x 10-1
lbf N 4.448
lbf- ft. N·m(orJ) 1.356
Ibflin.2 kgf/cm2 14.223
Ibflin. 3 kgflm3 2.768 x tOC
IbJft. 3 kg/m3 16.019
IbJin.' g/em3 2.768 x 1()l
tbJin. 3 kglm3 2.768 x tOC
gal (U.S. liquid) L 3.785
gal (U.S. liquid) m3 3.785 x 10-3
IbJgal gIL 119.826
ft/gal mIlL 748
of °C (OF - 32)/1.8
of K (OF + 459.67)/1.8
°C of (OC - 1.8) + 32
°C K °C+ 273.15
K °C °C-273.15

Multiple and Submultiple Units

1018 exa 1(t'l ••••••••••••••••••••••••••• deci
101' ••••••••••••••••••••••••••• peta 1(t'2 centi
1012 tera 1(t'3 •••••••••••••••••••••••••• milli
1(JJ •••••••••••••••••••••••••••• giga 1~ ......................•. micro
1()6 ••• ~ mega 1()-9 nano
1()3 kilo 1(t'12 •••••••••••••••••••••••••• pica
1()2 •••••••••••••••••••••••••• hecto 1(t'lS ••••••••••••••••••••••• femto
101 ••••••••••••••••••••••••••• deb 1(t'13 ••••••••••••••••••••••••••• atto

Abbreviations
J Joule m meter Pa ................•..•. pascal
kgf•..... kilogram force mm millimeter K................•...•. kelvin
L liter N newton W •.............•....•.... watt

Editor
Philip D. Harvey
Metals Engineering Institute, American Society for Metals

e 1981 American Society for Metals

This material may not be reproduced in whole or in part in any form whatsoever. Fully protected by copyright.
 Heat Treatment of Tool Steels
A definition of tool steel can best be reached
indirectly by considering how it differs from other
metals in the steel classification. Even then the
emphasis depends somewhat on the viewpoint of
the individual supplying the definition. To the
steel manufacturer, tool steel represents an extra
high quality product, not always used exclusively
for tools, which is closely controlled and inspected
throughout its manufacture, and is made and sold
in relatively small lots. The tool designer or tool-
maker, on the other hand, thinks of tool steel as a
group of special grades which, after proper heat
treatment, will provide the unique combinations of
properties required for various tool and die appli-
cations. Finally, the metallurgist might explain the
unique properties of tool steels as a group by point-
ing out that many generally tend to have excess
carbides in their structure" for improved wear
resistance.
Summarizing the above viewpoints, tool steel
might be defined as a group of special steels having
extra high quality and unique properties after heat
treatment which particularly adapt them for use in
tools and dies. In speaking of tools, it should be
remembered that we are primarily concerned with
industrial tools, such as lathe tool bits, milling
cutters, drills and forming dies for metals and
plastics, rather than ordinary household tools
which are frequently made from machinery steels.
Widely different properties are required of the
. steels from which tools are made. These greatly
divergent properties are secured by adding various
types and amounts of alloy additions to the steel
and subjecting the steel to suitable heat treatments.
In some cases, the total alloy content may be nearly
40070 of the total composition of the steel.
The presence of such large amounts of alloying
elements frequently has a pronounced effect on the
heating and cooling transformation characteristics
of the steel. Therefore, special procedures gener-
ally must be followed. Tool steels are steels of
special composition, require special heat.treatment,
and have special properties. Because of their great
importance and their special nature in regard to
heat treatment, this separate lesson is devoted
to them.
Types of Tool Steels
Table 15-1 lists compositions for the major types
of tool steels, classified into 14 groups. In most
instances, these steels are identified by both the
American Iron and Steel Institute (AlSI) and the
Unified Numbering System (UNS) designations.
Some of the steels listed are proprietary and do not
have an AISI designation. Through the years, tool
steels have been developed and assigned AISI
numbers. However, some are no longer listed by
 2
Table 15-1_ Classification and Approximate Compositions of Principal Types of Tool Steels
Identifying elements, per cent
AISI UNS C Mn Si Cr V W Mo Co Ni
Water-Hardening Tool Steels
W1 T72301 0.60-1.40 (a)
W2 T72302 0.60-1.40 (a) 0.25
W5 T72305 1.10 0.50
Shock-Resisting Tool Steels
51 T41901 0.50 1.50 2.50
52 T41902 0.50 1.00 0.50
55 T41905 0.55 0.80 2.00 0.40
56 T41906 0.45 1.40 2.25 1.50 0.40
57 T41907 0.50 3.25 1.40
Oil-Hardening Cold Work Tool Steels
01 T31501 0.90 1.00 0.50 0.50
02 T31502 0.90 1.60
06 (b) T31506 1.45 0.80 1.00 0.25
07 T31507 1.20 0.75 1.75
Air-Hardening Medium-Alloy Cold Work Tool Steels
A2 T30102 1.00 5.00 1.00
A3 T30103 1.25 5.00 1.00 1.00
A4 T30104 1.00 2.00 1.00 1.00
A6 T30106 0.70 2.00 1.00 1.25
A7 T30107 2.25 5.25 4.75 1.00 (c) 1.00
AS T30108 0.55 3.00 1.25 1.25
A9 T30109 0.50 5.00 1.00 1.40 1.50
A10 (b) T30110 1.35 1.80 1.25· 1.50 1.80
High-Carbon High-Chromium Cold Work Steels
02 T30402 1.50 12.00 1.00 1.00
03 T30403 2.25 12.00
04 T30404 2.25 12.00 1.00
05 T30405 1.50 12.00 1.00 3.00
07' T30407 2.35 12.00 4.00 1.00

L~w-Alloy Special-Purpose Tool Steels

L2 T61202 0.50-1.10 (a) 1.00 0.20
L6 T6'1206 0.70 0.75 0.25 (c) 1.50
Mold Steels
P2 T51602 0.07 2.00 0.20 0.50
P3 T51603 0.10 0.60 1.25
P4 T51604 0.07 5.00 0.75
P5 T51605 0.10 2.25
P6 T51606 0.10 1.50 3.50
P20 T51620 0.35 1.70 0.40
P21 T51621 0.20 1.20 (AI) 4.00

Chromium Hot Work Tool Steels

H10 T2081 0 0.40 3.25 0.40 2.50
H11 T20811 0.35 5.00 0.40 1.50
H12 T20812 0.35 5.00 0.40 1.50 1.50
H13 T20813 0.35 5.uu 1.00 1.50
H14 T20814 0.40 5.00 5.00
H19 T20819 0.40 4.25 2.00 4.25 4.25
Tungsten Hot Work Tool Steels
H21 T20821 0.35 3.50 9.00
 3
Table 15·1 (Continued)
Identifying elements, per cent
AISI UNS C Mn Si Cr V W Mo Co Ni
H22 T20822 0.35 2.00 11.00
H23 T20823 0.30 12.00 12.00
H24 T20824 0.45 3.00 15.00
H25 T20825 0.25 4.00 15.00
H26 T20826 0.50 4.00 1.00 18.00
Molybdenum Hot Work Tool Steel
H42 T20842 0.60 4.00 2.00 6.00 5.00

Proprietary Hot Work Tool Steels

6G 0.55 0.80 0.25 1.00 0.10 0.45
6F2 0.55 0.75 0.25 1.00 0.30 1.00
6F3 0.55 0.60 0.85 1.00 0.10 0.75 1.80

Tungsten High Speed Tool Steels

T1 T12001 0.75 (a) 4.00 1.00 18.00
T2 T12002 0.80 4.00 2.00 18.00
T4 T12004 0.75 4.00 1.00 18.00 5.00
T5 T12005 0.80 4.00 2.00 18.00 8.00
T6 T12006 0.80 4.50 1.50 20.00 12.00 -
T8 T12008 0.75 4.00 2.00 14.00 5.00
T15 T12015 1.50 4.00 5.00 12.00 5.00

Molybdenum High Speed Tool Steels

M1 T11301 0.80 (a) 4.00 1.00 1.50 8.00
M2 T11302 0.85-1.00 (a) 4.00 2.00 6.00 5.00
M3 Class 1 T11313 1.05 4.00 2.40 6.00 5.00
M3 Class 2 T11323 1.20 4.00 3.00 6.00 5.00
M4 T11304 1.30 4.00 4.00 5.50 4.50
M6 T11306 0.80 4.00 2.00 4.00 5.00 12.00
M7 T11307 1.00 4.00 2.00 1.75 8.75
M10 T11310 0.85-1.00 (a) 4.00 2.00 8.00
M30 T11330 0.80 4.00 1.25 2.00 8.00 5.00
M33 T11333 0.90 4.00 1.15 1.50 9.50 8.00
M34 T11334 0.90 4.00 2.00 2.00 8.00 8.00
M36 T11336 0.80 4.00 2.00 6.00 5.00 8.00

Ultrahard High Speed Tool Steels

M41 T11341 1.10 4.25 2.00 6.75 3.75 5.00
M42 T11342 1.10 3.75 1.15 1.50 9.50 8.00
M43 T11343 1.20 3.75 1.60 2.75 8.00 8.25
M44 T11344 1.15 4.25 2.00 5.25 6.25 12.00
M46 T11346 1.25 4.00 3.20 2.00 8.25 8.25
M47 T11347 1.10 3.75 1.25 1.50 9.50 5.00
Maraging Steels
Identifying elements, per cent
Type C Mn Si AI Ti Mo Co Ni
Grade 90 ............................ 0.03 max. 0.10 max. 0.12 max. 0.10 0.30 3.25 8.50 18.00
Grade 110 ..........................•. 0.03 max. 0.10 max. 0.12 max. 0.10 0.50 4.85 7.75 18.00
Grade 125 ............................. 0.03 max. 0.10 max. 0.12 max. 0.10 0.70 5.00 9.00 18.00
(a) Other carbon contents may be available.
(b) Contains free graphite in the microstructure to improve machinability.
(c) Optional.
 4
AISI because they are less commonly used and
their availability uncertain.
The steels are grouped according to their major
fields of application such as hot work, high speed
(cutting), spock resisting, etc. From this table it is
apparent that a variety of tool steels are available
for any particular job. Selection of the proper steel
for a specific job frequently is a difficult problem
because several different steels can be ·used, each
having slightly different properties and each re-
quiring somewhat different heat treatment proced-
ures. Anyone of the several steels may be satisfac-
tory for the job, but usually there is one steel
which, because of its combination of processing
characteristics and developed properties, is out-
standing for the specific application. Often selec-
tion is based on the number of parts that are to be
produced, or termed life cycle of the tool.
Properties and Characteristics
In an effort to simplify the selection of tool
steels, Table 15-2 has been prepared.
This table lists various characteristics of impor-
tance in selecting tool steel along with the relative
suitability of the various tool steel grades when
such characteristics are required of the tool. Briefly
the various characteristics considered important in
selecting a tool steel are as follows:
Nondeforming Properties. Nondeforming is a
term applied to tool steels that are air hardened
and thus deform less than water or oil quenched
tool steels. It does not infer that absolutely no
deformation occurs." A nondeforming property is
important for several reasons. Steels that show
little deformation or size change in hardening may
be machined closer to the finished size before hard-
ening than steels that show greater deformation.
Since machining of the unhardened steel is much
less expensive than removing metal from the hard-
ened steel, the advantages of a nondeforming steel,
with its smaller machining allowance, are evident.
Furthermore, the removal of larger amounts of
metal from a hardened part subjects the part to
possible cracking due to the heat generated. Also,
it is frequently difficult to correct for distortion in
an intricate tool.
Safety in Hardening. This factor is of extreme
importance in hardening tool steels where the
design and alloy considerations frequently make
cracking a problem. Those steels hardenable by
slow cooling rates (oil, polymers and air) provide
greater safety in hardening than steels hardenable
only by faster cooling rates, stich as result from
brine or water quenching. However, the oil or
polymer and air hardening steels must be properly
processed, for if such steels are water or brine
quenched, they will be even more susceptible to
breakage than a steel designed for water quenching.
Hot Hardness of High Speed Steels. These steel~
are characterized DY their ability to retain a high
hardness while operating at moderately elevated
temperatures such as would result during high
speed turning, hot piercing or forging operations.
This is sometimes referred to· as red hardness or
hot hardness. The red hardness of high speed steels
has been associated with the slowness with which
the transition carbides, M 2 C, grow at temperatures
below 1020 F (550 C).
Wear resistance is a self-explanatory term and is
a desirable property for any tooling application,
although it is far more important for some appli-
cations than for others.
Machinability. The ratings commonly given for
machinability are for steels in the annealed condi-
tion prior to hardening. For the grades of tool
steels having high alloy content and quantities of
excess carbide it is natural that they are difficult to
machine. However, there do exist differences in
.machinability which could well influence the selec-
tion of a steel, particularly where fine finish or
intricate design are concerned. Some grades of
tool steels are available with free graphite and free-
machining additives such as sulfur or lead.
. Temperature Required for Hardening. Because
of the limitations of available equipment, the hard-
ening range may be a factor in the selection of tool
steels, particularly when very high hardening
temperatures are involved. Decarburization and
scaling, more frequently associated with high tem-
peratures, are also problems to be considered.
Depth of Hardening. In many instances, service
conditions dictate the required hardenability for a
given tool. As a rule no more hardenability than
necessary should be specified. Use of steels having
greater hardenability than is actually required is
 5
Table 15·2. Approximate Comparison of Principal Types of Tool
Steels on the Basis of Some Metallurgical Characteristics
" Nonde- Resistance W Quench- Hardening Resist-
AISI forming" Safety in Tough - to Soften- ear Machin- Tempera- Depth of ance to
Harden-
Class. Prop- ing ness ing Effect Resist- ability Me:~ ture Hardening Decarbu-
erties of Heat ance Range F rization

w Poor Fair Good Poor Fair to Best Water 1400-1650 Shallow Best
Good
o Very Very Fair Poor Good Good Oil 1400-1625 Medium Good
Good Good
A Best Best Fair Fair Very Fair Air 1450-1800 Deep Good
Good to Fair
D Best Best Poor Fair Best Poor Air 1750-1900 Deep Fair'
W*Poor W*Poor Water 1550-1650 Poor
S O*Good Best Fair Fair Fair Medium
O*Fair Oil 1600-1700
H(a} Good Good Good Good Fair Fair Air 1650-2150 Deep Fair
H(b) Good Good Good Good to Fair to Fair Air, 1850-2350 Deep Fair
to Fair Very Good Oil
Good
M Good Fair Poor Very Very Fair Oil, Air, 2125-2275 Deep Poor
Good Good Molten
Salt
M(c) Good Fair Poor Best Very Fair Oil, Air, 2125-2275 Deep Poor
Good Molten
Salt
T Good Good Poor Very Very Fair Oil, Air, 2150-2375 Deep Good
Good Good Molten
Salt
T(d) Good Fair Poor Best Very Fair Oil, Air, 2250-2400 Deep Poor
Good Molten
Salt
P Fair Fair Good Poor Good Fair Water Carburize Shallow Good
1450-1550
L Good Good Very Fair Fair Fair Oil 1500-1700 Medium Fair
Good

·W indicates water quenched (a) Chromium Hot Work (c) Molybdenum High Speed with Cobalt
·0 indicates oil quenched (b) Tungsten Hot Work (d) Tungsten High Speed with Cobalt

not only poor economy, but may lead to cracking
or distortion or both.
Resistance to Decarburization. Consideration
of susceptibility to decarburization is important
since it determines how much material must be
removed from the surface of the steel before or
after hardening. In a case where the steel is suscep-
tible to decarburization, prior machining allow-
ance and both preheating and proper protection
must be supplied for the steel. Equipment limita-
tions may thus enter into consideration (note right-
hand column in Table 15-2).
Short Versus Long Production Runs. In many
cases, an important consideration in selecting a
t~ol steel for a given application is its expected
length of service. For example, a blanking die is
required, but it is estimated that no more than 100
pieces will be needed and a repeat order is improb-
able. Under such conditions a steel such as 01 may
be adequate. However, for the same blanking
operation, assuming 'continued production, a tool
steel having greater wear resistance, such as D2,
would be a more logical choice.
Atmosphere Protection
In selecting an atmosphere that will protect the
surface of tool steels against addition or depletion
of carbon during heat treatment, it is desirable to
choose one that requires no adjustnlent of compo-
sition to suit· various steels'. An ammonia-base
atmosphere (AGA class 601) meets this require-
ment and has the advantage of being sufficiently
 6
reducing to prevent oxidation of high-chromium Table 15·3. Ranges of Dew Point Temperature
steels. In the range of dew points generally found for Various Austenitizing Temperatures Used
in this gas (-40 to -60 F or -40 to -SO C), there is for Hardening Tool Steels (Furnace Dew Point;
no serious depletion of carbon, because the decar- AGA Class 302, Endothermic Atmosphere) (a)
burizing action is slow, and any loss of carbon at Furnace temperature, Dew point range,
F C F C
the surface is partially replaced by diffusion from
the interior. For applications in which high super- 1425 775 45 to 55 7 to 12
1475 800 45 to 55 7 to 12
ficial hardness is important, a carburized surface 1525 835 30 to 45 - 1 to 7
can be obtained by the addition of about 1% meth- 1575 855 25 to 35 - 4 to 2
ane to the atmosphere. Although ammonia-base 1600 870 22 to 30 - 5 to - 1
atmosphere costs more than endothermic gas, this 1700 925 20 to 30 - 7 to - 1
seldom becomes an important item of cost, because 1750 955 20 to 30 - 7 to - 1
1825 995 20 to 30 - 7 to - 1
tool treating furnaces generally are comparatively 2200 1205 5 to 10 -16 to -12
small and therefore require a correspondingly 2300 1260 o to 10 -18 to -12
small quantity of gas. An ammonia-base atmo- (a) For short times at temperature.
sphere is not recommended for very high austeni-
tizing temperatures, because of the undesirable that wrapping in a reflective foil greatly inhibits
reactions with hydrogen. rate of heat transfer by radiation so that longer
heating cycles are required compared with heating
Endothermic-base atmospheres are often used the same tool without wrapping.
for the protection of tool steel during heat treat-
ment. Suggested ranges of dew point for an AGA Vacuum atmospheres provide an excellent
class 302 endothermic atmosphere when used for means of protection for heat treating virtually any
hardening some common austenitizing tempera- tool steel. The degree of vacuum needed depends a
tures are listed in Table 15-3. great deal on the temperature range used and re-
quirements in terms of surface conditions. It is
Carbonaceous Liners. For small furnaces and recommended that the practice used for a specific
for austenitizing temperatures of about 1700 F application be suggested by the manufacturer of
. (925 C) to those temperatures used for heating the particular vacuum furnace, because vacuum
high speed steel, the use of expendable muffle equipment may vary. For quenching tool steels
liners made from a special carbonaceous material other than air hardening grades, integrated liquid
provides a satisfactory protective atmosphere. No quenching tanks are provided in the furnace design.
control is required, and it is simple to renew the
liner after it is no longer usable. This type of atmo- Molten Salt Baths. Molten salts of various com-
sphere generation is not sUItable for austenitizing positions are well adapted as normalizing and
the water-hardening and low-alloy oil hardening austenitizing mediums for tool steels. However,
types because it is decarburizing in this tempera- the same salts are not adaptable to all temperature
turerange. ranges. In general, salts composed of 3:pproxi-
mately 70070 barium chloride and 30010 soaium
Wrapping in Foil. In shops where a protective chloride are used for the working temperature of
atmosphere is not available, the practice of wrap- about 1300 to 1700 F (705 to 925 C), which accom-
ping s~all tools in stainless steel foil has proved modates a large portion of the carbon and low
successful for preventing decarburization and alloy tool steel compositions. For steels that re-
scaling during austenitizing. The chromium in the quire austenitizing temperatures of 1700 F (925 C)
stainless steel foil preferentially combines with and higher, a salt having a barium chloride content
oxygen, thus protecting the tooL This type of pro- of 92 to 96070 and a sodium chloride content of 4 to
tection can be used at all austenitizing temperatures 8070 is usually used. Such a mixture is virtually
(except tungsten-type high speed steel), although insoluble in water so that if the workpieces are air
for the temperatures used for austenitizing the cooled or oil quenched the adhering salt is ex-
molybdenum-type high speed steels it is common tremely difficult, if not impossible, to remove.
practice to use a dOll.hIe thickness of foil. In using The~efore, when tools are austenitized in salt they
this technique, it should be remembered, however, are quenched in salt which serves to wash the high-
 7
Table 15·4. Recommended Heat Treating Practice for Water-Hardening Tool Steels
Temperature, F Hardness
(based on carbon content) after treatment Procedure
Normalizing
1500 (0.60 to 0.75 C) Heat through uniformly; hold for 15 min
1450 (0.75 to 0.90 C) (light sections) to 1 hr (heavy sections),
1600 (0.90 to 1.10 C) then air cool.
1600 to 1700 (1.10 to 1.40 C)
Annealing
1360 to 1400 (0.60 to 0.90 C) 156 to 201 Bhn Heat through uniformly; hold for 1 to 4
1400 to 1450 (0.90 to 1.40 C) hr(a); furnace cool to 950 F at 40 of per
hr, then air cool.
Hardening(b)
1450 to 1550 (0.60 to 0.80 C) Rockwell C 65 to 68 .Hold at austenitizing temperature for 10 to
1425 to 1550 (0.85 to 1.05 C) 30 min; quench in water or brine (very
1400 to 1525 (1.10 to 1.40 C) small pieces may be oil quenched).
(a) Holding times vary from about 1 hr, for light sections and small furnace charges, to about
4 hr, for heavy sections and large furnace charges. (b) For large tools and tools with intricate
sections, preheating at 1050 to 1200 F is recommended.

temperature salt from the work and replace it with
a salt more soluble in water, thus permitting ·easier
and more efficient cleaning of "the work prior to
tempering.
For tools that cannot be ground after hardening,
or for tools that require the best possible surface
condition and the maintenance of sharp edges, salt
bath heating will give good results. With correct
operating conditions, tools can be produced with a
minim~ of carburization, decarburization and
scale. Three salt baths are generally used for aus-
tenitizing high speed steels: preheating, high-tem-
perature, and quenching baths. The function of
the quenching bath is to equalize the temperature
as well as to assure a clean surface after heat
treatment.
All the commercially used salts (except those
intentionally made carburizing) are mildly oxidiz-
ing to steel when they are used at recommended
austenitizing temperatures and in the open air. The
oxides produced by heating steel in molten salt are
largely soluble; hence, the steel surface is not vis-
ibly oxidized after heating. However, the accumu-
lation of oxide in the molten salt progressively
makes the salt more strongly decarburizing, and
for this reason baths must be periodically rectified.
Many salt baths can be rectified with either sili-
con or carbon. Baths in which the electrodes pro-
trude above the surface require daily rectification
with either ferrosilicon or silicon carbide. The
silicon combines with the dissolved metallic oxides
to form silicates, which settle to the bottom of the
pot in the form of a heavy, viscous sludge, which is
removed. Submerged-electrode baths can be recti-
fied by introducing a graphite rod into the molten
bath. The graphite reduces any oxides to metal,
which adheres to the rod. The metal can be scraped
off and the rod re-used.
Water-Hardening Tool Steels
Nominal compositions of the three principal
water-hardening tool steels are listed in Table 15-1;
and as indicated in the table, they can be obtained
with varying carbon contents. Types containing
0.90 to 1.0070 carb~n are most often used. Carbon
content and hardening depth markedly affect heat
treating temperatures, as indicated in Table 15-4
and Fig. 15-1, which outline heat treating practices
for these steels.
Water-hardening tool steels have a very high
critical. cooling rate; thus, they usually require
quenching in an aqueous medium in order to ex-
ceed .their critical cooling rates. However, thick-
ness of section and austenitizing temperature both
affect cooling rate. Oftentimes parts having thin
sections can be satisfactorily quenched in oil as
described in the following:
In one plant, it was desirable to harden small-
diameter punches in oil to reduce breakage and
consequent downtime of the presses. A study was
made to determine the maximum diameters of
water hardening tool steels that could be fully
hardened to a minimum of 60 HRC by oil quench-
ing. Results of this study, indicating the relati~n
 8

u..
0
(J)
'--
....0::J 1500
'-
Q)
Q.
E
{E 1400

1300 A,

1200

0 0.40 0.80 1.20 1.60

Percent Carbon
Fig. 15-1. Recommended normalizing~ annealing and hardening temperatures for water-hardening tool steels.

between austenitizing temperature, type of steel, uniform structure. Recommended normalizing

and punch diamter are shown in Fig. 15-2 temperatures are given in Table 15-4 and Fig. 15-1;
as indicated, optimum temperature varies with
It was determined that a greater degree of uni- . carbon content.
formity was obtained if the punches were normal-
ized prior to hardening (1600 F or 870 C for I
punches up to ~ in. or 6.35 mm in diam; 1650 F or '2 Water-hardening tool steels
900 C for. punches over ~ in. in diam). As indi- ~ Oil quenched to
cated in Fig. 15-2, austenitizing temperature varied Rockwell C 60 min

from 1450 to 1650 F (790 to 900 C), depending c:

.~ i3 t----+---+---+---+--_+__~
upon punch diameter. The punches were austeni- o
~ 5
tized by being heated vertically in a neutral salt ~ii
bath. They ~ere also quenched vertically, in a o
compounded oil containing additives. The quench- f/) .!. 1---+---+---+--2'::
~ 4
c:
ing oil was ~aintained at 120 to 140 F (50 to 60 C) ~

.~ 3 t----+--+-""-:
and circulated up and around workpieces at 50 gal .=i6
.per min. I L.-~
i
I
Normalizing. Except in special cases where expe- ii~~,--..--...~~_.........""","-,"""",--_,--~_--,
rience has proved it to be beneficial (as in the above 1450 1500 1550 1600 1650
case history), normalizing is not recommended for Austenitizing temperature, F
the water-hardening grades as received from the Fig. 15-2. Maximum section thickness of three classes
supplier. Normalizing is, however, recommended oj water-hardening tool steels that will develop mini-
for these steels after forging or before reheat treat-. mum hardness oj 60 HRC when oil quenched from
ment, for refining the grain and producing a more various austenitizing temperatures.

Decarburization during air cooling will be mini-
mized if parts are heated and cooled in a protective
atmosphere or heated in a neutral salt bath. Parts
heated in salt are protected during the air cooling
period by the film of salt that adheres to their sur-
faces when they are removed from the salt bath.
After parts have cooled, the film of salt can be
easily removed (except from recesses such as tapped
holes) by a water rinse.
Annealing. Water-hardening (W) type steels are
received from the supplier in the spheroidized
annealed condition. Thus, annealing by the user is
usually unnecessary. Annealing is applied to forged
or cold worked carbon tool steel to soften it for
easier machining, relieve residual stress, and pro-
duce a structure suitable for hardening. Annealing
may be done in an atmosphere furnace (provided
the furnace is of a type that can be cooled slowly to
below 1000 For 540 C), in a vacuum or in an ordi-
nary air furnace after the piece has been protected
against surface decarburization by being packed in
a suitable container with a protective material.
Protection against decarburization may be ob-
tained also by copper plating the surface or by
applying a surface-protecting paint. (Not all of
these paints are equally effective, and some are
difficult to remove; the prospective user should
investigate such a paint on a sample part during a
typical operating procedure and then inspect the
treated sample part for decarburization.)
When the steel has been placed in a pack to min-
imize surface reactions, a general rule of thumb is
to allow the assembly to soak at temperature for
one hour per inch of pack cross section. Work
should then be cooled in the furnace at a rate not
exceeding 40 F (22 C) per hour, to 950 F (510 C).
Below this temperature, cooling rate is not critical.
Hardness after annealing should be in the range of
156 to 201 HB. Precise temperature control is.
required to avoid carbide decomposition to graph-
ite, which may occur at higher temperatures.
Stress relieving prior to hardening is sometimes
employed to minimize distortion and cracking.
The procedure consists of heating the work to 1200
to 1325 F (650 to 715 C) and slow cooling. Usually,
stress relieving of water-hardening tool steel is
lilnited to complex or severely cold worked parts.
9
In most instances, stress relieving after harden-
ing and grinding is not employed. Periodic stress
relieving of tools that have been in service will
reduce the stresses imposed by such service, and
extend service life. Temperatures used for this
purpose should never exceed those used for tem-
pering the steel after hardening.
Austenitizing temperatures for the W type steels
may vary from 1400 to 1550 F (760 to 845 C), as
shown in Table 15-4. Sometimes somewht higher
temperatures are used (Fig. 15-2). Hardenability
increases as austenitizing temperature increases.
The optimum time at austenitizing temperature is
from 10 to 30 minutes. Preheating is unusual ex-
cept for very large tools or those with intricate
cross se~tions.
If surfaces are to be protected against scaling or
decarburization, an atmosphere or vacuum fur-
nace, lead bath or salt bath is required. It is partic-
ularly important to protect shallow-hardening
steels against scaling and decarburization. Severe
scaling can interfere with heat transfer during
quenching and slow down the required high rate of
cooling. Decarburization will produce a soft sur-
face on any tool steel, but in a deep hardening steel
it can be ground off until the underlying hard high-
carbon area is reached. Grinding a shallow-hard-
ening steel will frequently expose the soft core.
If scaling or decarburization will be removed by
grinding after heat treatment, austenitizing the
water hardening tool steels may be achieved in a
slightly oxidizing atmosphere, which can be ob-
tained by proper adjustment of the burners in
direct firing. Endothermic atmospheres are also
used, but close control is ,essential to match the
carbon potential of the atmosphere with the sur-
face carbon content of the workpiece.
Salt baths are also widely used and often pre-
ferred over other heating mediunls for austenitiz-
ing the W-type tool steels. Neutral .salts for the
required heating range are readily available.
Quenching. To produce m~mum surface hard-
ness as well as depth of hardness in water-harden-
ing tool steels, it is e~sential that they be cooled
rapidly from the austenitizing temperature. In
most instances, water or a brine solution consist-
ing of 10070 NaCI (by weight) in water is used.
Occasic;>nally, for an even faster quench, an iced
 10
Table 15·5. Recommended Heat Treating Practice for Shock-Resisting Tool Steels
Annealing Hardness
Tempering
Cool ing Annealed Holding Quenched
Tempera- Temperature, F
rate, hardness, time, Quenching hardness, Temper- Hardness,
Steel Normalizing ture, F (a) °F/hr (b) Brinell Preheat Austenitize min. medium Rockwell C ature, F Rockwell C
51 Not ree. 1450-1500 40 183-229 1650-1750 15-45 0 57-59 400-1200 58·40
52 ·Not rec. 1400-1450 40 192-217 1200 (c) 1550·1650 5-20 B,W 60-62 350- 800 60·50
55 Not rac. 1425-1475 25 192-229 1400 1600·1700 5-20 0 58-61 350- 800 60-50
56 Not rac. 1475-1525 25 192-229 1400 1675-1750 10-30 0 56-60 400- 600 56·54
57 Not rac. 1500·1550 25 187-223 1200·1300 1700·1750 15-45 (d) A,O 60-61 400·1150 57-45
(a) Lower limit of range should be used for small sections; upper limit for large sections. Holding time varies from about 1 hr for light sections
and small furnace charges to about 4 hr for heavy sections and large charges; for pack annealing, hold for 1 hr per in. (25 mm) of pack cross sec-
tion. (b) Maximum. Rate is not critical after work is cooled to about 950 F (510 C). (c) Preferable for large tools, to minimize decarburization. (d) For
open furnace heat treatment. For pack hardening, hold for V2 hr per in. (25 mm) of pack cross section.

brine solution is employed. Cooling rate is a func-

tion of size of workpiece as well as of quenching
medium; for this reason, small pieces can be
quenched in oil. Polymer quenchants are used for
rates between water or brine and oil.
Tempering. Water-hardening tool steels should
be tempered immediately after quenching, prefer-
ably before they reach room temperature (about
120 F or SO C is optimum). Salt baths, oil baths
and air furQaces all offer satisfactory heating me-
diums for tempering. However, working tempera-
tures for both oil and salt are limited; the minimum
for salt is about 325 F (165 C), and ~he maximum
for oil is usually about 400 F (205 C).
All parts made of these steels should be tempered
at temperatures not lower than 300 F (150 C). One
hour at temperature is usually adequate; additional
soaking is not desired because time at temperature
will further lower hardness.
Tools should be placed in a warm (200 to 250 F
. or 95 to 120 C) furnace immediately after quench-
ing and then be brought to the tempering tempera-
ture with the furnace. This is particularly neces-
sary when quenched tools are being accumulated
for tempering in a single batch. Allowing quenched
tools to stand at room temperature or placing them
in a cold furnace may lead to cracking. Except for
large pieces, the work will heat at about the same
rate as the furnace. The low temperatures used in
tempering eliminate the need for atmosphere con-
trol. The resistance to fracture by impact initially
increases with tempering temperature·to about 360
F (180 C) but falls off rapidly to a minimum at
about 500 F (260 C). This is known as 500 F (260
C) embrittlement. For tools subjected to impact
loading, tempering temperature should be selected
to give an optimum combination of hardness and
impact resistance.
Shock-Resisting Tool Steels
These steels may be obtained with several varia-
tions in composition, for specific applications; for
example, 81 steel is available with 0.30 or 0.500/0
Mo or with up to 0.90070 Si. The user of these non-
standard compositions should obtain from the
manufacturer information as to the modifications
required in heat treatment, or select a heat treat-
ment recommended for the shock-resisting tool
steel of standard compositIon that most closely
resembles the modified steel. The latter procedure
should be followed only after the treatment has
been tried on test samples.
Normalizing and Annealing. Normalizing is not
recommended for any of the shock-resisting tool
steels. Annealing of the high-silicon types 52, 85
and 56 can be critical because they are susceptible
to graphitization as well as decarburization. An-
nealing these types at temperatures higher than
those indicated in Table 15-5 may produce a softer
structure, but it will also increase the danger of
graphitization. The silicon types should not be
soaked at temperature. Surfaces should be pro-
tected against decarburization by heating in a pro-
tective atmosphere or a vacuum furnace, by the
use of pack annealing, or by the application of
proprietary paints.
Pack annealing is done by surrounding the parts
with a protective material inside a clos~d container,
heating the container to a recommended tempera-
ture, and slow cooling. Dry silica sand usually is
satisfactory for type 51, and a combination of new
and used carburizing compound usually is satisfac-
tory for S2 and S5. Burned-off cast iron chips,
spent pitch coke, lime and mica are sometimes
used also. Cast iron chips decrease in carbon con-

tent and should not be used indefinitely; lime and
mica should be used carefully, because they are
insulators. Excessive thicknesses of a protective
material should not be packed around parts, be-
cause this complicates handling and lengthens
heating time. Wrapping parts tightly in brown
paper before surrounding them with packing ma-
terial helps to keep the surfaces clean.
Proprietary paints are available that are intended
to protect steel surfaces from decarburization dur-
ing annealing. The use of such paints is simpler
than the use of a pack anneal, but not all of these
paints are effective. Moreover, considerable diffi-
culty may be experienced in removing such paints
after heat treatment. The prospective user should
test any such paint on a sample of steel prior to
adopting it in practice.
Stress relieving prior to hardening is seldom re-
quired for shock-resisting tool steel, except for
extremely intricate parts of widely varying section
thickness to minimize distortion and cracking and
parts subjected to excessive stock removal, to re-
lieve stresses induced by machining. Treatment of
such parts, which involves no austenite transfor-
mation, consists of heating them to 1200 F or 675.
C (soaking should be avoided), furnace cooling to
about 950 F (5~O C), and then removing them from
the furnace to cool in air.
Stress relieving of tools after tempering is sel-
dom done. In some instances, however, increased
tool life' has been obtained by removing tools from
service and stress relieving them (at a temperature
no higher than the original tempering temperature)
before returning them to service.
Austenitizing. Temperatures for austenltlzlng
the shock-resisting tool steels vary from 1550 to
1750 F (845 to 955 C). Preheating is not manda-
tory, but as indicated in Table 15-5, it is sometimes
desirable for large tools, to minimize distortion,
shC'rten time at the austenitizing temperature, and
speed up production.
These steels may be austenitized in electric or
fuel-fired furnaces or in salt or lead baths. Gener-
ally, for austenitizing temperatures below 1fOO F
(870 C), a slightly oxidizing atmosphere minimizes
scale and decarburization, whereas above 1600 F
(870 C) a reducing atmosphere is required. If a
11
semimuffle fuel-fired furnace is used, the required
atmosphere can be obtained at low cost by adjust-
ment of burners. However, if electrically heated or
full-muffle fuel-fired furnaces are used, a pre-
pared atmosphere from an external source is
required.
Neutral salt baths offer a practical means of
heating all of the S steels for austenitizing. A salt
mixture of approximately 700/0 barium chloride
and 30070 sodium chloride is generally preferred
for the higher silicon types such as S2, 85 and 56,
whereas a salt composed of approximately 85070
barium chloride and 150/0 sodium chloride is com-
monly preferred for the lower silicon types such as
81 and 57.
If atmosphere furnaces or neutral salt baths are
not available, the shock-resisting steels can be
heated in a pack of material such as burned pitch
coke or cast iron chips. Packing mediums must be
free of oil or other contaminants. Before being
placed in the pack, tools should be wrapped with
heavy brown paper, to prevent packing material
from adhering to them as they are removed for
quenching.
Types S2 and 55 should be quenched almost as
soon as they reach the austenitizing temperature;
types SI and 57 are soaked at temperature for 15
to 45' min before being quenched (Table 15-5).
Types SI and 87 have the highest hardenability of
these steels. The other types, although lower in
hardenability than 51 and S7, are higher in harden-
ability than the W steels. 57 is air hardening in
sections up to about 4 in. (101 mm) in thickness.
Tempering. All of the shock-resisting steels resist
softening from tempering to a greater degree com-
pared with carbon tool steels. However, the alloy
content of the shock-resisting steels is not suffi-
cient to cause any significant amount of secondary
hardening. The effect of tempering temperature on
the as-quenched bardness of three types of shock
resisting steels is given in Fig. 15-3. Also illustrated
in the figure is the effect of variations in composi-
tions on hardness.
TooJ<; made of shock-resisting steel should be
tempered immediately after quenching, or crack-
ing is likely to result, especially if they are quenched
in water or brine.
 12
70
SI 52 55

4i
3 Steels 9. and 10
~ 50
o
c::
45 ... 0Steel 7
40 '--_010..-_"'- ----.1
• Steel 8

AQ 200 400 600 800 AQ 200 400 600 800 AQ 200 400 600 800
Tempering temperature, F Tempering temperature, F

.--Steel~ Composition, % ~Quenching~

No. Type C Si WorMo Cr V Temp, F Medium

1 81 ...... 0.43 2.00W 1.30 0.25 1750

2 51 ...... 0.53 2.00W 1.65 0.25 1650
3 51 ...... 0.50 2.75W 1.25 0.20 1700
4 51 ••...• 0.55 2.50W 1.50 0.35 1700
5 81 .•••.. 0.50 0.75 2.50W 1.15 0.20 1750 Oil
6 81 ...... 0.58 0.95 2.25W 1.25 0.25 1700 Oil
7 82 .•••.. 0.50 1.10 0.50Mo 0~20 1575 Water
8 82 .•...• 0.50 1.10 0.50Mo 0.20 1650 Oil
9 55 •..•.. 0.50 1.60 . O.25Mo 1600 on
10 55 •..... 0.60 2.00 O.40Mo 0.28 0.20 1650 Oil

Fig. 15-3. Effect of tempering temperature on hardness of shock-resisting tool steels.

Surface treatments such as carburizing and car-
bonitriding are often applied to SI so as to increase
wear resistance for certain tools such as punches.
Such surface treatments are not usually used for
other grades of shock-resisting steels.
Oil-Hardening Cold
Work Tool Steels
.Recommended heat treating practices are sum-
marized in Table 15-6.
Normalizing is desirable and oftentimes neces-
sary for parts that have been forged or heated
previously to temperatures much higher than the
proper austenitizing temperature, because it pro-
"duces a more uniformly refined grain structure.
Work should be held at temperature for 15 minutes
to 1 hour, depending on section size; prolonged
soaking is not desirable. When workpieces are to
be hardened after normalizing, precautions must
be taken in order to avoid decarburization during
normalizing. If tools are to be subsequently ma-
chined, annealing is recommended in preference to
normalizing.
Annealing. Finished or semifinished tools made
from the type 0 steels should be protected to mini-
mize carburization or decarburization. This can be
accomplished by the use of dry exothermic furnace
atmospheres. More often, however, it is accom-
plished by pack annealing in a protective material,
such as clean cast iron chips or 6- to 8-mesh spent
pitch coke.
Type 01 steel may also be cycle annealed for
spheroidizing, as suggested in Table 15-7.
Stress Relieving. In most instances, stress reliev-
ing of finished tools prior to final hardening does
not noticeably lessen distortion du~ing hardening.
If extreme dimensional accuracy after hardening is
required, tools should be stress relieved after rough
machining but before final light machining. A
recommended stress-relieving treatment consists
of heating to 1200 to 1250 F (650 to 675 C), hold-
ing at temperature for one hour per inch of thick-
ness, and then air cooling.
Preheating of the type 0 steels will minimize dis-
tortion that may take place in hardening, especially
when austenitizing is done in a liquid bath. Open
furnaces can be used for preheating if some scale is
 13
Table 15-6. Recommended Heat Treating Practice for Oil-Hardening Cold Work Tool Steels
Annealing Hardening
Tempering
Normalizing Cooling Annealed Holding Quenched
Temperature, F
temperature, Temperature, rate, hardness, time, Quenching hardness, Temper· Hardness,
Steel F (a) F (b) °F/hr(c) Brinell Preheat Austenitize min. medium Rockwell C ature, F Rockwell C
01 1600 1400·1450 40 183-212 1200 1450·1500 10·30 0 63-65 350-500 62-57
02 1550 1375·1425 40 183·212 1200 1400·1475 5·20 0 63·65 350·500 62-57
06 1600 1410-1450 20 183-217 1450·1500 2· 5 0 63·65 350-600 63-58
07 1650 1450-1500 40 192-217 1200 1450-1525 10-30 W 64-66 350-600 63·58
1550·1625 10·30 0 64·66 (d) 350-550 64·58
(a) Holding time, after uniform through heating, varies from about 15 min. for small sections to about 1 hr for large sections. Work is cooled
from temperature in still air. (b) Lower limit of range should be used for small sections; upper limit for large sections. Holding time varies from
about 1 hr for light sections and small furnace charges to about 4 hr for heavy sections and large charges; for pack annealing, hold for 1 hr per
inch of pack cross section. (c) Maximum. Rate is not critical after cooling to below 1000 F. (d) Sections larger than 11/2 in. will be softer.

not objectionable, but if scale-free hardening is Martempering. If dimensional control is critical,

mandatory, some type of protection is required:
salt baths, painting with proprietary compounds,
or wrapping in stainless steel foil are methods
commonly used. Endothermic atmospheres are not
recommended at 1200 F (650 C) because they are
not safe.
Austenitizing. Workpieces that have been pre-
heated may be austenitized in the same furnace or
transferred to another furnace depending largely
on availability of equipment. Decarburization and
scaling can be effectively minimized in liquid salt
or lead baths, and in furnaces with controlled at-
mospheres (such as endothermic gas, dissociated
ammonia, and ~gon or other inert gases). How-
ever, in all of these there is some danger of decar-
burization if conditions are not controlled. Oxides
in the molten baths or excess water vapor in the
various gases will cause decarburization. All type
o steels may be austenitized in semimuffle fur-
naces if packed in protective materials such as
spent pitch coke and clean cast iron chips.
If salt baths are used, a salt mixture comprised
of barium chloride and sodium chloride, or one of
barium chloride, sodium chloride and calcium
chloride is satisfactory. The supplier of the specific
salt can furnish procedures for keeping the bath
rectified.
Quenching. The optimum temperature range for
quenching baths consisting of conventional oils is
100 to 140 F (38 to 60 C); agitation is recommended.
Quenching oils that contain additives (fast oils)
increase the cooling rate of the steel and permit
more latitude in the operating temperature of the
bath. Tools may be quenched in these oils at 180 F
(80 C), without loss of hardness.
martempering offers advantages. In martemper-
ing, the work is quenched in a bath of oil or molten
salt that usually is held about 25 to 50 F (15 to 28
C) above the M s temperature, and is held in the
bath long enough to allow the work to attain sub-
stantially equalized temperature throughout. The
work is then removed from the bath and air cooled.
The slow cooling through the martensitic transfor-
mation range permits the transformation of auste-
nite to martensite to take place uniformly through-
out the piece, thus minimizing distortion.
Tempering. The type 0 steels should be tem-
pered immediately after quenching (preferably
before they quite reach room temperature). These
steels usually are not tempered below 250 F (120 C)
or above 1000 F (540 C); the most commonly used
temperature range is from 350 to 400 F (175 t~ 205
C). Tempering times vary with section size. Often,
a time at temperature of one hour per inch of
thickness (minimum dimension of heaviest section)
or per inch of diameter, with a minimum of one
hour, is used.
Conventional tools made from the type 0 steels
are seldom subjected to multiple tempering or
subzero treatment. However, for some special
Table 15-7. Cycle Annealing Treatments
for Four Types of Tool Steel
Steel Treatment
01 ... Heat to 1350 F, hold for 4 hr; heat to
1440 P, hold for 2 hr; cool to 1275 F,
hold for 6 hr; air cool.
A2 ... Heat to 1650 F, hold for 2 hr; cool to
1400 P, hold for 6 hr; air cool.
A6 ... Heat to 1500 P, hold for 2 hr; cool to
1200 F, hold for 6 hr; air cool.
P2 ... Heat to 1650 P, hold for 2 hr; cool to
1425 P, hold for 6 hr; air cool..
 14
Table 15-8. Recommended Heat Treating Practice for Medium-Alloy Air-
Hardening and High-Carbon, High-Chromium Cold Work Tool Steels
Annealing Hardening
Tempering
Normalizing Cooling Annealed Quenched
Temperature, F Temper- Hardness.
temperature, .Tempera- rate, hardness, Holding Quenching hardness,
Steel F (a) ture, F (b) of Ihr (c) Brinell Preheat Austenitize time, min. medium Rockwell C ature. F Rockwell C
Medium·Alloy Air·Hardening Cold Work Tool Steels
A2 Not rec. 1550-1600 40 (d) 201-229 1450 1700-1800 20·45 (e) A 62-65 (1) 350-1000 62-57
A3 Not rec. 1550-1600 40 207-229 1450 1750-1850 25-60 (e) A 350-1000 62-57
A4 Not rec. 1360-1400 25 (g) 200-241 1250 1500-1600 15-90 A 61-64 (1) 350- 800 62-54
A6 Not rec. 1350-1375 25 217-248 1200 1525-1600 20-45 A 59-63 (1) 300- 800 60-54
A7 Not rec. 1600-1650 25 (d) 235-262 1500 1750-1800 30-60 (e) A 64-67 (1) 300-1000 67-57
A8 Not rec. 1550-1600 40 192-223 1450 1800-1850 20-45 (e) A 60-62 (1) 350-1100 60-50
A9 Not rec. 1550-1600 25 212-248 1450 1800-1875 20·45 (e) A 56-58 (1) 950-1150 56·35
A10 1450 1410-1460 15 235-269 1200 1450-1500 30-60 A 62-64 (1) 350- 800 62-55

High.Carbon, High·Chromium Cold Work Tool Steels

02 Not rec. 1600-1650 40 217-255 1500 1800-1875 15-45 (e) A 64 400-1000 61-54
03 Not rec. 1600-1650 40 217-255 1500 1700-1800 15·45 (e) 0 64 400-1000 61-54
04 Not rec. 1600·1650 40 217-255 1500 1775-1850 15·45 (e) A 64 400-1000 61-54
05 Not rec. 1600-1650 40 223-255 1500 1800-1875 15-45 (e) A 64 400-1000 61-54
07 Not rec. 1600-1650 40 235-262 1500 1850-1950 30-60 (e) A 65 300-1000 65-58
(a) Holding time, after uniform through heating, varies from about 15 min. for small sections to about 1 hr for large sections. Work is cooled
from temperature in still air. (b) Lower limit of range should be used for small sections; upper limit for large sections. Holding time varies from
about 1 hr for light sections and small furnace charges to about 4 hr for heavy sections and large charges; for pack annealing. hold for 1 hr per
inch of pack cross section. (c) Maximum rate, to 1000 F unless footnoted to indicate otherwise. (d) To 1300 F. (e) For open furnace heat treatment.
For pack hardening, hold for Y2 hour per. inch of pack cross section. (f) Hardness varies with austenitizing temperature. (g) To 1200 F.

tools, such as gages, where dimensional stability is
critical, multiple tempering is desirable. In such
instances, the workpieces should be cooled to
below 150 F (65 C) prior to each retempering. Sub-
zero cooling to -100 F (-75 C) or lower is also
helpful in achieving dimensional stability.
Cold Work Tool Steels
Nominal compositions of medium-alloy air-
hardening cold work tool steels (group A) and
high-carbon, high-chromium cold work tool steels
(group D) are listed in Table 15-1. Recommended
heat treating practice for steels in both groups is
summarized in Table 15-8.
Normalizing is not recommended for any of the
A or D grades, except for type AID, which is a
graphitic tool steel.
Annealing. These steels usually are supplied in
. the annealed -condition by the manufacturer. They
should, however, be annealed after forging, and
prior to rehardening. Annealing is also required
for hardened or welded tools that are to be re-
worked.
Tools should be heated slowly arid uniformlv to
the annealing temperatue. Slow heating is particu-
larly important if a hardened tool is being annealed.
Cycle annealing may be used for some of the A
and D steels (see Table 15-7).
Stress Relieving. Tools made from A and D
steels that cannot be ground after hardening are
sometimes stress relieved after rough machining.
This is particularly advisable for delicate tools and
tools that vary markedly in cross section. Stress
relieving is used also on tools that are machined to
final shape, if these tools can be straightened after
stress relieving and before final heat treatment.
There is little advantage in stress relieving com-
pleted tools if they cannot be straightened prior to
hardening, because preheating will relieve stresses.
Recommended temperatures for stress relieving are:
A2, A7 ... _.. 1200 to 1250 F (650 to 675 C)
A4, A5, A6 .. 1250 to 1300 F (675 to 705 C)
D1 to D7 .... 1250 to 1300 F (675 to 705 C)
Usually, tools can be stress relieved at these
temperatures without surface protection. Tools are
commonly held at temperature for one hour per
inch (25 mm) of cross section (minimum of one
hour) and then air cooled.
Preheating. Steels of the A and D groups usually

are preheated before being austenitized for hard-
ening. Preheating reduces subsequent distortion in
the hardened parts by minimizing nonuniform
dimensional changes during austenitizir.. g. Pre-
heating simpler tools made of grades A4, A6 and
AIO can often be eliminated if they are austeni-
tized in a furnace instead of a liquid bath, because
these steels are austenitized at lower temperatures.
Holding time at temperature is usually one hour
per inch of maximum cross section. Preheating
temperatures of 1450 to 1500 F (790 to 815 C) are
used for tools made from A2, A3, A7, A8 or A9,
or from any of the D steels. For these higher tem-
peratures a liquid bath or a protective furnace
atmosphere is required in order to prevent scaling
and decarburization.
Austenitizing. Steels of the A and D groups can
be austenitized in neutral molten salt, or in various
types of gaseous atmospheres or vacuum furnaces,
packing media or stainless steel foil wrap. Because
of their lower austenitizing temperatures, types
A4, A6 and AIO may also be austenitized in moltep.
lead, or in open-fired furnaces with slightly oxidiz-
ing atmospheres. However, the latter methods are
not suitable for the other A or any of th.e D steels,
because of their higher austenitfzing temperatures.
Neutral salt baths of 70 to 85070 barium chloride
and 15 to 30070 sodium chloride are commonly
used for austenitizing the A and D steels. Suitable
rectification procedures recommended by the salt
supplier should be followed.
Atmospheres that are used for austenitizing the
A and D steels are endothermic gas and ammonia.
Endothermic atmospheres are used becau.:e this
atmosphere can be adjusted for the desired carbon
potential and controlled by the dew point of other
monitoring systems.
Stainless steel foil wrap is used extensively for
small parts. Vacuum is particularly suitable for
these steels because their air-hardening character-
istics permit slow cooling rates induced by back-
filling with gases.
Like the 0 steels, the A and D steels may be
packed and then austenitized in semimuffle fur-
na~es. The packing materials and heat treating
procedures employed are similar to those described
15
in the previous section on austenitizing of the 0
grades.
Steels of groups A and D must be held at their
austenitizing temperatures long enough to obtain
required carbide solution if they are to attain max-
imum hardness. However, hardening from exces-
sively high austenitizing temperatures will increase
the retained austenite. Retained austenite can be
decreased by multiple tempering or subzero cool-
ing (or both).
Quencbing. With the exception of D3, all of the
type A and D tool steels will attain maximum
hardness by cooling in still or slightly agitated air
or gas because they are all characterized by their
relatively high hardenability.
Depending on section size, hardenability, and
complexity of shape, the following methods are
used to obtajn increasingly accelerated cooling of
nominally air-hardening steels:
1. Cool in still air; that is, atmospheric air un-
disturbed by artificial circulation.
2. Cool in fan air; that is, the current of air
discharged from a fan.
3. Cool in air blast; that is, the discharge from
a high-pressure line.
4. Oil quench to black; that is, quench in oil
until the steel is below the temperature at
which it g~ows dull red, then cool to room
temperature in air.
5. Oil quench by conventional practice.
To minimize decarburization or scaling during
cooling from high austenitizing temperatures, air
is substituted for by gases such as endothermic
gas, nitrogen or argon.
Tempering practices for the A and D steels par-
allel those recommended for the 0 types described
in the precedin~ ,section. Most of these grades
exniolt seconaary hardening characteristics, thus
making double tempering desirable. Tempering is
usually begun when the work reaches a tempera-
ture of about 120 to 150 F (SO to 65 C). However,
these st~els retain some austenite at this tempera-
ture range. To maximize transformation of auste-
nite to martensite, cooling to room temperature,
or to subzero temperature, j~ sometimes applied.
However, subzero cooling increases the probabil-
 16
T ~~ I J I I
II A2 Other A steels
65 t - -
0 P 0 (') 4.
0

60
4L
() ~~ n ()
~~
~

U)
U)
Q)
c:
I r
I
n t" ,. .4~ A
•
~) 0
~~
()
0 0
~~ 4~ 4

I ~~ ~~. i 0 9
"E 4.

~ 55
I ! 4~ 4) 0 I ~>
() ~

I I • ()
u
I I I ~;\
4~ ()
~ I
I 0 ~~

~ 50
3
I I ! j) I .
o
0::: I
I;
l-
I
Austenitized
I L~

I II I
45 ~ Austenitized t '- 0 A4 1550F I
01700F • A5 1500 i
I

~. 1750 I I- 6 A6 1550 I i 1 i
40 •1800
I
A A7 1800
i
I
I
!
!
I
i ~

AQ 200 400 600 800 1000 1200 AQ 200 400 600 800 1000 1200
Tempering temperature, F Tempering temperature, F

Fig. 15-4. Effect oj tempering temperature on hardness of medium-alloy air-hardening tool steels.

ity of cracking during the cooling cycle. Usual the group 0 steels with an increase in tempering
practice is to begin tempering when parts reach temperature (Figs. 15-4 and 15-5), higher temper-
about 120 to 150 F (50 to 65 C) and then double or ing temperatures can be used for the A and D steels.
triple temper. Multiple tempering is effective in A minimum tempering temperature of 400 F (205
decreasing the amount of austenite retained in A C) is a common requirement for A2, A7 and D
and D steels and is a common practice in heat ·steels. Tempering temperatures as high as 1025 F
treating them. The general precautions and tem- (550 C) are frequently used, and still higher tem-
. pering practices outlined for 0 steels in the preced- peratures are used for special requirements.
ing section are followed for the A and D steels.
However, because most of the steels in groups A The data in Figs. 15-4 and 15-5 indicate that
'and D (except A4 and A6) soften less rapidly than certain steels (notably A2 and D2) exhibit higher

70
02_ "'---+---i--+-+--+---+-~!--+-~--+-0 3

65 I !
"--- ...
--
- -- ---
-- Air quenched Jrom 1850 F
I
~ 60
~~ - I
c ~~ ~ .~ r ~
~
o
.J:
/
v· - ~
\\
u 55
Q)
1900 F../·
V \~
/~~
~
~

g 50
7 ~
a::
1800 F

45
I \
~
40
AQ 20a... 400 600 800 1000 1200· AQ 200 400 600 800 1000 1200
Temperina temperature, F Tempering temperature, F

Fig. 15-5. Relation between tempering temperatures on hardness for D2 and D3 tool steels.
 17
Table 15-9. Recommended Heat Treating Practice for Low-Alloy, Special-Purpose Tool Steels
Annealing Hardening
Quenched
Tempering
Normalizing Cooling Annealed Austenitizing Holding hardness,
tem perature, Tempera- rate, hardness, tempera· time, Quenching Rockwell Temper· Hardness,
Steel F (a) ture, F (b) °F/hr (c) Brinell ture, F (d) min. medium C(e) ature, F Rockwell C
L2 1600-1650 1400-1450 40 163-197 1450-1550 10-30 W 63 350-1000 63-45
1550-1700 10-30 0 63
L6 1600 1400-1450 40 183-255 1450-1550 10-30 0 62 350-1000 62-45
(a) Holding time, after uniform through heating, varies from about 15 min. for small sections to about 1 hr for large sections.
Work is cooled from temperature in still air. (b) Lower limit of range should be used for small sections; upper limit for large sec·
tions. Holding time varies from about 1 hr for light sections and small furnace charges to about 4 hr for heavy sections and large
charges; for pack annealing, hold for 1 hr per inch of pack cross section. (c) Maximum. Rate is not critical after cooling to below
1000 F. (d) These steels are seldom preheated. (e) Typical average values; subject to variations depending on austenitizing tem-
perature and quenching medium.

hardness after being-. tempered at about 1000 F Low-Alloy Special-

(540 C) than after being tempered at temperatures
100 to 200 F (55 to 115 C) lower. This reversal in
the \lsual relationship is known as secondary hard-
ening. When tempering of residual austenite takes
place at higher temperatures, a complex carbide
tends to precipitate which renders austenite less
stable, enabling it te-transform to martensite on
cooling from the tempering temperature. The pre-
cipitation of the carbide and the formation of the
additional fresh martensite partially accounts for
the secondary hardening noted after the steel cools
from the tempering temperature. When a steel can
be tempered to the same hardness at more than one
temperature (for instance, D2 to 58 to 59 HRC), it
is advisable to select the highest tempering temper-
ature that will produce the desired hardness. This
~ill yield added toughness and may prevent tool
b'reakage in service.
The data presented in Figs. 15-4 and 15-5 indi-
cate that austenitizing temperature has a marked
effect on tempering characteristics.
Types A2 and A 7 and the D steels are occasion-
ally nitrided after being hardened and tempered.
This extremely hard case is often influential in
prolonging die life. Nitriding may be done either in
a salt bath or in an atmosphere of ammonia.
Higher tempering temperatures (950 to 1000 F or
510 to 540 C) are used on steels that are to be ni-
trided. Nitriding temperatures above the recom-
mended range of 950 to 1000 F (510 to 540 C) will
reduce hardness of the base metal and should not
be used. Austenitizing at a higher temperature
when hardening prior to nitriding will minimize
loss of hardness during nitriding of some D steels.
Purpose Tool Steels
There are two low-alloy special-purpose steels
presently available, types L2 and L6. Type L2 is
available with a carbon content ranging from 0.50
to 1.10070. Type L2 with the higher carbon content
is similar to commercial 52100 in composition, and
very often it is used instead of 52100 where the
extra quality provided by tool steel practice is
desired.
Steel L6 is an 0.70070 carbon, chromium, nickel,
molybdenum steel which is used for tool applica-
tions where a high degree of toughness is required;
intricate punches as one example. .
Recommended heat treating temperatures are
given in Table 15-9.
Normalizing should follow forging or any other
operation in which the steel has been exposed to
temperatures substantially above the transforma-
tion range.· The use of a protective atmosphere is
recommended.
Annealing must follow normalizing and precede
any rehardening operation.
Stress relieving prior to hardening may be ad-
vantageous for complex tools, to minimize distor-
tion during hardening. A common practice for
complex tools is to rough machine, heat to 1150 to
1200 F (620 to 650 C) for 1 hour per inch of cross
section, cool in air, and then finish machine prior
to hardening. .
Austenitizing. Salt baths, lead baths and various
types of atmosphere fumaces are all satisfactorY
 18
for austenitizing L2 and L6. If a salt bath is used, for plastic molds, although P4, P20 and P21 are
it should be a neutral type such as described for the also used for die-casting of the low-melting alloys.
type Wand 0 tool steels. Keeping the salt bath
properly rectified is of paramount importance. The wide variations in composition, method of
Preheating is seldom used prior to austenitizing forming the mold cavity, molding method, and
these steels. material to be molded are major influences on
choice of mold material as well as method of heat
Quenching. Oil is the quenching medium most treatment. The two most common methods of heat
commonly used for the L steels. Water or brine treating the mold steels are to preharden the steel
sometimes may be used for simple shapes, or for (or partially machined mold or die) to about 30 to
large sections of L2 (lower carbon range) that do 36 HRC, finish machine, and use at this hardness
not attain full hardness by oil quenching. level; and case harden by carburizing. Nitrided
molds have proved successful in some instances,
Tempering. Tools made of the L steels should but nitriding is not used extensively. When molds
be quenc.hed only to a temperature at which they
can be handled with bare hands (about 125 F or 50 Low-alloy special-purpose tool steels
C), and should be tempered immediately there- 70
aft,er; otherwise, cracking is likely to occur. 1 Ii L2 L6
~ 65 (0.50% C) I
The tempering characteristics of L2 and L6 are .go 60·Ii
Q.) I I! I' I ,
plotted in Fig. 15-6. For most applications, the L
steels are used at near-maximum hardness. It is ~ 55 /
/1525F, water
I !
I
! .-:-.. J-=T·-
l i :/
recommended that tools made of any of these low- 1500F,oi1 J
Q,)

~ 5 0 ~--,---oa.r-~-~
alloy steels be tempered at a minimum of 250 F U
o
!
i
(120 C). Double tempering is recommended re- 0:: 45 I
-t--_. "-7 _._-+ --
I • I
I
gardless of the tempering temperature used. 4 a '---"- --'
I •

i 1 :

700 300 500 700

Mold Steels Tempering temperature, =
Fig. 15-6. Relation oj hardness and tempering temper-
The principal use of the mold (type P) steels is ature for low-alloy special purpose tool steels.

70 I i
65 ~ ---J,_-+--_ I I i P4
Reheated for hardening -i-
after carburizing 1
60 ~-o-------~-..
I ! I
~ 55 i--~--4-4-~-~-~~~~::;::=¢'-+--i
Q)
c:
~ 50 t---j---+--r---+-~~-6-===O:==t:"""""""',.;ci--+--i
o
.J:;
u 4 5 ...-.--+-----+-~-
~
~
40 ~.....j---+--+--+--r---+---+---+---+-~~+T-~
u
&. 3 5 I----+--+--+---+----!--+---+--.:..-~~-+\---\-I
:3 a 1---+---+-+--+--;---+---+--+-1~-~--;-~
25 ~+__-+--_1__--4----+-__+_--+- 1__+---l._;.___+l
I
2 a L - - - - L - . - . . . . I . _ . . . l - - " -_ _..-......--J.._~....-.___'__...._.__l
AQ 200 40.0 600 800 1000 1200 AQ 200 400 600 800 1000
Tempering temperature, F Tempering temperature, F
(Left) Upper curve represents steel carburized at 1725 to 1750 F, removed from pack, cooled in
in hardwood charcoal at 1675 to 1700 F for 8 air and tempered.
hr, air cooled in pack, reheated at 1725 to 1750 (Right) Surface hardness after heating at
F, cooled in air and tempered. Middle curve temperature for 2 hr in carburizing compound,
represents steel carburized in cast iron chips oil quenching, and tempering.

Fig. 15-7. Tempering characteristics of carburized mold steels.

 19
Table 15·10. Recommended Heat Treating Practice for Mold Steels
Annealing Hardening (after carburizing)
Normalizing Cooling Annealed Austenitiz· Holding Quenched Tempering
temperature, Tempera· rate, hardness, Carburizing ing temper· time, Quenching hardness, Temper- Hardness,
Steel F (a) ture, F (b) °F/hr (c) Brinell temperature, F ature, F min. medium Rockwell C ature, F Rockwell C
P2 Not reqd. 1350-1500 40 103·123 1650-1700 1525-1550 15 0 62-65 350- 500 64-58 (f)
P3 Not reqd. 1350-1500 40 109·137 1650-1700 1475-1525 15 0 62-64 350- 500 64-58 (f)
P4 Not rec. 1600-1650 25 116·128 1n5-1825 1775-1825 15 A 62-65 350- 900 64-58 (f)
P5 Not reqd. 1550-1600 40 105-131 1650-1700 1550-1600 15 0, W 62-65 350- 500 64-58 (f)
P6 Not reqd. 1550 15 183·217 1650-1700 1450-1500 15 A, 0 60-62 350- 450 61-58 (f)
P20 1650 1400-1450 40 149-212 1600·1650 (d) 1500-1600 15 0 58-64 900-1000 37-28
P21 1650 Not recommended Hardened by solution treating and aging (e)
(a) Holding time, after uniform through heating, varies from about 15 min. for small sections to about 1 hr for large sections. Work is cooled
from' temperature in still air. (b) Lower limit of range should be used for small sections; upper limit for large sections. Holding time varies from
about 1 hr for light sections and small furnace charges to about 4 hr for heavy sections and large charges; for pack annealing, hold for 1 hr per
inch of pack cross section. (c) Maximum. Aate is not critical after cooling to below 1000 F. (d) When applicable. (e) Solution treatment: Hold at
1300 to 1350 F for 1 to 3 hr, quench in air or.-oil; approximate solution treated hardness, 24 to 28 HAC. Aging treatment: Reheat to 950 to 1025 F;
approximate aged hardness, 40 to 30 HRC. (f) Carburized case hardness.

are carburized or nitrided, the same procedures are
used as for production steels.
Heat treating practice for the mold steels is
summarized in Table 15-10. P21 is a special type
solution heat treated by the manufacturer and
delivered ready for the user to machine, age harden,
and place in operation.
Annealing. For some types the annealing tem-
perature is not extremely critical. A more impor-
tant factor is surface protection, especially if the
mold cavities will be formed by hubbing. If surfaces
are allow~d to carburize, even slightly, during
annealing, subsequent hubbing will be impaired.
Usually, parts are packed in a protective mate-
rial such as spent pitch coke and are held at anneal-
ing temperature only long enough to become
heated through; they are then cooled in the pack to
below 1000 F (540 C), after which they may be
removed from the pack. If hubbing is to follow, it
is usually preferable to use the lower side of the
annealing temperature range to minimize the dan-
ger of carburizing, even though annealing at the
higher side of the range will result in slightly lower
hardness. Atmosphere-controlled furnaces that
can be programmed for slow cooling can also be
used for annealing. For hubbing deep cavities, two
or more in-process anneals are sometimes required.
When cavities will be formed entirely by ma-
chining (sometimes a combination of hubbing and
machining is used), annealing usually is neither
necessary nor desirable, because slightly harder
structures machine more easily. Steels as received
from the manufacturer are usually suitable for
machining. If hardened molds require reworking,
they can be annealed as recommended in Table
15-10.
P2 and P3 mold steels both have sufficient
hardenability so that they can be fully hardened by
oil quenching. The operating hardness range may
vary from 54 to 64 HRC, but common practice is
to temper at about 600 F (315 C) to achieve a final
hardness of 54 to 58 HRC.
P4 steel is the most resistant to wear and to soft-
ening by tempering of all the steels in this group.
Because of these properties, it is commonly used
for injection molding of plastics that require high
curing temperatures and for dies used for die cast-
ing low-melting alloys. For the latter application, a
common practice is to carburize P4 in cast iron
chips to obtain a slight increase in carbon content
at the surface. The effect of carburizing practice,
as well as case and core hardness values after tem-
pering, is shown in Fig. 15-7.
Because of its high alloy content, P4 steel can be
hardened by air cooling. However, it is sometimes
quenched in oil to minimize scaling during cooling.
For use in plastic molds, the most common work-
ing range is 56 to 60 HRC, which may be obtained
by tempering the carburized and hardened molds
at 400 to 600 F (205 to 315 C), as shown in
Fig. 15-7.
PS steel, in which chromium is the only specified
alloying element, generally is about the same as P2
for ease of hubbing, and has a core strength equiv-
aleJ;lt to that of P3. After carburizing, a surface
hardness of 65 HRC can be achieved by ,water
 20
quenching, or slightly lower values by oil quench-
ing. Choice of quenching medium depends on
mold configuration, allowable distortion, and
required hardness. Hardness of 54 to 58 HRC is a
common working range; this can be obtained by
tempering at about 500 F (260 C).
P6 steel, because it can seldom be annealed to a
hardness of less than 183 HB (Table 15-10), is dif-
ficult to hub, and hence it usually is used for ma-
chine-cut cavities. It can be carburized by conven-
tional practice. Because of its hardenability, heavy
sections of P6 can be oil quenched to full hardness
from 1450 to 1500 F (790 to 815 C). The as-
quenched surface hardness is not quite so high as
for some other types, because the high nickel con-
tent of P6 promotes retention of austenite. Some
of this retained austenite is transformed in temper-
ing, with the result that after tempering, up to
about 500 F (260 C) the hardness will be little or no
lower than that obtained after quenching. By tem-
pering at 600 F (315 C), the most common working
hardness range (54 to 58 HRC) is obtained. In
some plants, a working hardness range of 58 to 61
HRC (which is obtained by tempering at 500 F or
260 C) is considered preferable.
no Steel. For injection molding of the general-
. purpose plastics or die casting of low-melting
alloys, P20 is usually used in the prehardened con-
dition. It is available at hardness levels of about
300 HB or slightly higher. In this condition, cavi-
ties are machined and the dies or molds placed in
service without further heat treatment. Annealed
molds or dies can be austenitized at 1550 to 1600 F
(845 to 870 C), oil quenched, and tempered at 1000
F (540 C), to obtain a hardness of about 300 HB.
Type P20 is often carburized for molds used in
compression molding, particularly for molding the
more abrasive plastics. Carburizing· temperatures
no higher than 1650 F (900 C) are recommended
for this steel, because higher temperatures may
impair polishability; otherwise, conventional car-
burizing practice is used, and molds may be
quenched in oil directly from the carburizing tem-
perature. The common working hardness range is
54t058 HRC.
Tempering characteristics for P20 carburized at
two different temperatures are given in Fig. 15-7.
This steel is sometimes nitrided for special appli-
cations. Conventional nitriding practice is em-
ployed. Before being nitrided, P20 should first be
quenched and tempered to about 300 HB as out-
lined above, and cavities should be machined;
following this sequence will assure freedom from
carburization or decarburization.
Type P21, containing nickel and aluminum as
the principal alloying elements, is an age-harden-
ing steel in the same sense as certain aluminum and
stainless steel alloys are age hardening. It may be
hardened by austenitizing and quenching. A tem-
pering treatment at 1250 F (675 C) acts as a solu-
tion treatment and reduces hardness to a point
where the material can be machined. A subsequent
aging treatment causes precipitation hardening as
a result of the formation of a nickel-aluminum
intermetallic compound.
Type P21 cannot be carburized but may be ni-
trided during aging. If the steel is held in a gas
nitriding furnace for 20 to 24 hours at 950 to 975 F
(510 to 525 C), a case depth of approximately
0.006 to 0.008 in. (0.15 to 0.20 mm) will result with
a face hardness of 94 HRI5N, which is equivalent
t070HRC.
Hot Work Tool Steels
. Table 15-11 summarizes the heat treating prac-
tices commonly employed for this composite
group of tool steels.
Normalizing. Because these steels as a group are
either partially or completely air-hardening, nor-
malizing is not recommended.
Annealing. Heating for annealing should be
slow and uniform to prevent cracking, especially
when annealing hardened tools. Heat losses from
.the furnace usually determine the rate of cooling;
large furnace loads will cool at a slower rate than
light loads. For most of these steels, furnace cool-
ing to 800 F (425 C) at 40 F (22 C) max per hour,
and then air cooling, will suffice.
For types 6F2 and 6F3, isothermal annealing
may be employed to advantage for small tools that
can be handled in salt or lead baths or for small
loads in batc~-:t.Y.Pe furnaces; ~owever" iso~e~~
annealing has no advantage over conventional
annealing for large die blocks or large furnace
loads of these steels.
 21
Table 15·11. Recommended Heat Treating Practice for Hot Work Tool Steels
Annealing Hardening
Normaliz-
Tempering (n)
ing tem- Cooling Annealed Quenched
Temperature, F Hardness,
perature, Tempera- rate, hardness, Holding Quenching hardness, Temperature,
Steel F (a) ture, F (b) °F/hr (c) Brinell Preheat Austenitize time, min. medium Rockwell C F Rockwell C

Chro~ium Hot Work Tool Steels

H10 Not rec. 1550·1650 40 192-229 1500 1850·1900 15-40 (d) A 56-59 1000-1200 56-39
H11 Not rec. 1550·1650 40 192·235 1500 1825-1875 15·40 (d) A 53-55 1000·1200 54·38
H12 Not rec. 1550·1650 40 192·235 1500 1825-1875 15·40 (d) A 52-55 1000·1200 55-38
H13 Not rec. 1550-1650 40 192·229 1500 1825-1900 15-40 (d) A 49·53 1000·1200 53-38
H14 Not rec. 1600·1650 40 207-235 1500 1850-1950 15-40 (d) A 55·56 1100·1200 47-40
H19 Not rec. 1600·1650 40 207-241 1500 2000-2200 2- 5 A,O 52·55 1000·1300 57-40
Tungsten Hot Work Tool Steels
H21 Not rec. 1600·1650 40 207·235 1500 2000-2200 2· 5 A,O 43·52 1100·1250 54·36
H22 Not rec. 1600-1650 40 207·235 1500 2000·2200 2· 5 A,O 48·57 1100-1250 52-39
Not rec. 1600·1650 40 1550 33·35 (e) 1200-1350 47-34
H23
H24
H25
Not
Not
rec.
rec.
1600-1650
1600-1650
40
40
212·255
217·241
207-235
1500
1500
2200-2300
2000-2250
2100-2300
2·
2·
2·
5
5
5
°
A,O
A,O
44·55
46-53
1050-1200
1050-1250
55-45
44-35
H26 Not rec. 1600-1650 40 217·241 1600 2150-2300 2· 5 A,D,S 63·64 1050-1250 58-43

Molybdenum Hot Work Tool Steel

H42 Not rec. 1550-1650 40 207·235 1350- 2050·2225 2- 5 A,O,S 54-62 1050-1200 60·50
1550

Other Alloy Tool Steels

6G Not rec. 1450-1500 40 (f) 197-229 Not 155Q.1575 o (g) 63 min (h)
reqd.
6F2 Not rec. 1440-1460 40 (j) 223·235 Not 1550-1600 o (g) 63 min (h)
reqd.
6F3 Not rec. 1400-1425 40 (k) 235·248 Not 1650-1700 A(m) 63 min (h)
reqd.
(a) Holding time, after uniform through heating, varies from about 15 min. for small sections to about 1 hr for large sections. Work is cooled
from temperature in still air. (b) Lower limit of range should be used for small sections; upper limit for large sections. Holding time varies from
about 1 hr for light sections and small furnace charges to about 4 hr for heavy sections and large charges; for pack annealing, hold for 1 hr per
inch of pack cross section. (c) Maximum rate, to 800 F unless footnoted to indicate otherwise. (d) For open furnace heat treatment. For pack harden-
ing, hold for 1/2 hr per inch of pack cross section. (e) Temper to precipitation harden. (f) To 700 F. (g) To 400 to 350 F, then air cool. (h) Temper
immediately. m Fo! isothermal annealing, furnace cool to 1200 F, hold for 4 hr, furnace cool to 800 F, then air cool. (k) For isothermal annealing
furnace cool to 1240 F, hold for 4 hr, furnace cool to 800 F, then air cool. (m) Cool with forced·air blast to 400 to 350 F. (n) Double tempering sug-
gested but not less than 1 hr at temperature for each temper. .

To minimize scaling and decarburization, small
parts are usually pack annealed, while large and
heavy die blocks are more commonly annealed in
controlled-atmosphere furnaces.
Packing material should preferably be spent cast
iron chips or spent pitch coke. Lime, sand or mica
is sometimes used, but unless such material is
mixed with a small amount of charcoal or other
carburizing material, the steel may be decarbu-
rized. The hot work steels must have a neutral
packing material, because they are extremely sus-
ceptible to both carburization and decarburization.
Stress Relieving. It is sometimes advantageous
to stress relieve tools made of hot work steel after
rough machining, but prior to final machining, by
heating them to 1200 to 1350 F (650 to 730 C). This
treatment minimizes distortion during hardening,
particularly for dies or tools that have major
changes in configuration or deep cavities. How-
ever, closer dimensional control can be obtained
by hardening and tempering after rough machin-
ing and prior to final machining, provided that the
final hardness obtained by this method is within
the machinable range.
Preheating prior to austenitizing is nearly always
recommended for all hot work steels except 60,
6F2 and 6F3. These three steels mayor may not
require preheating, depending on size and config-
uration of the workpieces.
Die blocks or other tools for open furnace treat-
ment should be placed in a furnace that is not over
500 F (260 C); work that is packed in containers
may be safely placed in furnaces at 700 to 1000 F
(370 to 540 C). Once the workpieces (or container)
have attained furnace temperature, they are heated
slowly and uniformly (at 150 to 200 F or 80 to 110
 22
C per hour) to the preheating temperature, and
held for one hour per inch of thickness (or per inch
of container thickness, if packed). Thermocouples
should be placed adjacent to the pieces in contain-
ers. Controlled atmospheres or other protective
means must be used above 1200 F (650 C) to mini-
mize scaling and decarburization.
Austenitizing. Rapid heating from the preheat-
ing temperature to the austenitizing temperature is
preferred for most types, particularly types H19
through H42. Time at austenitizing temperature
should only be sufficient to heat the work com-
pletely through; prolonged soaking is not recom-
mended.
The equipment and method employed for aus-
tenitizing are frequently determined by the size of
the workpiece. For tools weighing less than about
500 lb, any of the methods would be suitable.
However, larger tools or dies would be difficult to
handle in either a salt bath or a pack.
Tools or dies made of hot work steel must be
protected against carburization and decarburiza-
tion when being heated for austenitizing. Carbu-
rized surfaces are highly susceptible to heat check-
ing. Decarburization causes decreased strength,
which may result in fatigue failures; and on die-
casting dies, the molten casting metal will "weld
on" to decarburized surfaces and may cause wash-
out because of poor wear resistance of the decar-
burized surface. However, the principal detrimen-
tal effect of decarburization is to mislead the heat
treater as to the actual hardness of the die. In order
to obtain specified hardness of the decarburized
surface, the die is tempered at too Iowa tempera-
ture, and the die goes into operation at excessive
internal hardness and may break at the first appli-
cation of load.
Vacuum furnaces are well suited to hardening
these steels.
An endothermic atmosphere produced by a gas
generator is a widely used protective medium. A
dew point of 35 to 45 F (1 to 7 C) for an AGA class
302 atmosphere has been used for type Hll, H12
and H13 ·steels when austenitized at 1850 F
(1010 C).
Packing of work in spent pitch coke before heat-
ing it for austenitizing has been used extensively in
small shops where it has not been feasible to invest
in special equipment. This procedure i~ generally
used for small dies.
For small shops where sophisticated atmosphere
control equipment is not available, a common
practice for austenitizing small tools and dies is to
wrap them in stainless steel foil. This affords good
protection against scaling and decarburization.
Quenching Practice. Most of the hot work steels
will achieve full hardness by cooling in still air;
however, even with those types having the highest
hardenability, sections of die blocks may be so
large that insufficient hardening results. In such
instances an air blast or an oil quench is required
to achieve full hardness. Hot work steels are never
water quenched.
If blast cooling is used, air should be blasted
uniformly on the surface to be hardened. All air
must be dry. When being air quenched, dies or
other tools should not be placed on concrete floors
or in locations where water vapor m'ay strike them. __
. Gases are often substituted ror air to reduce decar-
burization and scaling.
Some of the hot work steels (especially the tung-
sten and molybdenum types) will scale consider-
ably during cooling to room temperature in air. An
interrupted quench reduces this scaling by elimi-
nating the long period of contact with air at ele-
vated temperature, but it also increases distortion.
The procedure is best carried out by quenching
from the austenitizing temperature in a salt bath
held at 1100 to 1200 F (595 to 650 C), holding in
the quench until the workpiece reaches the temper-
ature of the bath, and then withdrawing the piece
and allowing it to cool in air. An alternative, but
less precise, procedure is to quench in oil at room
temperature or slightly above and judge by color
(faint red) when the workpiece has reached 1100 to
1200 F (595 to 650 C); the workpiece is then quickly
withdrawn and permitted to cool to room tempera-
ture in air. While cooling, the workpieces should
be placed in a suitable rack, or be ~upported by
wires, in such a manner that air is permitted to
come in contact with all surfaces.
Steel H23 requires a different type of interrupted
quench, because ferrite precipitates rapidly in this
steel at 1100 F (595 C), and Ms is below room tem-
perature. Type H23 should be quenched in molten

salt at 325 to 375 F (160 to 190 C), and then air
cooled to room temperature. This steel will not
harden in quenching but will do so by secondary
hardening during the tempering cycle.
Parts quenched in oil should be completely im-
mersed in the oil bath, held until they have reached
bath temperature, and then transferred immedi-
ately to the tempering furnace. Oil bath tempera-
tures may range from 130 to 300 F (55 to 150 C),
but should always be below the flash point of the
oil. Oil baths should be circulated and kept free of
water.
Tempering. Hot work tool steels should be
tempered immediately after quenching, even
though sensitivity to cracking in this stage varies
considerably among the various types. For exam-
ple, air-quenched H23 may be safely "kept at room
temperature for several hours before tempering,
whereas most others, and especially 60, 6F2 and
6F3, are susceptible to cracking if they are cooled
substantially below 350"F or 175 C before tem-
pering.
Hot work steels usually are tempered in air fur-
naces of the forced-convection type. Salt baths are
used successfully for smaller parts, but for large,
complex parts, salt bath tempering may induce too
severe a thermal shock and cause cracking.
Multiple tempering has proved particularly ad-
vantageous for large or sharp-cornered die blocks
that are not permitted to reach room temperature
before the first tempering operation.
Multiple tempering ensures that any retained
austenite that transforms to martensite during the
first tempering cycle is tempered before a tool is
placed in service. Multiple tempering also mini-
mizes cracks due to stress originating from the
hardening operation.
Most of the hot work steels have secondary
hardening characteristics; H23 is the most pro-
nounced in this respect. As with A2 and D2 (dis-·
cussed previously), these secondary-hardening hot
work steels should be tempered at the highest tem-
perature at which the desired hardness can be
produced.
Surface Treatments. Although tools and dies
made from the hot work steels usually have suffi-
ci~nt hardness to perform the tasks for which they
were designed, they are occasionally surface hard-
23
ened to acquire improved resistance to wear or
heat for special applications. The two principal
processes that have been used for this purpose are
carburizing and nitriding.
Carburizing usually is limited to hot work steels
having a carbon content of 0.35070 or lower. Type
H12 can achieve a carburized surface hardness of
60 to 62 HRC. The carburized case should be shal-
low (for example, 0.015 in. or 0.38 mm max), or
severe embrittlement will occur. The greater the
thermal shock (or gradient) present in service -. as
in die casting - the shallower the case must be.
Gas or liquid nitriding is sometimes applied to
the hot work steels to increase resistance to heat or
wear, or both, and to increase service life. One
disadvantage is that it may accentuate heat check-
"ing. Hot work steels should be hardened and tem-
pered before being nitrided, but should be neither
decarburized nor carburized.
The quality and depth of the nitrided case are
influenced by the chemical composition of the
steel and by the time and temperature of nitriding.
The presence of nitride-forming elements such as
chromium and vanadium is helpful to the attain-"
ment of a satisfactory case. The fact that most of
the hot work steels reach a secondary hardening
peak ~hen tempered in the vicinity of 1000 F (540
C) is beneficial, because nitriding usually is accom-
plished in a range of 950 to 1000 F (510 to 540 C)
over a period of 15 to 24 hours. The nitrided 'case,
in addition to being very hard, may be brittle. Brit-
tleness increases with depth of case; hence, shallow
(0.003 to 0.008 in. or 0.076 to 0.2 mm) nitrided
cases are usually 'applied.
High Speed Tool Steels
Heat treating practice for all types of high speed
tool steels is summarized in Table 15-12. Normal-
izing of the high speed steels is not recommended.
Annealing. High speed steels must be fully an-
nealed after forging or before rehardening. To
minimize decarburization, pack annealing in tightly
closed containers is recommended. The packing
material can be dry sand or lime to which a small
amount of charcoal has been added; burned cast
iron chips also are satisfactory.
 24
Ta~le ..15.12. ~.commended Heat Treatlng'Practice for High Speed Tool St~els
Annealing Hardening
Temperature, F Tempering (d)
Cooling Annealed Holding Quenched
Tempera- rate, hardness, Austen- time, Quenching hardness, Temperature, Hardness,
Steel Normalizing ture, F (a) °F/hr (b) Brinell Preheat itize (c) min. medium Rockwell C F Rockwell C
Tungsten ·Hlgh Speed Tool Steels
T1 Not rec. 1600-1650 40 217·255 1500·1600 2300-2375 2-5 O,A,S 63-65 1000-1100 65-60
T2 Not rec. 1600-1650 40 223-255 1500-1600 2300-2375 2·5 OtA,S 64-66 1000-1100 66-61
T4 Not rec. 1600-1650 40 229-269 1500-1600 2300-2375 2-5 O,A,S 64-66 1()()()'1100 66-62
T5 Not rec. 1600·1650 40 235-285 1500·1600 2325-2375 2·5 O,A,S 64-66 1000-1100 65-60
T6 Not rec. 1600-1650 40 248-302 1500-1600 2325-2375 2·5 O,A,S 64-66 1000-1100 65-60
T8 Not rec. 1600·1650 40 229-255 1500·1600 2300-2375 2·5 O,A,S 64-66 1000-1100 65-60
T15 Not rec. 1600-1650 40 241·2n 1500-1600 2200-2300 2·5 O,A,S 65-67 1000-1200 68-63
Molybdenum High Speed Tool Steels
M1 Not rec. 1500-1600 40 207·235 1350-1550 2150-22:25 2·5 O,A,S 64-66 1000·1100 65-60
M2 Not rec. 1600·1650 40 212·241 1350-1550 2175-2250 2·5 O,A,S 65-66 1000-1100 65.60
M3 Not rec. 1600·1650 40 223-255 1350-1550 2200-2250 2-5 O,A,S 64-66 1000·1100 66.61
M4 Not rec. 1600·1650 40 223·255 1350-1550 2200-2250 2·5 O,A,S 64-66 1000-1100 66.61
M6 Not rec. 1600 40 248·277 1450 2150-2200 2·5 O,A,S 63-66 1000-1100 66.61
M7 Not rec. 1500-1600 40 217·255 1350-1550 2150-2225 2·5 O,A,S 64·65 1000·1100 66.61
M10 Not rec. 1500·1600 40 207·255 1350-1550 2150-2225 2-5 O,A,S 64-66 1000·1100 65.60
M30 Not ree. 1600-1650 40 235-269 1350-1550 2200·2250 2-5 O,'A,S 64-66 1000·1100 65-60
M33 Not rec. 1600-1650 40 235-269 1350-1500 2200·2250 2-5 O,A,S 64·66 1000·1100 65-60
M34 Not rec. 1600·1650 40 235-269 1350-1550 2200-2250 2·5 O,A,S 64·66 1000-1100 65.60
M36 Not rec. 1600·1650 40 235-269 1350-1550 2225·2275 2-5 O,A,S 64-66 1000·1100 65-60
M41 Not rec. 1600·1650 40 235·269 1350-1550 2175-2220 2-5 e,A,S 63·66 1000·1100 70-65
M42 Not rec. 1600·1650 40 235.269 1350-1550 2125-2175 2-5 O,A,S 63-66 950·1100 70-65
M43 Not rec. 1600·1650 40 248·269 1350·1550 2100·2150 2·5 O,A,S 63·66 950-1100 70-65
M44 Not rec. 1600·1650 40 248·285 1350-1550 2190·2240 2·5 e,A,S 63-66 1000·1160 70·62.,
M46 Not rec. 1600·1650 40 235-269 1350-1550 2175-2225 2-5 O,A,S 67-69
M47 Not rec. 1600·1650 40 235·269 1350-1550 2150·2200 2·5 O,A,S 65·70
ta) Pack annealing is recommended, for minimum decarburization. Steels should be held at temperature for 1 hr per inch of thickness o~ the
container. (b) Maximum. Rate is not critical after work (in pack, if employed) has been furnace cooled to 1200 F. (c) If steels are austenitized in a
salt bath, austenitizing temperatures should be 25° F lower than those in the ranges given. (d) Double tempering is suggested for not less than 1
hr at temperature for each temper. The M40s are triple tempered.

After the steel has reached the annealing temper-
ature range (Table 15-12), it should be held at tem-
perature for one hour per inch of thickness of the
container·and.should then be slowly cooled in the
furnace (at a rate not exceeding 40 F or 22 C per
hour) until it reaches a temperature of 1000 F (540
C), when a faster rate of cooling is permissible.
Preheating. In most high speed steels austenite
begins to form. at about 1400 F (760 C) so that
preheating slightly above this temperature will
minimize stresses that might be set up because of
,the transformation. If the prevention of partial
decarburization is important, a preheating temper-
ature of 1300 to 1450 F (705 to 790 C) may be
used. When this is not a problem, preheating at
1500 to 1650 F (815 to 900 C) is usually satisfactory.
Double preheating for high speed steel tools is
common practice, especially for large tools. The.
tools are preheated at about 1200 F (650 C) in one
furnace (no prepared atmosphere unless it is a salt
bath, in which the tool is automatically protected),
then transferred to another furnace at 1550 to 1600
F (845 to 870 C), preferably with some kind of
atmosphere protection, then finally to the high
heat furnace. Thermal shock is greatly minimized
by use of double preheating.
If a single preheat is used, the T type steels are
preferably preheated at 1500 to 1600 F (815 to 870
C). For the M types, preheat temperatures ranging
from 1350 to 1550 F (730 to 845 C) are commonly
used. The principal reason for using a lower pre-
heat temperature for the M grades is because they
are more vulnerable to decarburization than are
the Ttypes.
For all grades of high speed steels, common
practice is to preheat for twice the length of time
required at the austenitizing temperature for the
same workpiece. Accordingly, to ensure a uniform
flow of work, the capacity of the preheating instal-

68
en
~ -
c: 66 ",-
/
'"0
] 64
u V
/
~ 62
~
3
g 60
/'"
a:::
58
1900 2000 2100 2200
Austenitizing temperature, F
Fig. 15-8. Effect of austenitizing temperature on the
as-quenched hardness oj M2 steel.
lation is generally twice that of the austenitizing
installation.
Although preheating is recommended for all
high speed steels, small tools and those that do not
incorporate sharp notches or abrupt changes in
section, such as small tool bits and solid drill rod
blanks, may be placed directly into the' austenitiz-
ing furnace with reasonable safety. If consumable
carbonaceous muffles are used, the preheating
temperature must not exceed about 1200 F (650 C)
because the type of atmosphere they provide is
ineffective in preventing decarburization at con-
ventional preheating temperatures such as 1350 to
1550 F (730 to 845 C).
Austenitizing.. Accurate temperature· control is
required·in austenitizing high speed steel. Steels
containing about 3070 or more vanadium may be
held at the austenitizing temperature approximately
50070 longer than the lower-vanadium types. The
relatively pure vanadium carbide phase inherent in
the microstructure of the high-vanadium steels is
virtually insoluble at temperatures below the melt-
ing point and acts to restrict grain growth, thus
permitting longer soaking times without detriment.
80 r--..,.....-,---,----r-,.---,-......-----,--,.-.....---
70 t---t--+-...,....:=¥~~4---4---l-~-~
1000 1050 (100
Tempering temperature, F
Fig. 15-9. Effect of austenitizing and tempering tem-
peratures on impact strength of M2 high speed steel.
2~
T
However, the recommended austenitizing temp~t­
M2
atures for these steels should not be exceeded.
..............
""
Single-point tools intended for heavy-duty cut-
ting often can be effectivelyaustenitized at IS to 30
F (8 to 16 C) above the nominal austenitizing tem-
perature. The higher temperature increases alloy
solution, temper resistance, and hot hardness, but
2300 it also results in some sacrifice in toughness. To
impart added toughness, fine-edged tools, such as
taps and chasers, may be hardened at temperatures
25 to SO F (13 to 27 C) below the nominal austeni-
tizing temperature. Punches and dies that do not
require maximum hardness may be austenitized
for maximum toughness at temperatures 100 to
200 F (54 to 110 C) below the nominal temperature.
Other adjustments in austenitizing temperature
sometimes depend on the type of heating equip-
ment employed. Full muffle furnaces employing a
controlled atmosphere rich in carbon monoxide
are usually operated at the high end of the recom-
mended temperature range. Salt baths usually are
operated 30 to 50 F (16 to 27 C) below the top of
the range.
The effects of austenitizing temperature on the
as-quenched hardness of M2 steel is shown in Fig.
15-8. Below 2150 F (1175 C) M2 cannot develop
full hardness on quenching, because of insufficient
carbide solution. At temperatures above about
2250 F (1230 C) the as-quenched hardness of M2
decreases because of too much carbon and .alloy
solution and an excess of retained austenite in the
as-quenched steel.
Figure 15-9 illustrates the improved toughness
of M2, as measured by the Izod unnotched impact
test, that results from the use of lower-than-normal
austenitizing temperatures. The optimum means
for attaining maximum toughness in high speed
steel is through reduced austenitizing temperatures
rather than by full austenitizing and overtemper-
ing to an equivalent hardness level.
Protection against carburization, decarburiza-
tion and scaling of high speed steels may be pro-
vided by: (1) an endothermic atmosphere with a
Izod unnotched
low dew point (see Table 15-3); (2) use of carbona-
1150 1200 ceous liners in a muffle furnace; (3) dry hydrogen;
(4) dissociated ammonia; (5) vacuum; and (6)
wrapping the workpieces in a double thickness of
stainless steel foil.
 26
70 I

M2
I

I
Tempered
65
"-~""7 ~~ for -
....

.~/ /\' \~ " i'\,

,~, '- '.~
~ I""
~ ,, V O. lhr
en
en
CD
~~~
" "'....... ~ /
," ,,'
~..' /'" / / \ \ I
1\\ '\ \\ \\\,y- 07I
c: "-... '''',
'"e ,~ "~",,,, r~ V
"
~
.......
o
~...... - :~ ......
'~
--- d. "'---~ ~ \ ~
.c ~,.. 0 hr
..... _~ ... ~ \
(.) 60
\ \ \
\
\

\ \
~.
Vr.~hr
1

,-
\ \
\
I
55
100hr~N
,
\ \. \\ /2.5hr

\
I hY\ /fOr-
50
Austenitized at 2225 F
I I I
\ \{\
100 300 500 700 900 1100 1300
Tempering temperature, F

Fig. 15-10. Effect oj tempering temperature and time on hardness of M2 high speed steel.

Quenching. High speed steels, as a group, are
extremely high in hardenability; thus, they can be
fully hardened by quenching in air, oil or molten
salt. However, except for these workpieces, which
are air quenched between plates to keep them
straight, it is customary to quench in oil from muf-
fle or semimuffle furnaces and in molten salt from
a high-temperature salt bath. After its temperature
has been equalized in the salt quench, the tool is air
cooled. For large cutters heated in a furnace, an
interrupted oil quench is often· used to minimize
quenching strains and prevent cracking. This con-
sists of cooling the cutters in the oil only until they
lose color (about 1000 F or 540 C), and then cool-
ing in air.
Salt baths used for austenitizing of high speed
steels are usually composed of 92 to 960/0 barium
chloride and 4 to 8070 sodium chloride.
After quenching, high speed steel tools 'usually
possess high residual stress, and to prevent crack-
ing, it is good practice to transfer them from the
quenchant to a tempering furnace before they have
cooled to below 150 F (65 C). This is particularly
important for large or intricate tools, for which a
delay between quenching and tempering or permit-
ting the work to cool to too Iowa temperature may
induce cracking. If the work cannot be transferred
to a tempering furnace at once, it should be put in
.a holding furnace maintained at 250 to 400 F (120
to 205 C) until a tempering furnace is available.
Bainitic Hardening. This method of hardening
high speed steels has been used for some applica-
tions, but its merit is controversial. To produce a
primary bainitic structure, this treatment is per-
formed by arresting the quench from the austeni-
tizing temperature at approximately 500 F (260 C),
holding for 4 hours, then cooling to room temper-
ature. This produces a structure with about 55070
bainite and the remainder retained austenite. Sub-
sequent tempering at normal tempering tempera-
ture transforms the retained austenite and tempers
the bainite to a Rockwell C hardness 1 to 3 points
lower than normal for the selected tempering tem-
perature.
Tempering. As shown in Fig. 15-10 for an 1\12
steel austenitized at 2225 F (1215 C), the hardness
of high speed steel is directly affected by tempering
temperature and time. From the slope of the curves
in Fig. 15-10, it can be seen that M2 undergoes
secondary hardening at temperatures above about
700 F (370 C), and that secondary hardening pro-
ceeds at higher temperatures up to about 1100 F
(595 C) depending on time at temperature. These
temperatures approximate the practical limits for

most tempering operations; lower temperatures do
not evoke the secondary hardening response, and
higher temperatures produce hardnesses consider-
ably lower than are usually desired. A properly
tempered high speed steel tool is usually achieved
by tempering it at near the peak of the secondary
hardening, which occurs after tempering in the
range of about 950 to 1100 F (510 to 595 C) for a
conventional austenitizing temperature of 2225 F
(1220 C), but depending also on the time at tem-
pering temperature. For a time of 2.5 hours, which
is a reasonable time, the peak of the secondary is
about 1050 F (565 C). In common practice the
temperature range of 1025 to 1050 F (550 to 565 C)
is probably used more than any other range for
tempering most of the high speed steels.
Care must be used in selecting and controlling
the tempering temperature. Figure 15-10 shows
that the increase in hardness that can be attained
from about 700 F (370 C) to the peak of secondary
hardening can ,be substantial, and if the tool is not
fully hard it may be possible to increase its hard-
ness by retempering at a higher temperature. How-
ever, if the tool has been tempered at some temper-
ature beyond the peak of the secondary, hardness
cannot be increased by tempering, and the only
procedure that can be used to gain full hardness is
by fully annealing and rehardening. As can be seen
in Fig. 15-10, the hardness decrease beyond the
peak of the secondary is very rapid, an an error of
even 25 F (13 C) can mean a substantial loss of
hardness.
High speed steels normally are subjected to two
separate tempering treatments within the range of
1000 to 1100 F (540 to 595 C). The duration of
each treatment is usually 2 hours or more at tem-
perature. It is essential that the time-temperature
combination of the first tempering operation be
adequate to condition the retained austenite. Con-
sequently, the first tempering treatment is some-
times longer and at a slightly higher temperature
than the second, because the latter is used to tem-
per the freshly formed martensite that develops on
cooling from the first temper.
Tempering at too low a temperature or for too
short a time, or both, may not adequately condi-
tion the 20 to 30070 retained austenite present after
initial temper. This austenite will not transform
until the steel is cooled from the second temper
27
and a third temper is then required to temper the
martensite so formed. In order to carry these reac-
tions as near to completion as possible, high speed
steel should be cooled to near room temperature
between tempers.
In some instances certain high speed steel tools
are tempered as many as five times in order to
develop their maximum hardness (up to 70 HRC
for types such as M46 and M47).
Forced air furnaces are generally considered. to
be the most desirable for tempering high speed
steel, because the heat is transmitted from the
heating elements to the work by convection; conse-
quently, the transfer of heat is gradual, and there
is little danger of the work crackIng as the result of
thermal shock. It is advisable to place the work in
a tempering chamber maintained in the tempera-
ture range of 400 to 500 F (205 to 260 C) and to
bring the work up to the tempering temperature
slowly with the furnace. This is particularly impor-
tant for large or intricate tools, because too rapid a
heating rate may lead to cracking.
The very rapid heating rates of molten lead or
salt baths, and the attendant thermal shock, make
them unsatisfactory for tempering high speed steel
tools of other than simple shape and design, unless
the tool~ are preheated to about 600 F (315 C) be-
fore being introduced into the bath.
Refrigeration treatment may be employed to
transform retained austenite. The hardened or
hardened and tempered tool is cooled to at least
-120 F (-84 C) and then tempered or retempered at
normal tempering temperatures. Carburized sur-
faces will respond satisfactorily to the -120 F (-84
C) treatment, even when they have been tempered
prior to refrigeration.
This treatment is not intended to be used as a
means of correcting overheated workpieces that
have retained excessive amounts of austenite.
Nitri~ing. Liquid bath nitriding is preferred to
gas nitriding for high speed steel cutting tools be-
cause it is capable of producing a more ductile case
with a lower nitrogen content.
Although any of the liquid nitriding baths or
processes may be used to nitride high speed steel,
the commercial bath consisting of 60 to 70070 so-
dium salts and 30 to 40070 potassium salts is most
 ·28
commonly employed. The nitriding cycle for high
speed steel is of relatively short duration, seldom
exceeding 1 hour; in all other respects, however,
the procedures and equipment are similar to those
used for low-alloy steels.
The cyanide baths sometimes employed in liquid
nitriding introduce both carbon and nitrogen into
the surface layers of the nitrided case. Normally,
the highest percentages of both elements are found
in the first O.OOl-in. (0.025-mm) surface layer.
Cyanide baths are used to a lesser extent because
of the health hazard.
High speed steel tools that are nitrided in fresh
baths or for short times show steep nitrogen and
hardness gradients. To avoid these steep gradients,
which are believed responsible for the brittleness
of the case after such treatments, the use of longer
immersion time, higher temperature, or a thor-
oughly aged bath is recommended. To avoid brit-
tleness of case when relatively short immersion
times are used, the cyanate content of the bath
should exceed 6070. These conditions often will
lower the surface hardness as well as the hardness
gradient.
Nitriding of decarburized high speed steel tools
should be avoided, because it results in a brittle
surface condition. For those surfaces that have
been softened from grinding, nitriding is frequently
employed as an offsetting corrective measure.
Liquid nitriding provides high speed steel tools
~ith high'hardness and wear resistance and a low
coefficient of friction. These properties enhance
tool life in two somewhat related ways. The high
hardness and wear resistance lower the abrading
action of chip and work on the tool, and the low
frictional characteristics serve to create less heat at
and behind the tool point, in addition to assisting
in the prevention of chip pickup.
Steam treating produces a uniform layer of soft
iron oxide on the surface of finished high speed
steel tools. This layer, approximately 0.0002 in.
(0.005 mm) thick, has lubricant-retaining and anti-
galling properties, and in some applications will
improve tool life by reducing tool-edge buildup.
The oxide layer is removed from the tool after a
short interval of operation; during this interval,
the cutting surfaces of the tool develop a burnished
surface that adds further to antigalling character-
istics.
A typical processing cycle involves .placing the
work in a furnace with a sealed retort, heating to
approximately 700 F (370 C), and equalizing.
After a suitable equalizing time, which depends on
the load, the steam is admitted at controlled rates
for approximately ~ hour. The furnace is then
partly sealed to develop positive steam pressure,
and the temperature is raised to 975 F (525 C). The
steam can then be shut off and work removed
from the furnace and cooled normally.
The treatment produces a blue-black film whose
appearance is improved by subsequent dipp~ng in
oil. This treatment may sometimes be combined
with normal tempering treatments, because the
type of film produced is relatively insensitive to
temperature up to approximately 1075 F (580 C).
Carburizing is not recommended for high speed
cutting tools because of the extreme brittleness of
the case so produced: However, it is suitable for
applications requiring extreme wear resistance in
the absence of impact or highly concentrated load-
ing, such as are encountered with certain types of
cold work dies made from high speed steel. At the
same level of hardness, the carburized layer does
not have the heat resistance of normal high speed
steel because carbides in the microstructure are
predominantly Fe3C, rather than the complex
alloy carbides characteristic of high speed steel.
Carburizing cycles for high speed steel consist of
packing in a carburizing medium, heating to ap-
proximately 1900 to 1950 F (1035 to 1065 C) long
enough to develop the depth of case desired, and
air cooling. The usual holding time at carburizing
temperature is from 10 to 60 minutes, to produce a
case 0.002 to 0.035 in. (0.05 to 0.89 mm) deep.
Deeper cases should be avoided because of the
extreme brittleness developed. This treatment car-
burizes the surface and serves as the austenitizing
treatment for hardening the entire piece. The car-
burized layer will harden to 65 to 70 HRC at the
surface.
Maraging Steels
In contrast to most other steels used for tooling
 29
o. 75r--_ _A_"_"_e_a_I M_a_r_a.,:g_e...,;,(_ha_f_d_e.....
n) -.,

1hr at 815°C

0·50

IHeating
3hr at 480°C
::
C)
0·2
c::
~
.5
CD
~
co
..c
0
Air
cooling
I Heating

1COOling~
Air

-0·25

- 0·50'-------..L.-----'------!----~-T

Fig. 15-11. Typical heat treating cycle for maraging steel, including indicated volume changes at each stage.

applications, carbon is essentially omitted from
maraging steels. Instead, these steels utilize the
soft, ductile, iron-nickel martensite, which can be
precipitation hardened by other alloy additions.
Unlike ·the austenite-ferrite changes which char-
acterize most tool steels, the maraging steels,
which contain approximately 180'/0 nickel, do not
decompose into equilibrium austenite and ferrite
around 840 F (450 C) even if held for long periods.
Instead, with further cooling the austenite trans-
forms to martensite with a body-centered crystal
structure.
Mter the aging treatment about one-half of the
yield strength of the 18070 nickel grades is derived
from the iron-nickel martensite formed upon cool-
ing from the annealing temperature. Upon aging,
the strength is more than doubled as a result of
precipitation of intermetallic compounds.
Molybdenum exerts a pronounced effect on the
maraging reaction. Also, the Mo-Co combination
is coupled with high toughness.
The high strength and nondeforming properties
make these steels an attractive material for many
tooling applications, such as plastic molds, cold
work dies, and dies for hot work, as long as die
temperature does not exceed 840 F (450 C). Tool
components are machined or otherwise fabricated
from these steels in their annealed condition.
These steels can also be hubbed in the annealed
condition. After finish or near finish machining
they are hardened by aging at about 900 F (480 C);
therefore, the amount of deformation is small, and
little if any finish machining is required.
Limitations of maraging steels for tooling appli-
cations are: (1) maximum obtainable hardness is
about SO HRC; (2) high cost; and (3) generally
restricted availability.
Heat Treating Procedures
A typical heating cycle, which includes data on
dimensional change, is presented in Fig. 15-11.
Annealing or solution treatment consists of
heating uniformly to about 1500 F (815 C). This·
temperature can vary some depending upon the
specific alloy. Soaking time should be about one
hour per inch of section, followed by air cooling to
room temerature. The product of this phase of the
heat treatment is soft so that it can be machined or
hubbed.
 30
First Quench Second Successive
Quench
Fig. 15-12. Grain growth of high speed tool steels after requenching without proper precautions. 300 x ..
Stress Relieving. If heavy and.rigorous machin-
ing operations have been performed it is often
advisable to remove stresses by repeating the an-
nealing operation.
Hardening (Maraging). These steels are hard-
ened by subjecting them to a temperature of about
900 F (480 C) and cooling in air. This temperature
may also vary slightly, depending upon the specific
alloy. Precise annealing and aging temperatures
can easily be obtained from the supplier of the
alloy. Soaking time depends upon size of the work-
piece and desired hardness. However, the time at
temperature for development of full hardness is
usually at least 3 hours and can be more for large
workpieces. This simple treatment, which does not
involve quenching, virtually eliminates distortion.
Age hardening can be accomplished. without
precautions against carburization or decarburiza-
tion (carbon is an impurity ·and limited to 0.03010
max.). Therefore, aging can be done in anyone of
several types of furnaces provided the temperature
can be closely controlled.
Nitriding. The conventional nitriding cycle (gas
or molten salt) and the maraging cycle are suffi-
ciently close to permit simultaneous aging and
nitriciing. Shallow, hard cases can be obtained by
nitriding. Surface hardness values as high as 860
HV have been reported.
Third Successive
Quench
Trouble Shooting
The difficulties that arise in heat treating tool
steels are usually in one of two categories: (1)
cracking, and (2) failure to meet the required hard-
ness. It is not possible within the scope of this les-
son to deal with all of the causes of cracking which
lead to failure. This is covered in MEl Course 13,
Lesson 10, Failure of Tools and Dies. However, in
the following paragraphs some of the common
reasons for difficulty will be indicated.
Cracking. In many instances poor design is the
major cause of cracking. This includes drastic
changes in section thickness, poorly spaced holes,
. small fillet radii and sharp corners, all of which
contribute to unbalanced stresses during heating
arid cooling which may result in cracking.
Also, directly related to design is selection of
material. For example, intricate tools are some-
times made from a water-hardening grade when an
oil hardening steel would have prevented cracking.
Further, an oil-hardening grade may have been
selected when an air-hardening grade would have
been a better choice. A capable heat treater often
can, by certain techniques, compensate for selec-
tion errors, but not always.
 ·31

'::~:8~'
Fig. 15-13. Development of network in high speed steel and its correction by annealing and requenching. (a) High
speed steel properly heated. (b to e) Progressive development of network due to holding the high speed steel jor
increasingly excessive lengths oj time at the Qustenitizing temperature. (f) Annealed and properly heated high speed
steel that previously contained moderate network structure due to improper heating practice.

Assuming that design and material are within
reason there are several other areas to investigate
when cracking occurs.
1. The tool may have been rehardened without
annealing, which results in accumulated
grain growth and is likely to crack - if not
in quenching, probably soon after it is placed
in service. Figure 15-12 illustrates the grain
growth which can take place with successive
quenching without intermediate annealing.
2. Poor austenitizing practice - too much time
at temperature or temperature too high, or
 32
both. The high speed tool steels, because of
their high austenitizing temperatures., are
especially crack sensitive. Microstructures
0 2. Failure to heat to a sufficiently high temper-
such as shown in Figs. lS-13d and e are
highly susceptible to cracking.
3. Allowing the tool to become too cold after
quenching and before tempering; or too long
between quenching and tempering.
4. Placing in the temperature furnace too soon
after quenching - sufficient time should be
allowed for transformation to reach comple-
tion before tempering begins.
5. Quenching was too severe for the steel com-
position and/or $e. design.
Failure to Meet Hardness Requirements. Some
of the more common causes of tools not hardening
properly are listed as follows:
1. Accidental mixture of steel grades.
ature for austenitizing.
3. Failure to quench rapidly enough for the
0
hardenability of the grade of steel involved.
4. Decarburization resulting in erroneous hard-
ness readings - grind off some material a~d
retest.
S. Retention of excessive amounts of austenite
from austenitizing at too high a temperature.
6. Tempering at temperatures that were too
high for the as-quenched hardness values for
carbon or low-alloy tool steels.
7. Tempering at a temperature which was too
low - or too high for steels having second-
ary hardening properties (see Fig. 15-10).
 MATERIALS ENGINEERING INSTITUTE
1
I
. ~10L 14, C41 L13
Test Series G HEAT TREATMENT OF TOOL STEELS Page 1

PLEASE PRINT YOUR NAME AND ADDRESS BELOW RETURN TO: ASM International®
MATERIALS PARK, OH 44073-0002
NAME REGISTRATION NO. _

STREET _

CITY STATE ZIP DATE _

MUL TIPLE CHOICE

Place the appropriate letter (A, B, C, D) in the box. (Only one answer is correct
and each question has a value of10 ]Joints.)

1. In tempering high-alloy tool steels:

(A) temperature should always be higher than 230°C (450°F).

D
(B) a protective atmosphere must be used.
(C) multiple tempering using two or more complete cycles is recommended.
(D) quenching from the first cycle is the best practice.

2. As a class, water-hardening tool steels are characterized by their:

(A) high manganese content.

D
(B) large grain size.
(C) low carbon content.
(D) shallow hardening.

3. Intentional carburizing of tool steels is:

(A) usually restricted to applications requiring heat or wear resistance.

D
(B) never used for shock-resisting types.
(C) used only for low-alloy types.
(D) most often used for W types.

4. The basic purpose of incorporating cold treating in a heat treating cycle is to:

(A) allow lower austenitizing temperature.

D
(B) maximize transformation of retained austenite.
(C) compensate for over heating in austenitizing.
(D) eliminate need for double tempering.
 MATERIALS ENGINEERING INSTITUTE

C10L14,C41L13
Test Series G HEAT TREATMENT OF TOOL STEELS Page~

5. When there is no protective atmosphere available, an effective means ofprotecting

tool steel surfaces during austenitizing is: D
(A) coating with sodium silicate.
(B) wrapping in aluminum foil.
(C) wrapping in stainless steel foil.
CD) close spacing of the tools in the furnace.

6. Annealing of certain types of shock-resisting steels can be critical because they are
susceptible to: D
(A) graphitization.
(B) loss of silicon.
(C) carburization.
(D) cracking during heat up.

7. When dimensional control is critical for tools made from oil-hardening cold work
tool (type 0) steels it is common practice to employ: D
(A) austempering.
(B) abnormal austenitizing temperatures.
(C) martempering.
(0) lower than normal tempering temperatures.

8. Tool steels of the A and D groups are:

(A) usually preheated at 790 to 815°C (1450 to 15()()°F) before austenitizing.

D
(B) always preheated at 540°C (lOOO°F) prior to austenitizing.
(C) seldom preheated before austenitizing.
(0) preferably preheated at 650°C (1200°F) maximum prior to austenitizing.

9. All of the A and D types of tool steels are characterized by their:

(A) low carbon content.

D
(B) high silicon content.
(C) tine grain size.
(D) relatively high hardenability.

10. The practice of arresting the quench of high speed steels at approximately 260°C
(500°F) for 4 hours is known as: D
(A) carbide stabilization.
(B) equalizing.
(C) bainitic hardening.
(D) martempering.
 ASM International® is a Society whose
mission is to gather, process, and
disseminate technical information.
ASM fosters the understanding and
application of engineered materials and
The Materials their research, design, reliable manu-
Information Society
facture, use, and economic and social
benefits. This is accomplished via a
unique global infonnation-sharing
network of interaction among members
in forums and meetings, education
programs, and through publications
and electronic media.

Printed on DocuTech October 1994

 The Materials
Information Society

Constitution and Heat

Treatment of
Stainless Steels
by
Paul G. Nelson, B.S.

Course 10
Lesson, Test 15
Course 41
Lesson, Test 14
 Metric Conversion Factors
To convert from To Multiply by

in. mm 25.4
in. m 25.4 x 10-3
mil JJm 25.4
J,Jin. JJm 25.4
in.2 m2 6.45x1~
in.3 m3 1.64 x 10-~
ft. m 3.048 x 1~1
ft. 2 rrr 9.29 x 10-2
ft.3 m3 2.831 x 1~2
oz. g 2.834 x 101
lb. kg 4.536 x 10-1
Btu J 1.054 x 1()3
Btullb.• of Jlkg - K 4.18 x l(Y
Btulft. • hr - of W/m-K 1.730
10 inJin.JOF 1()-6 mmlIDInJ'OC 1.8
psi Pa 6.895 x 1()3
psi kPa 6.895
ksi kPa 6.895 x 1()3
ksi MPa 6.895
ksi - in. l12 MPa-m212 1.099
ksi~iD. MPa- -Jm 1.099
oz.f gf 28.4
lbf kg! 4.536 x 10-1
Ibf N 4.448
Ibf - ft. N-m(orJ) 1.356
IbfJin.2 kgf/cm2 14.223
IbfJin. 3 kgf/m3 2.768 x lOC
IbJft.3 kg/m3 16.019
IbJin. 3 g/cm3 2.768 x 101
IbJin. 3 kg/m3 2.768 x lOC
gal (U.S. liquid) L 3.785
gal (U.S. liquid) m3 3.785 x 10-3
IbJgal gIL 119.826
ft./gal mlIL 748
OF °C (OF - 32)/1.8
OF K (OF + 459.67)11.8
°C OF (OC - 1.8) + 32
°C K °C + 273.15
K °C °C -273.15
Multiple and Submultiple Units
1018 •••••••••••••••••••••••••••• exa 10-1 ••••••••••••••••••••••••••• deci
101~ ••••.••..••••••.••••••••••• peta 10-:: centi
1012 ••••••••••••••••••••••••••• tera 10-3 •••••••••••••••••••••••••• mil1i
1(JJ •••••••••••••••••••••••••••• giga 1Q-6 •••••••••••••••••••••••• Illicro
1()6 mega 1 Q-9 •••••••••••••••••••••••••• nano
1()3 kilo 10-12 •••••••••••••••••••••••••• pica
1()2 •••••••••••••••••••••••••• hecto 10-1~ femto
101 ••••••••••••••••••••••••••• deb 10-13 ••••••••••••••••••••••••••• ano

Abbreviations
J joule m meter Pa pasca1
kg! kilogram force mm millimeter K kelvin
L liter N newton W watt.

by
Paul G. Nelson, B.S.
The Budd Company

e1969 American Society for Metals

This material may not be reproduced in whole or in part in any form whatsoever. Fully protected by copyright.
 Constitution and Heat TreatDlent
of Stainless' Steels
Iron has been alloyed with practi-
cally every mown element in an effort
to improve its corrosion resistance.
The discovery that relatively large addi-
tions of chromium (about 12% or more)
markedly improved the corrosion re-
sistance of iron under atmospheric con-
ditions resulted in the development of
the group of modern alloys popularly
termed "stainl.ess steels." The classi-
fication "stainless" is not a precise one
but rather a relative one, because all of
these alloys will corrode under certain
conditions. However, this in no way re-
duces the industrial importance of these
alloys. They do have outstanding corro-
sion resistance in a wide variety of en-
vironments. Under normal atmospheric
conditions corrosive attack is so little
on many of the stainless steel alloys
that it is easy to understand why they
have been termed II stainless."
The development of stainless steels
cannot be attributed to anyone individual
but must be shared by a number of in-
vestigators. The production of low car-
bon fe rrochromium and chromium
metal by aluminothermic reduction was
certainly a tremendous factor. This
process, developed by Goldschmidt in
1895, is still in use today. Later, using
Goldschmidt chromium, Guillet produced
all three of the present-day important
classes of stainless steel-martensitic,
ferritic, and austenitic-and examined
them metallurgically and mechanically.
His discoveries were complemented by
Monnartz, whose observations of fac-
tors affecting the corrosive behavior of
these alloys were not fully appreciated
until more modern theories of corrosion
were developed.
CONSTITUTION
The three important classes of stain-
less steel which have been developed
have been classified according to their
mic rostructure as follows:
1. Martensitic
2. Ferritic
3. Austenitic
A newcomer into the stainless steel
family is a type of alloy mown as pre-
cipitation-hardening stainless steel.
These alloys will also be discussed.
Each of these classes contains a
minimum of 12 per cent chromium, that
amount being necessary to obtain the
unique passivity or stainless character
possessecr by these alloys. The balance
of the composition has been modified
with other alloying. elements to obtain
the characteristic microstructure and
desired mechanical properties.
Iron-Chromium Phase Diagram
To better understand the phase re-
lationships in iron-chromium alloys, it
is helpful to refer once again to the
2

iron-cemenite phase diagram, Fig. 14:1. form to the gamma (1') or face-centered
This diagram shows that on heating from cubic form at the A 3 temperature. On
room temperature iron transforms from further heating to the A 4 temperature
the alpha (a) or body-centered cubic the gamma form transforms to the delta

3200

3000

~-...
2802-F. 280 ~
~~ &+L L • 4- PERITECTIC
r FERRITE 8 .::..~:" ...., I 2711-.=-.
LIQUID

Y+L
1
LaJ
a::
~
~
<{
a::
I.&J
Cl.

.~~-----
_-+----+----;...-~--+-_t___;__ ___t-'t\~~NT!KJ
IRON CARBIDE
Fe3C

A
2
1414-F.

IFERFCITE
1400
cr,J
i..
':."
-.d.:..:A2 ..:•...•
.
A,
;,.-:~"':1M!': ·" "'+i"':o': ~:'i.: ~:.A~;r·
~
t T ~
••

T
-;-1_+-i'---+1_+-,1
I
t
I I I
~--;------il--~---+--+---+-+----+-~
I
•. - ~ ... ~ ---- A,. "'.'
..
and A. ~
--EUTECTOID
....---
1333 -F.
! ; I ! I I
'-.:
1200~;:
:.
tT;
I I

J
I

I
I;:
i
I
;
: ~! I " I. !I,I
1000.:; I I I
i I
800~:.·_+--4-~-......j..--+-+-~-+---+--+---"--+----+------+--~-+---+--+---II
I a+ em :
•
! !! , I
I
I
600 Ho•-1---+--....f-!--~·-r-·"'"""1i-~-+------iI--+--+-T---+---I·--+I--+---+-+----+-----t
, I I t I
300: ~~--t---t---+.~~I
I ~I-
II' I

- __._ i---If) +---+-----1-!-

o l i! I
05
! I i I
6
~

Fig. 14:1. The Iron-Cementite Phase Diagram.

The critical points for pure iron are shown on the left. Changes appear as lines pro-
• ANNEALED MICROSTRUCTURES - - - - - - - - - -

gressing to the right as carbon is added. Note the location of the A 4 critical line and the
influence of carbon on the delta ((~) field.

(6) form which persists until the melt-
ing point. The delta form, like the alpha
form, has a body-cente red cubic lattice;
in fact they are identical phases. They
have been given different names simply
because they form at different tem-
peratures.
It will be noted from the iron-ce-
mentite phase diagram that increasing
the carbon content raises the A 4 tem-
perature but lowers the A g temperature
until the delta phase has disappeared.
Unlike carbon, there are a number of
alloying elements which when added to
iron in increasing amounts have the
effect of lowering the A 4 temperature
and raising the A s temperature. Since
the phase existing below As is ferrite
and that above A 4 is also ferrite, these
alloying additions have the effect of
narrowing the temperature range in
which austenite can exist. When enough
of the proper alloying element is added,
A 4 merges with As; and there is no
transformation of the type ferrite to
austenite (or austenite to ferrite). With
such compositions the only solid phase
that can exist is ferrite. This is shown
for chromium additions to iron in
Fig. 14:2.
The carbon-free iron-chromium
phase diagram in Fig. 14:3 shows in
more detail how chromium affects the
A 3 and A 4 critical points. It will again
be noted that increasing chromium con-
tent lowers A 4 and raises A 3 until at a
chromium content exceeding 13 per cent,
the two critical points disappear. Iron-
chromium alloys with less than 12 per
cent chromium, when heated to above
A g but below A 4 , transform to austenite.
When cooled at the proper rate from
this temperature, assuming some carbon
is present (usually 0.10 per cent or
more), transformation to martensite
Occurs.
Stainless steels that harden by such
martensitic transformation are known as
3
• ; I
F~eezint1 Point : I
i""~----'-""""",
I !
i •
:.4 4 ;
I
i
32r:=:==~==;_--;-~~~~~_~~"""'....J
2 4 6 8 10 12 14
Chromium, Weight Per Cent
Fig. 14:2 The Effect of Chromium on the
Constitution of Iron.
Note that chromium (a) lowers the
freezing point only slightly, (b) rapidly
lowers A 4 , and (c) first lowers, then
raises As until it joins with A 4 , thus
closing off the field of austenite in what
is termed Hthe gamma loop.i'
martensitic stainless steels. Low car-
bon compositions (0.10 per cent maxi-
mum) which exceed 14 per cent chro-
mium do not transform from ferrite to
austenite but remain ferritic at all
temperatures. These alloys and their
modifications which do not harden sub-
stantially because they remain ferritic
at all temperatures are the class of
stainless steels commonly mown as
ferritic stainless steels or more prop-
erly ferritic stainless irons.
Effect of Chromium often
Alloying Elements is termed a fer-
ritizing element
because increasing amounts of chro-
mium tend to increase the amounts of
ferrite phase which are present. There
are a number of other elements which
have a similar effect. Some examples
are the elements tungsten, molybdenum,
silicon, vanadium, and aluminum which
4

form smaller and smaller gamma loops
as seen in Fig. 14:4.
There are other alloying additions
which tend to stabilize ferrite because
they form carbides and thus reduce the
austenite-forming effect of carbon. How-
ever, before discussing them further it
will be well to consider the alloying
elements which have effects opposite to
those of chromium. These elements
lower A 3 and raise A 4 • When As is
lowered sufficiently only austenite
exists, even at room temperature. When
A4 is raised sufficiently, ferrite dis-
appears entirely so that only austenite .
is present when the melting point is
reached. The effect of carbon in raising
A4 and lowering Ag is shown in
Fig. 14:5. It will be noted that the 100

3200 -
-
2800 -
-

o
Alpha
-
-
-
-
-
-
-
-
-
-
400~ -
-
I I

20 40 60 80 100
Chromium-Per Cent
Fig. 14:3.. Phase Diagram for Carbon-Free Iron-Chromium Alloys.
 5
sigma phase varies over a wide range of
3000 composition as seen in Fig. 14:3. Sigma
phase is usually objectionable because
of its brittle nature. However, one com-
mercial alloy containing 24 per cent
chr9mium plus a few per cent molyb-
denum and nickel can be converted to
100 per cent sigma by suitable heat
treatment. This alloy finds extensive
use in special automotive valve service
because of its unusual resistance to
I OOO""--ol,,---.a...-oIo-.A.-~IIIIIo.-...A.-...A.-..........a-..a.- ~ wear and burning.
o 2 4 6 8 10 12
Alloy Addition, Weight Per Cent
Iron- Nickel Phase Diagram
Fig. 14:4. Elements That Form a Gamma
Loop When Added to Iron. Because only limited amounts of car-
bon can be used in stabilizing austenite,
Carbon-free iron-base alloys are face-
centered cubic (gamma) inside each early investigators experimented with
loop. Outside the loops the alloys are many other elements in an effort to
body-centered cubic. stabilize the austenitic phase. Nickel
was found to have the desired effect of
per cent austenite range is extended to lowering As and raising A4 • Unlike car-
18 per cent chromium when 0.60 per
cent carbon is present. Increasing the
3100
carbon content further does not result in
increased austenite stabilization be- 2700
cause the chromium forms carbides in-
stead of going into solid solution. Other
elements which raise A4 and lower A 3 ,
as seen in Fig. 14:4, are nickel and
manganese.

Sigma Phase A phase of interest and

importance particularly
in the. higher chromium (27 per cent or
more) stainless steels, but also under 700
some circumstances in the more com-
plicated alloys of lesser chromium con- 300
tent, is sigma phase. Referring to o 2 4 6 8 10 12 14 16 18
Fig. 14:3, it will be noted that sigma Chromium, Weight Per Cent
phase appears in the iron-chromium Fig. 14:5. The Effect of Carbon on the
system at chromium contents varying Gamma Loop of Iron-Chromium
from about 20 to 70 per cent chromium. Alloys.*
Sigma, a brittle phase, is a compound
The inner loop is that for carbon-free
of iron and chromium, FeaCrb. The sub- iron-chromium alloys. The outer loop
scripts a and b represent the propor- shows the maximum expansion of the
tions of iron and chromium. Theoreti- loop to about 18 per cent chromium
cally, a and b tend toward equality but when 0.60 per cent carbon is added.

*A similar plot showing the relationship between chromium, carbon, and the gamma field
is given in the 1948 Metals Handbook (Fig. 3, p. 459).
 6

bon it is a substitutional-solid-solution erally cannot be hardened by heat treat-

alloying element in iron with complete
solubility in alloys from 100 per cent
iron to 100 per cent nickel. The com-
plete iron-nickel phase diagram is pre-
sented in Fig. 14:6. Sufficient additions
. of nickel to iron-chromium alloys con-
taining 12 per cent or more chromium
will eventually stabilize the austenite
phase. However, the most important
alloys which have been developed have a
minimum chromium content of approxi-
mately 17 per cent. The alloys which
are essentially all austenite are known
as the austenitic stainless steels.
Like the ferritic stainless steels, the
standard austenitic stainless steels gen-
ment. However, as will be shown later,
the austenitic stainless steels are par-
ticularly advantageous for many applica-
tions. Nickel, because of its effect in
stabilizing austenite, is known as an
austenitizing element. There are other
elements which behave in a similar
manner. The noble metals platinum,
palladium, rhodium, and osmium which
are austenitizers are too costly to have
wide use. Cobalt, which raises both As
and A 41 is also too expensive for wide
use. Manganese and nitrogen have been
widely used to produce both chromium-
manganese-nickel and chromium-man-
ganese austenitic stainless steels.
These alloys have now assumed con-

2800 ...
_-----
~

--- Liquid -~

~ ...........
.. Ferrite (8)
- - - --- ----
~

-
2400 - -
1600 Austenite (y)
lJ...
0

Q) 1200
...'-
::J
0
'-
Q)
a. 800
E
~
400 Magnetic Austenite (y)

o ....--_-....-_~_ ....
~..-..-..-..._..... __ ~,_ . . . . . ._ _ ~_-..... Ioo._ ___

o 20 60 40 80 100
Nickel, Weight Per Cent
Fig. 14:6. Complete Iron-Nickel Phase Diagram.
Austenite is the preferred form. With increasing nickel content, the gamma phase be-
comes stable at room temperature.. The area above the enclosed ferrite (ct) zone is a
necessary two-phase field; the c~oss-hatched bands indicate actual measurements of
transformation on heating and cooling.

siderable commercial importance. How-
ever, the role of nickel is a dominant
one because it has been so widely used
and understood. Nickel also supplements
and improves the excellent corrosion
resistance of the high-chromium alloys.
In summary, it can be seen that in
the stainless steel alloys there is an
atomic balance between two classes of
ele.ments which have been termed aus-
tenitizers and ferriti'zers. The principal
austenitizers are carbon, nickel, nitro-
gen, and manganese; the principal
ferritizers Cl.re chromium, aluminum,
phosphoros, and silicon. The carbide-
forming elements molybdenum, co-
lumbium, titanium, tantalum, tungsten,
vanadium, and zirconium also have a
lesser but sometimes important role as
ferritizers. A proper balance of these
elements must be maintained in pro-
ducing any of the three important
classes of stainless steel-austenitic,
ferritic, or martensitic.
STAINLESS STEEL TYPES AND USES
In present-day industrial manufac-
ture and use, the American steel indus-
try refers to various types of stainless
steel by,their type numbers. These type
numbers were established by the Amer-
ican Iron and Steel Institute (AISI) to
define chemical composition limits. If
the stainless steels are purchased only
to commercial specifications the AlSI
has also established other commercial
standards for these steels. In the AISI
series of stainless steels, the 200 series
is reserved for chromium-manganese
or chromium-nickel-manganese types.
Basically, the 300 series is reserved
for chromium-nickel types, and the 400
series for straight chromium stainless
steels.
Table 14:1 lists the type numbers
and chemical compositions of the three
classes of wrought stainless steel. Some
typical uses are indicated in Table 14:2.
7
MARTENSITIC ALLOYS
The so-called martensitic stainless
steels are identified by their martens-
itic mic rostructure in the hardened con-
dition. Referring back to Fig. 14:3 and
Fig. 14:5, it will be noted that typical
martensitic stainless steels with com-
positions such as Types 410 or 420
freeze as ferrite from the melt. Careful
consideration of the various commercial
analyses will also show that the mar-
tensitic types are limited to the narrow
range of 12 to 18 per cent chromium.
This is also evident from Fig. 14: 5 be-
cause only in this range can a fully
austenitic structure be obtained on re-
heating. As with carbon and alloy steels,
100 per cent austenite is required prior
to quenching if full hardening response
is to be obtained.
Since carbon and chromium offset
one another in the austenite-ferrite
struggle, a proper balance of carbon to
chromium is required if a fully mar-
tensitic structure is to be obtained. With
low carbon (0.15 per cent maximum),
there cannot be' more than about 13.5
per cent chromium, or excess ferrite
will be present which will not transform
to austenite on heating. On the other
hand, if carbon is increased to a large
value, for example one per cent, free
chromium carbides will be present
which do not go into solution in the aus-
tenite. Similarly when the chromium
content is increased to 18 per cent, car-
bon must be increased to approximately
0.60 per cent or free ferrite will be
present.
This atomic struggle is rather un-
fortunate in its effect on corrosion re-
sistance. Raising chromium content
increases corrosion resistance, but in-
creasing carbon content lowers it.
Therefore, one element tends to offset
the other. Principal advantages in using
higher carbon, higher chromium alloys
are increased hardness for cutting and
8

Table 14:1. Types and.Compositions of the Standard Wrought Stainless Steels

AISI Nominal Composition, * weight per cent

Type Carbon Chromium Nickel Manganese Other Elements
Martensitic Alloys
Turbine quality
403 0.15 max. 11.5 to 13.0 1.0 max.
Si - 0.50 max.
410 0.15 max. 11.5 to 13.5 1.0 max.
414 0.15 max. 11.5 to 13.5 1.25 to 2.50 1.0 max.
416 0.15 max. 12.0 to 14.0 1.25 max. S - 0.15 min.
416 Se 0.15 max. 12.0 to 14.0 1.25 max. se - 0.15 min.
420 over 0.15 12.0 to 14.0 1.0 max.
(usually 0.30
to 0.40)
431 0.20 max. 15.0 to 17.0 1.25 to 2.50 1.0 max.
440A 0.60 to 0.75 16.0 to 18.0 1.0 max. Mo - 0.75 max.
440B 0.75 to 0.95 16.0 to 18.0 1.0 max. Mo - 0.75 max.
440C 0.95 to 1.20 16.0 to 18.0 1.0 max. Mo - 0.75 max.
Ferritic Alloys
405 0.08 max. 11.5 to 14.5 1.0 max. Al - 0.10 to 0.30
430 0.12 max. 14.0 to 18.0 1.0 max.
430F 0.12 max. 14.0 to 18.0 1.25 max. S - 0.15 min.
430FSe 0.12 max. 14.0 to 18.0 1.25 max. se - 0.15 min.
446 0.20 max. 23.0 to 27.0 1.5 max. N 2 - 0.25 max.
Austenitic Alloys
201 0.15 max. 16.0 to 18.0 3.5 to 5.5 5.5 to 7.5 N2 - 0.25 max.
202 0.15 max. 17.0 to 19.0 4.0 to 6.0 7.5 to 10.0 N2 - 0.25 max.
301 0.15 max. 16.0 to 18.0 6.0 to 8.0 2.0 max.
302 0.15 max. 17.0 to 19.0 8.0 to 10.0 2.0 max.
302B 0.15 max. 17.0 to 19.0 8.0 to 10.0 2.0 max. Si - 2.00 to 3.00
303 0.08 max. 17.0 to 19.0 8.0 to 10.0 2.0 max. S - 0.15 min.
303 Se 0.08 max. 17.0 to 19.0 8.0 to 10.0 2.0 max. Se - 0.15 min.
304 0.08 max. 18.0 to 20.0 8.0 to 12.0 2.0 max.
304L 0.03 max. 18.0 to 20.0 8.0 to 12.0 2.0 max.
305 0.12 max. 17.0 to 19.0 10.0 to 13.0 2.0 max.
308 0.08 max. 19.0 to 21.0 10.0 to 12.0 2.0 max.
309 0.20 max. 22.0 to 24.0 12.0 to 15.0 2.0 max.
309S 0.08 max. 22.0 to 24.0 12.0 to 15.0 2.0 max.
310 0.25 max. 24.0 to 26.0 19.0 to 22.0 2.0 max. Si - 1.5 max.
310S 0.08 max. 24.0 to 26.0 19.0 to 22.0 2.0 max. Si - 1.50 max.
314 0.25 max. 23.0 to 26.0 19.0 to 22.0 2.0 max. Si - 1.50 to 3.00
316 0.08 max. 16.0 to 18.0 10.0 to 14.0 2.0 max. Mo - 2.00 to 3.00
316L 0.03 max. 16.0 to 18.0 10.0 to 14.0 2.0 max. Mo - 2.00 to 3.00
317 0.08 max. 18.0 to 20.0 11.0 to 15.0 2.0 max. Mo - 3.00 to 4.00
321 0.08 max. 17.0 to 19.0 9.0 to 12.0 2.0 max. Ti is 5 x carbon
(min.)
347 0.08 max. 17.0 to 19.0 9.0 to 13.0 2.0 max. Cb is 10 x carbon
(min.)
*Only principal elements are indicated. Limits on sulfur, phosphorus, and silicon
can be obtained from AISI specifications.

Table 14:2. Typical Uses of the Standard Wrought
Stainless Steels
Type Typical Uses
Martensitlc Alloys
403 Turbine buckets and valves.
410, 414 Nuts and bolts, kitchen tools, tableware,springs,
rules and tapes" furnace parts and heat resisting
applications to 1400°F, cutlery, pump parts,
fastenings, screws, steam turbine parts, gage
parts, oil burner parts, scissors, ship bells, etc.
416 Bolts and nuts, screws, intricate machinedparts
such as pump and valve parts, outboard motor
parts, etc.
420 Cutlery, heat treated springs, pump parts,
knives, dental and surgical instroments, self
tapping screws, etc.
431 Pump parts, steering wheel spokes, windshield
wiper arms, marine propeller shafting, aircraft
stress members, etc.
440A, Bushings and bearings, cutlery valve seats and
440B, trim, heat treated springs, gages, surgical and
440C dental instruments.
Ferritic Alloys
405 Cast-in-place turbine blades, linings for petro-
leum industry, heat exchanger tubes, welded
assemblie s operating below 1400 Cl F that cannot
be annealed after welding.
406 Grid resistors for starting heavy duty motors,
heating elements for furnaces, toasters, waffle
irons, etc.
430 Automotive trim and molding, interior decora-
tive work, refrigerator trays, chemical and
processing towers" condensers, bubble caps,
tanks and screens. Furnace parts subject to
temperatures up to 1550°F, trim on irons, cab-
inets" mixers, toasters, etc.
430F Screws, nuts, bolts, fittings, and other machined
parts that require corrosion resistance similar
to, that of 430.
446 Primarily for heat resisting applications such as
annealing boxes, baffle plates, glass molds,
heaters, oil burner parts, stirring rods, valves
and fittings, x-ray tube bases and lead-in wires.
Austenitic Alloys
201, 202, Aircraft Industry-Screws, fuel lines, engine
301, 302, parts, firewalls, etc.
304
Architectural- Unlimited application, store fac-
ings, frames, rails, trim walls, etc.
Chemical Processing-Tanks, piping towers,
heat eXChangers, bubble caps, etc.
Food Processing-Unlimited applications, ket-
tles, impellers, tanks, refrigerator cars, fer-
mentation vats, pasteurizers, buckets, shovels,
barrels, etc.
Household Items-Tableware, cooking utensils,
mixing bowls, refrigerators, evaporators, wash-
ing machine tubs, etc.
Dairy Industry-Milk cans, pasteurizers, heat
exchangers, freezers, coolers, tanks, and
separators.
Textile Industry-Dye and bleach equipment.
9
Table 14:2 (cont'd)
Type Typical Uses
Austenitic Alloys (cont'd)
Transportation Industry-Trailer bodies, rail-
way passenger cars, structures, passenger car
finish, automobile hub caps, bumpers and trim.
302B Annealing cove'rs, burner sections, furnace
parts.
303 Free-machining 18-8 screws, bolts and nuts,
shafts, carburetor parts, and other machined
parts with corrosion resistance of 18-8.
305 Spun parts, drawn parts, cold heading operations.
308 Welding rods.
309, 310 Chemical and photographic processing equip-
ment. Armealing boxes, covers, welding rods,
furnace parts resistors, retorts, for resistance
to oxidation to"2000° F.
316, 31 7 For chemical processing equipments which re-
quires greater corrosion resistance than that of
18-8. Chemical processing, paper processing,
and soap manufacture equipment.
321, 347 Similar applications to those for 301, 302, and
304 where fabrication by welding would cause
harmful carbide precipitation. They are also
used where considerable corrosion resistance is
required and regular or intermittent heating in
the 800 to 1500t;l F range occurs.
wearing applications or improved
strength for applications such as bolts,
nuts, and other highly stressed parts.
In Types 414 and 431, the nickel con-
tent improves corrosion resistance and
toughness while the alloys can still be
made to transform to martensite. In
Type 431 the nickel content permits full
hardening in an alloy whose chromium
content is higher than that of othermar-
tensitic alloys of similar carbon con-
tent. This alloy combination results in
the most corrosion resistant of the
martensitic steels. The amount of nickel
added is restricted by the fact that
nickel lowers the M s temperature; with
too much nickel full hardening cannot
be achieved.
Annealing The martensitic stainless
steels are annealed or
softened by two principal methods. In
full annealing, they are heated to the
austenitizing temperature followed by
slow cooling to a temperature below the
austenite region. This results in a
 10
structure of -carbide particles in a fer-
rite matrix, a microstructure which
possesses the best ductility and ma-
chinability for many fabrication
processes. In the second method, called
process annealing~ the martensitic alloy
. is heated to a temperature below A3 ,
held for specified time, and then cooled
in air, oil, or water. The resultant mi-
crostructure is a tempered martensite
which possesses sufficient ductility and
machinability for many fabrication
processes.
Table 14:3 outlines typical annealing
cycles for the martensitic types as well
as the hardness values to be expected
from each treatment.
Hardening As might be expected from
their high chromium con-
tents, the hardenability of the martens-
itic stainless steels is quite high.
Therefore, drastic quenching is not re-
quired. Oil cooling or marquenching are
usually adequate for the lower-carbon
grades such as Types 403, 410, 414, and

Table 14:3. Annealing Treatments and Resultant Hardness for Martensitic

Stainleft Steels

Time at Resultant
AISI Type of Temperature, Temperature, Method of Brinell
Type Anneal OF hours Cooling Hardness
403 Process 1350 to 1450 1 to 3 air, oil or water 170 to 195
Full 1550 to 1650 1 to 3 25 to 50°F per hr 135 to 160
to 1100°F then
air, Oil, or water
410 Process Same as for Type 403 170 to 195
Full Same as for Type 403 135 to 160
414 * 1200 to 1300 4 to 8 air, oil or water 235 to 25u
416 Process Same as for Type 403 170 to 195
Full Same as for Type 403 135 to 160
420 Process 1350 to 1450 2 to 6 air, oil or water 205 to 225
Full 1600 to 1650 1 to 2 25 to 50°F per hr 170 to 200
to 1100°F then air~
oil or water
431 * 1150 to 1225 4 to 8 air or oil 225 to 240
440A Process 1350 to 1450 2 to 6 air, oil or water 230 to 245
Full 1625 to 1675 1 to 2 25 to 50°F per hr 205 to 230
to 1100°F then air 1
oil or water
440B Process Same as for Type 440A 235 to 250
Full Same as for Type 440A. 215 to 240
440C Process Same as for Type 440A 250 to 270
Full Same as for Type 440A 230 to 255
*Does not respond to slow cooling. from above the critical.

416; the higher-carbon grades such as
Type 431 and the 440 series can also be
air cooled. A complication in hardening
these higher-chromium steels is the
tendency for austenite retention after
quenching. This tendency is aggravated
by using hardening temperatures which
are higher than necessary. Because of
the sluggishness of these steels it is
generally desirable to use somewhat
longer holding times at hardening tem-
perature than with plain carbon steels
while keeping the hardeningtemperature
on the low side.
Retained austenite is objectionable
because it reduces the hardness of the
part. In addition, it may transform to
martensite when the alloy is in service.
This phase transformation results in a
volume change which can cause con-
siderable internal stress and possible
distortion in the part. Retained austen-
ite, if permitted to occur, may be elim-
inated by subzero cooling to -100 of or
by multiple tempering of the hardened
steel. These methods are similar to
those used for high speed steels.
11
employed in hardening all types and is
almost mandatory for Types 420, 431,
440A, 440B, and 440C if parts are large
or at all complex in shape. Preheating
is accomplished by heating slOWly and
holding for about one hour at 1000 to
1450°F prior to austenitizing.
After hardening, the martensitic
steels are either stress relieved or
tempered. It is extremely importantthat
these steels are not quenched to a tem-
perature below 150°F prior to either of
the se treatments. Quenching to room
temperature may result in increased
distortion or cracking of the part.
Because the higher-carbon steels
such as 440A, 440B, and 440C have
rather low M f temperatures, double
tempering or double stress relieving
is particularly recommended to elim-
inate retained austenite in these alloys.
Afte r the fir st treatment some retained
austenite will transform to martensite
-accompanied by internal stress. The
second temper or stress relief relieves
these stresses.

Hardening treatments for the mar- Stress Relief The term stress relief
tensitic stainless steels are listed in as applied to as-
Table 14:4. Preheating is frequently quenched steels indicates a heat treating

Table 14:4. Hardening Treatments for Martensitic Stainless steels

Time at
Austenitizing ,Temperature, Quenching Brinell
Type Temperature, OF minutes Medium Hardness
403 1700 to 1850 15 to 30 oil 380 to 415
410 1700 to 1850 15 to 30 oil 380 to 415
414 1800 to 1950 15 to 30 oil (air 400 to 450
for small sections)
416 1700 to 1850 15 to 30 oil 380 to 415
420 1800 to 1900 15 to 30 air 530 to 560
431 1800 to 1950 15 to 30 air 400 to 440
440A 1850 to 1950 15 to 30 air 550 to 580
440B 1850 to 1950 15 to 30 air 570 to 590
440C 1850 to 1950 15 to 30 air 600 to 620
12
process which relaxes internal stresses erties. The effect of tempering tem-
without permitting important micro- perature on the properties of several
-structural changes to occur. In general, common martensitic stainless steels is
this process is carried out at tempera- illustrated in Fig. 14:8. It will be noted
tures below those which are used for that the etfects are similar to those
tempering. Common stress relief treat-- obtained in tempering low alloy and car-
ments employ temperatures from about bon steels. Higher tempering tempera-
300 to 700°F for times of 1 to 3 hours. ture reduces hardness, yield strength,
The variation in hardness and impact and tensile strength, but increases
strength of Type 420 martensitic stain- elongation and impact strength.
less steel at various stress-relief tem-
peratures is shown in Fig. 14:7. It is
interesting to note that there is a jump ~--------------t300 :g
in hardness at about 800°F. This hard- ...
~
0» ~
CD
c::
ness increase is believed to result from C
CD 2000
::I:
transformation of retained austenite. ~ 125
~
Q; -q;
-- CL
Tempering Tempering involves >-0

higher temperatures than 8 100

-gc_
stress relief (usually 1000 to 1400°F) o!!
at times of from 1 to 4 hours. Temper- 25 ~
~
50 o
c
ing treatments, unlike stress relieving, 20 N
result in marked changes in hardness, :a
microstructure ~ and mechanical prop-
...
I
-...
c
15;i
o
t;
10_~
c
c lmpac1
~ ."...",,, "0
E
_---- 5 g'
0_-----_.----
(.)
"'C o
Q;
3
55 o
N
_--.... --'--_ _..... Ow
~ 1000 1200 1400
0
0 50 Tempering Temperature, OF
a::
0
(J)
Q)
45 Fig. 14:8. The Effect of Tempering Tem-
c:
~
~
perature on the Mechanical Prop-
0
J: 40 erties of several Martensitic
Stainless Steels.
:9 15 Note the close conformity of Type 416
...
I
-...
(the free-machining grade) with Type
...o 410 except for impact strength. All
samples were tempered for four hours.
c
~
E

O~_--.... ---a. ~~
475°C Brittleness considerable
Of
100 300 500 700 900 iInportance in
Stress-Relieving Temperature, OF
stress relieving and tempering martens-
Fig. 14:7. The Effect of Stress-Relieving itic steels is the phenomenon of 475°C
Temperature on the Hardnessand brittleness. The brittleness, which is
Impact Strength of Type 420 also accompanied by some loss of cor-
Stainless Steel. rosion resistance, results when fer-
Samples were held at temperature for ritic and martensitic steels are cooled
four hours and then cooled in air. slowly through the temperature range

600 to 400°C (1100 to 750°F). The ef-
fects are most noticeable on high-chro-
mium (16 per cent or more) ferritic
steels but are also observed to some
extent in martensitic steels containing
13 per cent or more chromium. This
phenomenon can be prevented by cool-
ing rapidly through the 1100 to 750 0 F
temperature range. Air cooling is suf-
ficiently rapid for the martensitic and
ferritic steels, and 475°C brittleness
can be eliminated in these. steels by
heating to above 1100°F and air cooling.
Corrosion Although heat treatment
Resistance is essential for providing
the wide range of me-
chanical properties in the martensitic
stainless steels, it is also important in
its effect on corrosion resistance. Al-
though this class of stainless steels is
not subject to the carbide sensitization
encountered in austenitic steels, free
carbides do reduce corrosion resist-
ance. Because the carbon is dissolved
in hardening, the hardened steels have
better corrosion resistance than the an-
nealed steels.
FERRITIC ALLOYS
The ferritic stainless steels possess
a mic rostructure which is primarily
13
ferritic. It will be noted that these
alloys have not been defined ~s non-
hardening stainless steels. Actually, in
some of these alloys there are minor
proportions of austenite which can
transform to martensite with accom-
panying hardening. These alloys also
work harden, although not as signifi-
cantly as the austenitic alloys.
Referring to Fig. 14:5, comparison
of the compositions of the ferritic alloys
from Table 14:1 will show that most
ferritic alloys are to the right of the
gamma loop. Type 430, however, has a
chromium and carbon range which can
result in a mixture of martensite and
ferrite. The chromium content can be
varied from 14 to 18 per cent. With
carbon on the low side and chromium
on the high side, the structure is largely
ferritic and nonhardenable. On the other
hand, with chromium on the low side
and carbon on the high side, the alloy
can be hardened. Table 14:5 shows typ-
ical mechanical properties of the fer-
ritic ·alloys. The hardened properties of
Type 430 are fo.r compositions that will
harden significantly.
Large quantities of the 430 alloy are
used in nitric acid manufacturing plants.
Type 430 is also a popular alloy for
automotive and other trim applications

Table 14:5. Mechanical Properties of Ferritic Stainless Steels

Type Modulus of Ultimate 0.2% Offset Elongation Izod
and Elasticity, Tensile Yield in 2 inches, Brinell Impact,
Condition million psi Strength, Strength, per cent Hardness ft-Ib
psi psi
405 (Annealed) 28 to 29 60,000 35,000 . 20 160 to 180 20 to 35
406 (Annealed) 29.0 85,000 25
430 (Armealed) 29 to 30 65,000 35,000 20 to 35 130 to 165 15 to 35
430 (Hardened) 150,000 110,000 3 255 to 300 15 to 35
430F (Annealed) 29.0 70,000 45,000 15 to 30 150 to 190 15 to 35
430F (Hardened) 150,000 110,000 3 275 to 300
446 (Annealed) 28.0 80,000 50,000 25 to 30 160 to 185 1 to 10
14
.and accounts for alm,ost 75 per cent of
the tonnage of ferritic alloys. The an-
nealed ferritic alloys have better duc-
tility than the martensitic alloys and are
therefore more satisfactory for' cold
forming applications.
AIUlealing Except for Type 430 there
is no application of hard-
ening, or tempering and stress relieving
of ferritic steels. The only commonly
used heat treating process is annealing.
Annealing is largely restricted to proc-
ess annealing such as previously de-
scribed for martensitic steels. A com-
mon treatment consists of heating to
0
1450 to 1600 F, holding for 1 to 2 hours,
and cooling in air (or oil or water to
avoid 475 0 C embrittlement). Cold-rolled
strip steels may be heated for shorter
periods of time (10 to 15 minutes). The
free-machining grade 430F· is annealed
at a somewhat lower temperature (1250
to 1450°F).
Grain Growth The ferritic alloys,
when heated above
1650°F, are subject to grain coarsening
which greatly reduces notch toughness
as measured by such tests as the Izod
impact test. With most ferritic alloys,
this coarse' grain structure cannot be
refined by heat treatment because there
is no ferritic to austenitic transforma-
tion as with the martensitic alloys.
Where welding is performed, Type 430
is often specified in the hardenable
composition so that subsequent heat
treatment can be used to reduce the
grain size.
Because large grain size has such a
detrimental effect on notch toughness,
it must be avoided in the ferritic alloys.
There are no methods which reduce the
grain size in most alloys except hot or
cold working, and these are not always
possible. In Type 446. nitrogen is used
as a grain refiner. However, this alloy
is still subject to grain growth if heated
above 1650°F.
Hardening Except for modifications
of Type 430 which are only .
partly ferritic above As, the ferritic
stainless steels cannot be hardened by
heat treatment. They are subject to
475°C embrittlement as preViously de-
scribed for martensitic steels. This
enibrittlemerlt is avoided and eliminated
in the same manner as with the mar-
tensitic steels. Ferritic alloys are also
subject to sigma phase formatio'n as
described earlier.
Cold Working Ferritic stainless
steels can be hardened
by cold working. However, the rates of
work hardening are much lower" than
those 9btained with austenitic stainless
steels. It is possible to cold reduce
ferritic alloys as much as 90 per cent
prior to annealing. Because less an-
nealing between stages is required, the
ferritic alloys are more economical to
cold reduce than are the austenitic
alloys.
AUSTENITIC ALLOYS
The austenitic stainless steels are
characterized by their austenitic (face-
centered cubic) structure. They are
also, with minor exceptions, nonhard-
enable by heat treatment and nonmag-
netic. As previously explained, these
alloys result from modifying the iron-
chromium alloys with austenitizing ele-
ments (nickel being the most important)
until the austenitic microstructure is
retained at room temperature. Because
of the sluggishness of transformation in
these alloys, it is not necessary to add
sufficient austenitizers to produce an
equilibrium austenite. Many austenitic
alloys are actually metastable alloys;
that is, they are in unstable equilibrium
with carbides in solution and the mar-
 15
tensite transformation suppressed to as structural materials in railway
well below room temperature. equipment, trailers, aircraft compo-
nents, and other highly stressed parts~~
In the borderline alloys, such as
Type 301 and Type 201, the nickel con- The higher-nickel steels, particu-
tent is so low that the austenite is rel- larly those with 10 per cent or more
atively unstable and transforms readily nickel, Wldergo almost no work hard-
to martensite when cold worked. The ening from austenite transformation and
combination of cold work and transfor- are te rmed free spinning because of
mation results in considerable strength- their improved behavior in this opera-
ening as indicated for Type 301 in Table tion. Typical mechanical properties of
14:6. Types 304, 202, and 302 behave the common austenitic steels in the an-
similarly but to a lesser extent. These nealed condition are given in Table 14:7.
cold worked alloys have found wide use
The austenitic alloys. are widely used
because of their superior ductility,
Table 14:6. Variation of Mechanical Properties of toughness, and weldability as compared
Type 301 with Cold Working
to the fe rritic and martensitic alloys.
Ultimate 0.2% Offset They are also more corrosion resistant
Tensile Yield Elongation
Strength, Strength, in 2 inches, in many applications.
Condition psi psi per cent
1/4 hard 135,000 90,000 35 The 200 series of austenitic alloys
1/2 hard 165,000 120,000 25
Full hard 195,000 155,000 15 are high in manganese, the manganese
Extra full hard 240,000 200,000 4 being used in place of part of the re-
Maximum hard 295,000 260,000
quired nickel content. Type 201 is sim-

Table 14:7. Typical Mechanical Properties of AIUlealed Austenitic

Stainless Steels

Ultimate 0.2% Offset

Elastic Tensile Yield Izod
Modulus, Strength, Strength, Elongation, Brinell Impact,
Type million psi psi psi per cent Hardness ft-Ib
301 28/29 100,000 40,000 50/60 155/175 70/110
201 28/29 110,000 50,000 50/60 165/185 70/110
302 28/29 85,000 35,000 50/60 140/160 70/110
302B 28.0 85,000 40,000 50/60 150/170 80/100
303 28/29.0 90,000 35,000 30/55 155/175 70/110
304 28/29.0 85,000 30,000 50/60 140/160 70/110
304L 28/29.0 75,000 28,000 50/60 120/140 70/110
305 28/29.0 85,000 38,000 50/60
308 28/29.0 85,000 35,000 50/60 145/165
309 29.0 90,000 40,000 45/50 165/185 70/110
310 29/30.0 90,000 40,000 45/50 165/185 70/110
316 27.5/28.5 80,000 30,000 50/60 140/160 70/110
317 27.5/28.5 80,000 35,000 50/60 140/160 70/110
321 28/29.0 90,000 35,000 50/55 145/160 70/110
347 28/29.0 90,000 40,000 45/55 155/180 70/110
16
ilar in properties to Type 301, its nickel
counterpart, whereas 202 is similar to
302. Types 303 and 303 Se are free-
machining alloys to which sulfur or
selenium has been added to improve
machinability. Similar improvement to
that observed in martenSitic alloys 416
and 416Se and ferritic alloy 430F re-
sults. Alloys 309 and 310 are widely
used in high temperature applications.
Their higher chromium and nickel con-
tents give higher strength and higher
resistance to scaling at elevated tem-
peratures.
Types 316 and 317 contain molyb-
denum, which is added to improve the
corrosion resistance of these alloys (as
compared to regular 18-8 alloys) in
many applications. High temperature
strength is also improved by the molyb-
denum addition. Because molybdenum
disturbs the austenitic balance, the
nickel content of these alloys must be
increased appreciably to insure a fully
austenitic structure.
Sensitization
The overall corrosion resistance of
austenitic stainless steels and factors
affecting it are complex subjects which
are too lengthy to discuss fully in this
lesson. However, the greatest threat to
the corrosion resistance of austenitic
alloys is sensitizations which is aharm-
ful precipitation of a grain boundary
constituent in a certain temperature
range. This phenomenon, as seen in
Fig. 14:9, reduces grain boundary cor-
rosion resistance so that under many
corro·sive conditions rapid and prefer-
ential corrosion may occur in those
general areas.
The precipitated constituent in the
past has been identified as a nitride and
as a metallic phase, but now is usually
conceded to be a complex carbide of
chromium, iron, and carbon. It is b~-
Fig. 14:9. Sensitization in 18-8 Austenitic
Stainless Steel. 400X.
Note grain boWldary areas which have
been corroded after electrolytic-oxalic
acid etch.
lieved that precipitation of the carbide
depletes the grain boundary area of
chromium, thus making it less corro-
sion resistant.
The rate of precipitation, which is a
time-temperature phenomenon and oc-
curs over a temperature range of 850
to 1475°F, is most rapid in the range of
900 to 1300°F. With alloys such as Type
302, only a few seconds to minutes at
120QoF are required to reduce. corro-
sion resistance drastically under cer-
tain conditions. Numerous tests such as
the Huey nitric acid test, the Strauss
copper sulfate-sulfuric acid test,
microexamination and more recently the
'. ferric sulfate-sulfuric acid and elec-
trolytic-oxalic acid etcn tests have been
used to determine the extent of harmful
precipitation that occurs during sen-
sitization.
Eliminating There are a number of
Sensitization methods which have
been employed to elim-
inate sensitization or prevent its harm-
ful effects. In Types 321 and 347 stain-
Ie ss steels, ~tanium or columbium has

been added for this purpose. These
elements combine with carbon to form
titanium or columbium carbides which
are much more stable than chromium
carbide. When carbon is tied up in the
form of a strong carbide, the possibil-
ity of forming chromium carbide later
when the alloy is heated to sensitization
temperatures is greatly reduced. Be-
cause carbon is an austenitizer, its re-
moval by these additions makes it nec-
essary to increase the nickel content if
ferrite is to be avoided. This method of
eliminating sensitization is particularly
important in welding applications where
post-weld heat treatments cannot be
used.
Frequently with the stabilized steels,
a stabilization heat treatment is given
after the normal solution anneal is em-
ployed. By heating Type 347 to approx-
imately 1600 to 180QoF for several
hours all of the carbon is converted to
columbium carbide. Similarly, titanium
can be converted to titanium carbide by
heating at 1550 to 1650°F for a similar
period of time. Later heating in the
sensitization range then fails to affect
corrosion resistance because the carbon
is already tied up as columbium or
titanium~carbide, and chromium carbide
cannot form.
Another approach to eliminating sen-
sitization' or at least to minimizing its
effect, is lowering the carbon content.
The idea here is that the lower the car-
bon content the lower the amount of
chromium carbide that can be formed.
Type 304, which has a maximum carbon
content of 0.08 per cent, is widely used
in moderately corrosive environments
after heating, as in arc welding for ex-
ample. Even lower-carbon 18-8 steels,
such as 316L and 304L (0.03 per cent
maximum carbon), are being used for
severe corrosive conditions after rel-
atively short sensitization times such
as in arc welding. They are not rec-
17
ommended where the alloys are heated
for extensive periods of time in the
sensitization range and then intermit-
tently exposed to severe corrosion con-
ditions. Types 321 or 347 with stabiliza-
tion heat treatment after welding would
be 'preferred for these applications.
Sensitization can and often is elim-
inated through the use of the solution
anneal. By heating well above the sen-
sitization range, usually 1850 to2000°F,
and cooling rapidly (air or water de-
pendent on section and analysis), all
chromium carbides can be taken into
solid solution and retained there. This
treatment also eliminates cold work
and most internal stresses. The micro-
structure of properly solution heat
treated 18-8 is shown in Fig. 14:10.
(Compare with the sensitized sample in
Fig. 14:9.). All mill annealed austenitic
stainless steel is given a solution an-
nealing treatment.
It is also possible to eliminate sen-
sitization of 18-e steels, such as 301
and 302, by holqing at temperatures of
1200 OF for extensive periods. This
method merely permits the damaging
Fig. 14:10. Microstructure of Properly
Solution Heat Treated 18-8 Aus-
tenitic Stainless Steel. 400X.
There is no evidence of sensitization
after electrolyfic-oxalic acid etch.
18
precipitation to continue until completed.
Further holding at temperature then
permits chromium from the chromium-
rich austenite to diffuse into the chro-
mium-depleted areas, thus largely re-
storing corrosion resistance. The treat-
ment, unfortunately, is generally not
very practical because of the extensive
time that it takes to complete it.
Sigma Phase
There have been some reported
cases in which sigma phase has occur-
red in fully austenitic allays. Generally
these alloys had undergone considerable·
cold work prior to heating in the tem-
perature range in which sigma phase
was formed. However, in mostinstances
ferrite is present prior to sigma for-
mation in the austenitic alloys. It would,
therefore, be expected that alloying
elements which are ferritizers, as men-
tioned before, might unbalance aus-
tenitic alloys sufficiently to promote
sigma formation. Such elements are
silicon, chromium, molybdenum, tung-
sten, titanium, and colum1;>ium.
Actually, Type 309 (25 per cent
chromium and 12 per cent nickel) is
more susceptible to sigma formation
than is Type 310, which contains 25 per
cent chromium, 20 per cent nickel, and
higher carbon. Similarly, a modified
Type 317, which contained 3 per cent
molybdenum and 0.3 per cent titanium,
developed sigma phase rather readily. It
is evident that sigma will occur at
rather low chromium contents in the
austenitic alloys if ferritizers are not
offset by" austenitizing ad~tions such as
nickel and carbon.
An excellent contribution in estab-
lishing the relative role of each element
in promoting the presence of 100 per
cent austenite is that of SChaeffler. His
complete equation, Which is based on
weld metal, is as follows:
I (Cr + 1.8 Mo + 2.5 5i + 2Cb _16}2
Ni+- Mn +30c = -- + 12
2 12
Thi~ equation indicates the minimum
amounts of nickel + manganese + carbon
required to assure that weld deposits
will be' completely austenitic and contain
no ferrite.
Unfortunately, this equation does not
include nitrogen which is particularly
important in the chromium-manganese-
nickel and chromium-manganese stain-
less steels. Experimental work indi-
cates the effect of nitrogen to be about
equal to .that of carbon. It will be noted
from this equation that some possible
compositions of alloys such as 304
would have free ferrite in weld metal.
Wrought alloys require less nickel or
other austenitizer content than do cast-
ingsor weld metal to produce a lOOper
cent austenitic condition. Hot working
operations serve to help break down
delta ferrite into austenite.
Most commercial austenitic alloys
are properly balanced to produce 100
per cent austenite. This is done not only
to avoid sigma formation but also to im-
prove hot working operations, since
two-phase alloys are more difficult to
roll. The presence of delta ferrite is
highly undesirable, particularly in hot
working operations such as continuous
hot rolling of strip. On the other hand,
arc welding manufacturers often have
deliberately designed welding electrodes
to produce weld deposits which contain
appreciable ferrite. In arc welding, the
presence of ferrite helps prevent crack-
ing of weld deposits. Unfortunately such
weld deposits are susceptible to sigma
formation. In high temperature applica-
tions, the ferrite content is often re-
stricted to 1 to 4 per cent, and some
sigma is tolerated.
Sigma formation is austenitic alloys
can occur in the temperature range of
925 to 1775°F, with maximum formation

in the range 1375 to 1650°F. With one
alloy, less than one hour at 1550°F re-
sulted in 100 per cent sigma. Annealing
above 1830°F has restored ferrite in all
cases. The presence of sigma phase re-
duces the corrosion resistance in sul-
furic and nitric acids to an important
degree, reduces ductility, increases
hardness, and drastically reduces im-
pact resistance. With one 317 alloy
heated for four hours at i550~F, notch
impact energy dropped from 120 to 10
foot-pounds. Our present understanding
of this phase has helped prevention of it
in wrought alloys. The problem of pro-
ducing all-austenitic weld metal with
good weldability and no tendency to
sigma formation, however, is still not
completely solved.
PRECIPIT AT ION-HARDENING ALLOYS
It has been mentioned before that the
austenitic steels normally cannot be
hardened by heat treatment, and it is
true that stable austenitic steels cannot
be hardened by martensitic transfor-
mation. However, recent development
of the precipitation-hardening stainless
steels have made hardening possible in
both austenitic and semiaustenitic
alloys. The precipitation hardening
stainless steels can be divided into three
classes-martensitic, semiaustenitic,
and austenitic. Compositions of some of
the more common alloys are given in
Table 14:8.
Martensitic Alloys
The martensitic stainless steels
were the first to which precipitation-
hardening additions were made. This
was a reasonable choice because mar-
tensite is a strong base for further
hardening. Also, the solid solubility of
many elements is much less in ferrite
and martensite, which ~re both body-
centered-type crystals, than it is, in
face-centered cubic austenite. There-
19
fore, it was logical to conclude that an
alloy might be precipitation hardened
after martensitic transformation where-
as it could not be hardened before
transformation when it was in the aus-
tenitic state.
Stainless W is solution annealed by
heating to 1900 ± 50°F for a minimum
of 15 minutes, cooling to room temper-
ature, and holding for two hours. In this
treatment, some of the titanium and
carbon are taken into solution with the
formation of austenite on heating. Ap-
proximately 10 per cent ferrite remains
With the austenite. After cooling to room
temperature and holding two hours, the
austenite transforms to martensite. Be-
cause the martensite is rather low in
carbon it is also low in strength, but
possesses fair ductility. Typical prop-
erties of solution annealed stainless
Ware:
Yield Strength (0.2% offset) 75,000 psi
Ultimate Tensile strength 120,000 psi
Elongation in 2 inches 25%
For maximum properties Stainless
W is then aged at 950°F for 30 minutes.
A wide range of properties can be ob-
tained by overaging at temperatures
from about 1150 to 1250°F for various
times. During aging treatments it is
believed that titanium carbide is pre·
cipitated. Because greater hardening is
obtained when aluminum is present, it is
probable that this element also plays a
part in precipitat~on, possibly as an
aluminum-nickel compound. Typical
aged properties of Stainless Ware
shown in Table 14:9.
Heat treatments for 17-4 PH are
similar to those for Stainless W. Solu-
tion heat treatment is usually 1900 ±
25 0 F for 30 minutes followed by cooling
to 90°F or below. This treatment, re-
ferred to as Condition A, takes the
principal hardening element copper into
solution. Some carbides are largely un-
 ~
o

Table 14:8. Compositions of Representative Commercial Precipitation-Hardenable Stainless Steels

Nominal Composition, * weight per cent

Alloy C Mn 91 Cr Ni AI Mo Other Elements Fe
Martensitic Types
Stainless W 0.12 max. 1.0 max. 1.0 max. 16.0 to 18.0 6.0 to 8.0 1.0 max. - 1.0 Ti max., 0.2 N max., Balance
1.0 AI max.
17-4 PH 0.07 max. 1.0 max. 1.0 max. 15.5 to 17.5 3.0 to 5.0 - - 3.0 to 5.0 eu, 0.15 to
0.45 Cb + Ta
Balance

Semlaustenitic Types
17-7 PH 0.09 max. 1.0 max. 1.0 max. 16.0 to 18.0 6.5 to 7.75 0.75 to 1.50 - - Balance
PH 15-7 Mo 0.09 max. 1.0 max. 1.0 max. 14.0 to 16.0 6.5 to 7.75 . 0.75 to 1.50 2.0 to 3.0 - Balance
AM 350 0.12 max. 0.90 0.50 max. 16.0 to 17.0 4.0 to 5.0 - 2.5 to 3.25 0.10 N 'Balance
AM 355 0.15 max. 0.95 0.50 max. 15.0 to 16.0 4.0 to 5.0 - 2.5 to 3.25 0.10 N Balance

Austenitic Types
A-286 0.08 max. 1.0 to 2.0 0.40 to 1.00 13.5 to 16.0 24.0 to 28.0 0.35 max. - 1.0 to 1.5 Ti, 0.10 to Balance
0.50 V
17-10P 0.10 to 0.14 0.50 to 1.00 0.60 max. 16.5 to 17.5 9. 75 to 10. 75 - - 0.25 to 0.30 P Balance
HNM 0.30 3.50 0.50 18.50 9.50 - - 0.25 P Balance
*Unless indicated otherwise, phosphorus and 8ulfurare nominally 0.04 or 0.03 max.

Table 14:9. Typical Mechanical Properties of Aged Stainless W
Property
0.2% Offset Yield Strength, psi
Ultimate Tensile Strength, psi
Elongation in 2 inches, per cent
dissolved because columbium and tan-
talum carbides are quite stable. Like
Stainless W, some delta ferrite also re-
mains in 17-4 PH after the solution
treatment. Typical solution treated
(Condition A) properties for 17-4 PH
are indicated below:
Yield Strength (0.2% offset) 110,000 psi
Ultimate Tensile Strength 150,000 psi
Elongation in 2 inches 12%
Also, like Stainless W, 17-4 PH can
be aged at various temperatures to pro-
duce a range of mechanical properties.
Some typical mechanical properties are
shown in Table 14: 10.
Principal applications of these alloys
have been in bar forms for various
high-strength corrosion-resistant ap-
plications.
Semiaustenitic Alloys
The semiaustenitic precipitation
hardenable stainless steels have a
unique combination of properties that
has made them increasingly more pop-
ular in aircraft structures, missile
Table 14:10. Typical Mechanical Properties of Aged 17-4 PH
Property
Yield Strength (0.2% offset), psi
Ultimate Tensile Strength, psi
Elongation in 2 inches, per cent
21
Aged at 950°F Overaged at 1150 ° F
for 30 minutes for 2 hours
180,000 100,000
195;000 134,000
3 to 10 20
parts, honeycomb, skins, ducts, and
springs. Nonmilitary products include
such uses as handsaws, garbage dis-
posal units, valve diaphragms, com-
pressor discs, and nuclear reactor
components.
When heated to about 1950 °F and
cooled to room temperature, these al-
loys are soft and ductile because the
microstructure is either entirely or
largely austenitic. In this condition they
can be fabricated in a manner similar
to the other ductile austenitic stainless
steels. Typical annealing cycles and
mechanical properties that result are
shown in Table 14:11.
With all of these alloys, marten-
sitic transformation is made possible by
precipitating chromium carbides, thus
raising the M s and M f temperatures.
With one treatment at about 1400 ° F, the
M f is raised to room temperature; with
a second treatment, near 1700 ° F I cool-
ing to -100°F is required to reach the
M f temperature. Typical austenite-to-
martensite transformation condition-
ing cycles with accompanying mechan-
ical properties are given in Table 14:12.
0
Aged at 900 F Aged at 1150°F
for 1 hour for 1 hour
180,000 125,000
195,000 145,000
13 19
22
Table 14:11. Annealing Cycles and Typical Mechanical Properties of
Semiaustenitic Precipitation Hardening Steels

17-7 PH PH 15-7 Mo AM 350 AM 355

Armealing Cycle 1950 ± 25°F for 1850 to 1950°F 1875°F for
30 min. per in. of fpr 45 min. per 10 to 15 min.
thickness. Cool in. of thickness. Cool rapidly.
rapidly. Cool rapidly.
0.2% Offset
Yield Strength. psi 40,000 55,000 63,000 56,000
Ultimate Tensile
strength, psi 130,000 130,000 149,000 187,000
Elongation in 2 inches,
per cent 35 35 39 28.5

Table 14:12. Austenite Conditioning Cycles and Resultant Typical Mechanical

Properties of Semiaustenitic Precipitation-Hardening Steels

1. For Alloys 17-7 PH and PH 15-7 Mo

Condition T
Heat to 1400 ± 25°F
for 90 minutes. Transform
by cooling to 60°F within
1 hr. Hold 30 minutes.
Condition R-I00
Heat to 1750 ± 25 of
for 10 minutes. Mr cool.
Transform by cooling to
-100° F. Hold 8 hours.

17-7 PH PH 15-7 Mo 17-7 PH PH 15-7 Mo

0.2% Offset
Yield Strength, psi 100,000 90,000 115,000 125,000
Ultimate Tensile
Strength, psi 145,000 145,000 175,000 180,000
Elongation in 2 inches.
per cent 9 7 9 7

2. For Alloys AM 350 and AM 355

L annealed and single aged (Condition SA)" Condition SC - L annealed,
Heat to 1710°F for 10 min. Air cool to subzero cooled. Heat to
room temperature. Heat to 1375°F. Hold 3 1710°F for 10 min. Air cool
hours. Cool to room temperature (1710°F to room temperature.
treatment optional for AM 350 but necessary Cool to -100°F for 3 hours.
for best properties in 355).
AM 350 AM 350 AM 355
(Condition SC) (Condition SA) (Condition SC)
0.2% Offset Yield
Strength, psi 110,000 105,000 130,000
Ultimate Tensile
Strength, psi 200,000 160,000
Elongation in 2 in-
ches, per cent 12 10
 23

Table 14:13. Typical Properties of 17-7 PH and PH 15-7 Mo in Conditions

TH 1050 and RH 950
Condition TH 1050 Condition RH 950
Property 17-7 PH PH 15-7 Mo 17-7 PH PH 15-7 Mo
0.2% Offset
Yield Strength, psi 185,000 200,000 220,000 225,000
Ultimate Tensile
Strength. psi 200,000 210,000 235,000 240,000
Elongation in 2 in-
ches, per cent 9 7 6 6

After martensite transformation, two
common aging cycles are commonly
employed on PH 15-7 Mo and 17-7 PH.
To achieve condition TH 1050, material
in Condition T is heated to l050°F for 3
hours. To achieve Condition RH 950,
steel in Condition R-loo is heated to
950 OF for 60 minutes. Typical mechan-
ical properties of these alloys are given
in Table 14:13.
Unlike 17-7 PH and PH 15-7 M0 1
AM 350 and AM 355 are aged in the
same manner regardless of prior mar-
tensite transformation treatment.
Treatment is 850°F for 3 hours. Typical
mechanical properties are shown in
Table 14:14. Condition DA (double aged)
had the" prior SA treatment whereas
Condition SeT (subzero cooled, tem-
pered) had the prior se treatment.
In addition to the above treatments,
other treatments which involve substan-
tial cold reduction (40 to 60 per cent)
plus aging have been developed for all of
these alloys. In some of these treat-
ments, tensile strengths in excess of
300,000 psi with hardness values of 55
Rockwell C have been developed. These
alloys are promising materials for
rocket and high-speed-aircraft con-
struction because of their high strength
and relatively good welding properties.
The molybdenum-containing alloys als~
retain good strength to temperatures of
900 to 1000°F.
There are numerous complications
that can arise in the rather involved
heat treatment of these alloys. It is vital
that the martensitic transformation go
to completion if best properties are to
be obtained by aging. Heat treating
atmospheres are also importantbecause
absorption of carbon and nitrogen will
disturb the austenitic balance and may

Table 14:14. Typical Mechanical Properties of AM 350 and AM 355 in

Conditions DA and SeT

Condition DA Condition SeT

Property AM 350 AM 355 AM 350 AM 355
0.2% Offset
Yield Strength. psi 155,000 155,000 172,000 181,000
Ultimate Tensile
Strength, psi 195,000 195,000 201,000 216,000
Elongation in 2 inches,
per cent 10.5 10 12.8 11.2
24

even make these alloys (and the mar-
tensitic alloys) nonhardening. Generally,
air is a most satisfactory atmosphere
for these allays. Vacuum, argon, he-
lium, and dry hydrogen are also used. A
further complication in the semiausten-
itic types is the volume increase ac-
companying martensitic transformation.
Allowance must be made for this in
parts which are held to accurate di-
mensions. In these alloys, the expansion
amounts to about 0.0045 inch per inch.
Subsequent aging results in a contrac-
tion of about 0.0004 to 0.0005 inch per
inch.
Austenitic Alloys
There are a number of available
austenitic precipitation hardening stain-
less steels. In one group, consisting of
A-286, Discaloy, W-545, and J-1300,
each contain about 15 per cent chro-
mium and 25 per cent nickel plus 2 per
cent titanium and various amounts of
Alloy HNM is representative of
another group of alloys which includes
17-10P and contains 18 per cent chro-
mium, 10 per cent nickel and 0.25 per
cent phosphorus. Each of these groups
of allC?ys is nonmagnetic and not hard-
enable by martensitic transformation.
The austenitic precipitation-hard-
enable alloys are hardened by first
solution annealing and then aging at a
lower temperature. As might be ex-
pected~ they are quite ductile in the
solution treated condition. Solution
treatments and resultant typical me-
chanical properties are given in Table
14:15.
After solution annealing, A-286 is
aged at 1325 OF for 16 hours; HNM is
aged at. 1350°F for 16 hours. Typical
mechanical properties after these aging
treatments are shown below:
A-286 HNM

other elements such as molybdenum~
chromium, etc. These alloys, which are
modifications of a German alloy called
Tinidur, have seen extensive use in jet
engine components because of their
moderately high strength at tempera-
tures over, 1000 OF and good oxidation
resistance at high temperatures.
Yield Strength O.2(~ offset, psi 100,000 124,000
Ultimate Tensile Strength, psi 156.000 168,000
Elongation in 2 inches, '7(. 25 19.5
As mentioned before, A-286 hasbeen
Widely used in high temperature ap-
plications. In general, its fabrication
properties are good. Some difficulty has
been encountered in welding, particu-

Table 14:15. Solution Treatments and Resultant Mechanical Properties for

Alloys A-286 and HNM

A-286 HNM
Solution anneal at 1650 ± Solution anneal
25°F or 1800 ± 25°F for 2050 ± 25°F for 60
90 minutes per inch of minutes per inch of
thickness. Cool rapidly thickness. Cool
Property rapidly.
0.2% Offset
Yield Strength~ psi 37,000 56,000
Ultimate Tensile
Strength, psi 91,000 116,000
Elongation in 2 inches,
per cent 47.5 57.5

larly in welding sections over It inches
thick. Alloy HNM is even more trouble-
some in welding, probably because of the
formation of a low-melting phosphide at
welding temperatures. Reduction of
phosphorus in this alloy minimizes weld-
ing problems but leads to a loss of me-
chanical properties.
These alloys do not require heat
treating procedures as complicated as
those for the semiaustenitic types. Dis-
tortion is also not a problem. Unfortu-
nately, the alloys of this type developed
thus far possess considerably lower
room temperature heat treated proper-
ties than the semiaustenitic alloys. At
temperatures over 900 to 1000°F, how-
ever, A-286 is in many cases stronger
than the semiaustenitic allays. It can be
expected that there will be further new
alloy development with this group of al-
loys because of their great potential
value.
MARAGING STEELS
A new class of Nickel-rich steel is
now being used which has numerous
unique characteristics:
Chemical analyses for three steels of
this class' (the 18% Ni grade seems most
popular) are:
Ni 18% Ni 20%Ni 25% Ni
c 0.03 max. 0.03 max. 0.03 max.
Co 9 0 0 the as- cooled "annealed" condition. It
Mo 5 0 0 forms isothermally and atherm ally. This
Ti 0.2-0.7 1.5 1.5
Al 0.1 0.2 0.2
Cb 0.0 0.5 0.5
Like ordinary steels, they transform
from fcc austenite (150QoF) to martensite
at temperatures of about 300°F. Quench-
ing is not required, because this class of
steel without much carbon, has no pearl-
ite nose to by-pass; therefore, cracking
or distortion from thermal stresses are
25
eliminated. Newly-formed room tem-
perature martensite is not very hard
(about R c 25), but it is very tough; it can
be deformed with comparative ease.
Reheating this martensite is not tem-
pering, but aging; in fact, it is knoWilas
maraging. Hardness rises to about R c 52
after 3 hours at 900°F. This hardness is
retained even when steel is maintained
for many hours between room tempera-
ture and 750°F.
If welded, these steels can be uni...
formly hardened merely by aging again
at about 900°F.
Some New Terms
ausaging - (aus-aging) - aging at elevated
temperature to cause precipi-
tation hardening in an austen-
ite matrix
ausforming - deforming metastable steel
while in austenitic phase
maraging - aging at elevated tempera-
ture to cause precipitation
hardening in a martensite
matrix
marforming - deforming steel - while
martensitic
Transformation
The martensitic crystal structure of
maraging steels is body-centered cubic
(not body-centered tetragonal) when in
martensite work-hardens very little,
making it quite formable prior to marag-
ing. The mechanical properties can be
further improved with coldworking of at
least 25% prior to the maraging treat-
ment.
Recommended austenitizing is 150Q oF
(one hour per inch thickness) and special
atmospheres for any of the maraging
steels are ulUlecessary.
26

Because of the higher alloy content, refrigerating 8 hours at 10QoFor lower is

the 25% Ni steel does not transform to necessary for full martensite of the 25%
100% martensite at room temperature; Ni steel, or the steel should be cold
the 18% Ni and 20% Ni steels do. Ausag- worked at least 25% to induce complete
ing 4 hours at 1300°F followed by austenite-to-martensite transformation.

Typical Properties of Maraging Steels

18%Ni 20%Ni 25% Ni

Cond.-
PROPERTIES Annealed Mar age d Annealed Mar aged Annealed Maraged

Tensile Strength,
Ksi 140 250-275 153 255-268 132 260-290
Yield Strength,
Ksi 95 240-268 115 243-262 40 240-270
%Elong.· 17 10- 12 25 12- 13 30 10- 15
%RA 75 48- 58 65 58- 59 72 40- 60
R c Hardness 28 50- 52 29 50- 52 12 48- 52
Notched Tensile
Str. 0.3" bar 375 365 280-360

HEAT TREATMENT

Austenitize 150QOF, Air Cool 150QOF, Air Cool 1500°F, Air Cool
Ausage none none 1300°F, 4 hrs, Air
Cool; or CW 25%
Recommended
Maraging Temp. 900 800 900 800 900
Recommended or or
Maraging Time 3 Hrs. 4 hrs. 1 hr 4 hrs. 1 hr
 MATERIALS ENGINEERING INSTITUTE

C10L15,C41L14
Test Series E Constitution and Heat Treatment of Stainless Steels Page 1

PLEASE PRINT YOUR NAME AND ADDRESS BELOW RETURN TO: ASM IntemationaJ®
MATERIALS PARK, OH 44073-0002
NAME REGISTRATION NO _
STREET _

CITY STATE ZIP DATE _

MULTIPLE CHOICE

Place the appropriate letter (A, B, C, D) in the box. (Only one answer is correct
and each question has a value of10 points.)

1. Annealing response is not obtained on slow cooling of martensitic stainless steel

Types 414 and 431 from above the critical temperature. This is caused by: D
(A) the nose of the TIT curves shifting to excessively long times.
(B) inadequate solution of chromium in the austenite.
(C) retention of 100% austenite on slow cooling.
(D) cooling too slowly when a rapid quench anneal is specified.

2. Parts have been precision machined from 17-7 PH stainless steel in the solution-
annealed condition. After thermal treatments of Condition T and Condition TH
1050 have been completed, you would anticipate:
D
(A) a decrease in size.
(B) an increase in size.
(C) an increase in ductility.
(D) a decrease in yield strength.

3. Annealing of a cold-worked Type 301 stainless steel has resulted in grain growth.
Grain refinement can be accomplished by: D
(A) process annealing.
(B) ausaging.
. (C) cold working to full hard followed by controlled annealing.
(D) transfonning to 100% sigma phase.

4. Sigma phase is fonned occasionally in austenitic alloys when exposed to elevated

temperatures. An alloying element that tends to promote sigma phase is: D
(A) molybdenum.
(B) manganese.
(C) nickel.
(0) carbon.
 MATERIALS ENGINEERING INSTITUTE

C1 OL15, C41 L14

Test Series E Constitution and Heat Treatment of Stainless Steels Page~

Aluminum additions are specified in the chemical composition of Type 405 to serve
5.
as a: D
(A) deoxidizer.
(B) austenitizer.
(C) ferritizer.
(D) carbide former.

Nickelmaraging steels when fully hardened have a microstructure of:

6.

(A) austenite and bainite.

D
(B) aged marten~ite.
(C) tempered martensite.
(D) retained austenite and ferrite.

A substitute austenitizer that can be used to conserve nickel in the austenitic class
7.
·of stainless steels is: D
(A) molybdenum.
(B) silicon.
(C) manganese.
(D) vanadium.

Sulfur and seleniumin controlledamounts are addedto some types ofstainless steels
8.
to improve: D
(A) machinability.
(B) weldability.
(C) castability.
(D) hardenability.

9. In the AISI system of stainless steel type designations; the various series numbers
indicate that: D
(A) an alloy in the 400 series is always martensitic.
(B) an alloy in the 300 series is austenitic.
(C) the alloys in the 200 series have a manganese content of 2.0%.
(D) the last two digits represent the carbon content.

10. Copper is the principal hardening element used in the martensitic precipitation-
hardening steel designated: D
(A) A-286. (C) 17-7 PH.
(B) 17-10P. (D) 17-4 PH.
 ASM Intlmational~ is a Society whose
mission is to gather, process and
disseminate technical information.
ASM fosters the understanding and
application of engineered materials and
their research, design, reliable manu-
The Materials facture, use and economic and social
Information Society benefits. This is accomplished via a
unique global information-sharing
network of interaction among members
in forums and meetings, education
programs, and through publications
and electronic media.

Printed on DocuTech July 1994

 ASH\.
c:zThe Materials
Information Society
Materials Park, OH 44073-0002
www.asminternational.org

Introduction to Heat Treating

~Course 0740, Final Exam, Test Series A~

1. Five choices are listed below. Which is NOT a reason for heat treating:
a. Fluidity
b. Hardness
c. Ductility
d. Strength
e. Toughness

2. How is hardness defined with respect to metals?

a. By the amount a bar will elongate before fracturing
b. By the test and behavior of the metal
c. As a combination of strength and ductility
d. As a measurement of the resistance to scratching or indentation
3. How is toughness defined with respect to metals?
a. By the amount a bar will elongate before fracturing
b. By the test and behavior of the metal
c. As a combination of strength and ductility
d. As a measurement of the resistance to deformation
4. What is an alloy? (
a. A combination of metals
b. A combination of metals and non-metals
c. A combination of metals or metals and non-metals
d. A combination of non-metals
5. Which of the following is not a goal of annealing?
a. Refining grain size
b. Elimination of gas
c. Hardening
d. Solution treatment
6. What elements are combined to create steel? (Select all that apply).
a. Silver
b. Chromium
c. Iron
d. Copper
e. Carbon
f. Titanium

© 2006 ASM International®

2 ~ Course 0740, Final Exam, Test Series A

7. Calculate stress based on the following information: Load·SOO Lbs, cross sectional area.6
square inches.
a. 83.33 Pounds per square inch
b. 0.012 Pounds per square inch
c. 83.33 Foot Pounds
d. 0.012 Foot Pounds
e. 0.56 Pounds per square inch
8. Fatigue fractures are caused by shock loading
a. True
b. False
9. Hardness values that are derived from similar basic fundamental calculations would be
expressed as:
a. DPH and Scleroscope
b. Rc, KHN and Mohs
c. BHN,DPH,KHN
d. BHN and Rc
10. When steel is slowly heated, the rearrangements of atoms from body-centered cubic to face-
centered cubic:
a. induces ferromagnetism
b. causes a dimensional contraction
C. inhibits diffusion
d. removes atoms from the crystal lattice
11. When optimum machinability or formability is desired, the most important goal of heat
treating is to:
a. obtain a spheroidlzed microstructure
b. reduce the hardness of the surface
c. provide a structure of lamellar pearlite
d. avoid graphitization
12. The tendency for steel to crack in quenching is increased by:
a. high Ms temperature
b. lower carbon content
c. high austenitizlng temperature
d. higher quench bath temperature

13. What is meant by the term secondary hardening?

a. Rehardening
b. Austenitize a second time
c. Precipitation of carbide forming elements
d. Tempering at low temperature
14. The first stage of tempering involves:
a. The decomposition of martensite
b. The decomposition of Bainite
c. The decomposition of chromium carbides
d. The decomposition of austenite

© 2006 ASM International®

3 ~ Course 0740, Final Exam, Test Series A

15. When the carbon potential of the furnace atmosphere is lower than that of the steel being
processed, the surface carbon content of the steel will:
a. Increase
b. Decrease
c. Remain the same
d. form methane

16. The Maraging tool steels will generally contain an approximate Nickel content of:
a. 4%
b. 2.5%
C. 18%
d. 12%

17. P20 Mold steel can be used for:

a. Magnesium die casting dies
b. Aluminum extrusion dies
c. Compression molding
d. Cutting tools

18. The martensitic group of stainless steels are identified as:

a. 600 Series Stainless
b. 300 Series Stainless
c. Precipitation hardening stainless
d. 400 Series Stainless

19. A stainless steel with a nickel content of 20% will experience a magnetic change at the
following apprOXimate temperature:
a. 950 F
b. 650 F
c. 800 F
d. 820 F

20. Retained austenite in a stainless steel 440:

a. Is necessary
b. is objectionable
c. cannot be transformed
d. will not occur

© 2006 ASM International®