Professional Documents
Culture Documents
P.W. Schilke
GE Energy
Schenectady, NY
Contents
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..........1
Materials Philosophy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..........1
Turbine Buckets and Nozzles. . . . . . . . . . . . . . . . . . . . . . . . . . . ..........2
Processing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..........2
Bucket Materials . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..........4
Directionally Solidified — GTD-111 Buckets . . . . . . . . . . . . . . . . . . ..........5
Equiaxed Buckets — GTD-111 . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..........6
IN-738 Buckets . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..........6
U-500 Buckets . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..........6
Future Buckets . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..........7
Nozzle Materials . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..........8
FSX-414 Nozzles. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..........8
GTD-222 Nozzles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..........8
N-155 Nozzles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..........8
Future Nozzle Materials and Coatings . . . . . . . . . . . . . . . . . . . . . . . ..........8
Bucket Coatings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..........9
Hot Corrosion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..........9
High-Temperature Oxidation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
High-Temperature Coatings. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
Platinum-Aluminide Coatings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
PLASMAGUARD™ Coatings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
Low-Temperature Coatings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
Shroud Coatings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
Future Coatings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
Combustion Hardware . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
Combustion Liners. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
Transition Pieces . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
Turbine and Compressor Wheels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
Turbine Wheel Processes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
Turbine Wheel Alloys . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
Alloy 706 Nickel-Base Alloy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
Cr-Mo-V Alloy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
12 Cr Alloys . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
A286 Alloy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
Other Rotor Components . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
Rotor Developments. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
Contents (cont’d)
son with the current baseline material. duction, specimens produced with certain
More than 50 Rainbows have been installed forgings and select-ed castings are destructively
since the 1950s, covering compressor blade, tested to ensure specification compliance.
compressor coating, combustor, nozzle, nozzle Critical rotating components are subjected to
coating, bucket and bucket coating materials. non-destructive inspection techniques such as
Virtually all of the improved hot gas path ultrasonics, liquid penetrant, magnetic particle
materials that we now regard as standard were and X-ray examination, depending upon the
rain-bow tested in Rainbow rotors prior to component. Proof testing is also performed on
introduction. Typically, the Rainbow materials the most critical components.
are removed and evaluated periodically, then This philosophy of material development and
replaced with standard parts. Current Rainbow production qualification has existed since GE
tests include bucket and nozzle coatings, com- began building gas turbines in the 1950s, and
bustion components and bucket and nozzle it will continue in the future to meet the needs
materials. for improved materials in new and uprated
The Rainbow rotor tests, including the long- machines.
term laboratory tests, constitute the corner-
Turbine Buckets and Nozzles
stone of the materials development philoso-
phy. They have successfully provided a continu- Processing
ous stream of carefully evolved materials and GE has used investment cast nozzles and buck-
processes for GE heavy-duty gas turbines. ets made by the lost wax technique since the
After a material has been proven in a Rainbow mid-1960s. This casting process allows the use
rotor, producibility is verified through exten- of alloys that are difficult to form or machine
sive first piece qualification tests and pilot lot and provides great design flexibility for inter-
evaluations. Components must continue to nal cooling schemes. For example, ceramic
meet rigorous production non-destructive and coring is used extensively in these castings to
destructive test requirements. Extensive work form air-cooling passages and to provide
with suppliers is completed in order to qualify weight reduction.
parts that use a new material. During this time, Most nozzle and bucket castings used by GE
trial parts are destructively tested and analyzed are made by using the conventional equiaxed
to determine that the properties meet the investment casting process. In this process, the
requirements defined by the GE specifications. molten metal is poured into a ceramic mold at
Hundreds of bucket and nozzle castings and a pressure below 10-2 torr (10-2 mm Hg).
many wheel forgings have been cut up for Vacuum is used in most cases, except for some
analysis to verify that the processing (starting of the cobalt alloys, to prevent the highly reac-
stock, casting/forging parameters, heat treat- tive elements in the superalloys from reacting
ment, etc.) is correct. with the oxygen and nitrogen in the air. With
Once a supplier becomes qualified, the proper control of metal and mold thermal
processes used to make that component are conditions, the molten metal then solidifies
"frozen" for production and can not be from the surface toward the center of the
changed without GE approval. Once in pro- mold, creating an equiaxed structure. To pre-
vent shrinkage porosity, care is taken to allow tion of these transverse grain boundaries con-
proper feeding of molten metal to the casting fers additional creep and rupture strength on
while it solidifies.A variety of investment cast the alloy, and the orientation of the grain
buckets and nozzles has been produced during structure provides a favorable modulus of elas-
the past 30 years. The examples in Figures 1 ticity in the longitudinal direction to enhance
and 2 indicate the process flexibility in accom- fatigue life. More recently, GEPG has worked
modating design and size variations. with its suppliers to develop large, single-crystal
Directional solidification (DS) is also being castings that offer additional creep and fatigue
employed to produce advanced technology benefits through the elimination of grain
buckets. First used in aircraft engines more boundaries.
than 25 years ago, it was adapted for use in The MS5002C directionally solidified bucket
large air-foils through the efforts of GEPG and was the first large land-based gas turbine DS
its suppliers several years ago. By exercising bucket made on a production basis and has
careful control over temperature gradients, a been in commercial service since 1989. Figure 3
planar solidification front is developed in the shows three recent examples of directionally
bucket, and the part is solidified by moving solidified stage 1 buckets: an MS9001FA, an
this planar front longitudinally through the MS7001FA and an MS6001FA. All are etched
entire length of the part. The result is a bucket to show the directional grain structure.
with an oriented grain structure that runs par- Secondary operations include electrochemical
allel to the major axis of the part and contains and electrodischarge machining, hard-coating
no transverse grain boundaries. The elimina- on some components and conventional and
MS7001E MS7001F
creep feed grinding. These processes and subse-
1st Stage 2nd Stage
MS7001F MS9001F quent coatings for corrosion and oxidation pro-
MS6001 1st Stage 1st Stage
1st Stage tection are fully qualified for each design to
MS5002
1st Stage ensure that metallurgical quality is maintained,
adverse residual stresses are not introduced and
overall properties are not degraded. In addition,
dovetails are shot-peened to provide residual
compressive stresses for improved fatigue
MS3002 MS5001
Uprate Uprate strength.
1st Stage 1st Stage
MS5001
MS7001 1st Stage
2nd Stage
MS9001F
3rd Stage
MS6001 MS3002
1st Stage Uprate
1st Stage MS9001FA MS7001FA MS6001FA
1400
2600
Temp. °C
Temp. °F
1100 Steam
2000 1.4 kg/cm2 x 10-3 (20 ksi)
Cooling
severe combination of temperature, stress and 1800
1000
(100,000 hrs)
Conventional
Advanced
Air Cooling SC
DS Alloys
Air Cooling
environment; it is generally the limiting 1600
900 RENE 77 IN738
(U700)
GTD-111
GTD-111
700
1200
Since 1950, turbine bucket material tempera- 1950 1960 1970 1980 1990 2000
Year
ture capability has advanced approximately
850°F/472°C, approximately 20°F/10°C per Figure 4. Firing temperature trend and
year. The importance of this increase can be bucket material capability
appreciated by noting that an increase of
100°F/56°C in turbine firing temperature can until the 1970s, when air cooling was intro-
provide a corresponding increase of 8% to duced, which decoupled firing temperature
13% in output and 2% to 4% improvement in from bucket metal temperature. Also, as the
simple-cycle efficiency. Advances in alloys and metal temperatures approached the
processing, while expensive and time-consum- 1600°F/870°C range, hot corrosion of buckets
ing, provide significant incentives through became more life-limiting than strength until
increased power density and improved efficien- the introduction of protective coatings.
cy. During the 1980s, emphasis turned toward two
Figure 4 shows the trend of firing temperature major areas: improved processing to achieve
and bucket alloy capability. The composition greater bucket alloy capability without sacrific-
of the new and conventional alloys discussed is ing alloy corrosion resistance; and advanced,
shown in Table 1. The increases in bucket alloy highly sophisticated air-cooling technology to
temperature capability accounted for the achieve the firing temperature capability
majority of the firing temperature increase required for the new F generation of gas tur-
COMPONENT Cr Ni Co Fe W Mo Ti Al Cb V C B Ta
BUCKETS
U500 18.5 BAL 18.5 4 3 3 0.07 0.006
RENE 77 (U700) 15 BAL 17 5.3 3.35 4.25 0.07 0.02
IN738 16 BAL 8.3 0.2 2.6 1.75 3.4 3.4 0.9 0.10 0.001 1.75
GTD111 14 BAL 9.5 3.8 1.5 4.9 3.0 0.10 0.01 2.8
NOZZLES
X40 25 10 BAL 1 8 0.50 0.01
X45 25 10 BAL 1 8 0.25 0.01
FSX414 28 10 BAL 1 7 0.25 0.01
N155 21 20 20 BAL 2.5 3 0.20
GTD-222 22.5 BAL 19 2.0 2.3 1.2 0.8 0.10 0.008 1.00
COMBUSTORS
SS309 23 13 BAL 0.10
HAST X 22 BAL 1.5 1.9 0.7 9 0.07 0.005
N-263 20 BAL 20 0.4 6 2.1 0.4 0.06
HA-188 22 22 BAL 1.5 14.0 0.05 0.01
TURBINE WHEELS
ALLOY 718 19 BAL 18.5 3.0 0.9 0.5 5.1 0.03
ALLOY 706 16 BAL 37.0 1.8 2.9 0.03
Cr-Mo-V 1 0.5 BAL 1.25 0.25 0.30
A286 15 25 BAL 1.2 2 0.3 0.25 0.08 0.006
M152 12 2.5 BAL 1.7 0.3 0.12
COMPRESSOR BLADES
AISI 403 12 BAL 0.11
AISI 403 + Cb 12 BAL 0.2 0.15
GTD-450 15.5 6.3 BAL 0.8 0.03
bines. (See Figure 5.) The use of steam cooling impact strength, hot corrosion and oxidation
to further increase combined-cycle efficiencies resistance, producibility, coatability and physi-
will be realized in the 1990s. cal properties must also be considered.
All GE gas turbine bucket alloys are vacuum- Directionally Solidified—GTD-111
cast, nickel-base superalloys that are strength-
Buckets
ened through solution and precipitation-hard-
ening heat treatments. Figure 6 shows the stress The stage 1 bucket material currently in pro-
rupture strength of these alloys and the super- duction is directionally solidified GTD-111. This
alloys used for nozzle applications: GTD-222, is the same as GTD-111 equiaxed except for
FSX-414 and N-155. This comparison is pre- tighter control on the alloy chemistry. This
sented in the form of the stress required for bucket material is currently being used on the
rupture as a function of a parameter that 6FA, 7FA and 9FA turbines, and on the 6B, 9EC,
relates time and temperature (the Larson- 7EA and on the 5/2C and D and 3/2J uprated
Miller Parameter). turbines. DS GTD-111 is also being applied to
stage 2 and stage 3 buckets of the 7FA and 9FA
This parameter is one of several important
gas turbines.
design parameters that must be satisfied to
ensure proper performance of the alloy in a As discussed earlier, the use of directionally
bucket application, especially for long service solidified GT-111 results in a substantial
life. Creep life, high- and low-cycle fatigue, increase in the creep life, or substantial
thermal fatigue, tensile strength and ductility, increase in tolerable stress for a fixed life.
Figure 7 shows the advantage of directionally
solidified GTD-111 compared to equiaxed.
This advantage is due to the elimination of
transverse grain boundaries from the bucket,
the traditional weak link in the microstructure.
30 2
Stress KSI
20
1
Figure 5. Advanced air-cooling technology 10
60
4
0 0
10 15 20 25 30 35
3
FSX-414
Creep Advantage Temp °C
40
IN-738
Kg/cm2 x 10-3
GTD-222
Stress KSI
2 20 30 40 50 60
N-155 U-500 Creep Advantage Temp °F
GTD-111
Buckets • Increased Tensile Strength: ~ 25%
1 Nozzles
• Increased Tensile Ductility: ~ 100%
10 • Increased Fatigue Strength: ~ 900%
Temp.
500 600 700 800 900
100,000 Hrs °C (Strain Controlled)
Life
1000 1200 1400 1600 • Increased Impact Strength: ~ 33%
°F
• Increased Creep Strength: ~ 22°C (40°F)
Figure 6. Stress rupture comparison –
bucket and nozzle materials Figure 7. Directionally solidified GT-111 vs. equiaxed
In addition to improved creep life, DS GTD- required for optimizing the processing of these
111 possesses more than 10 times the strain parts.
control or thermal fatigue compared to
equiaxed GTD-111. The impact strength of DS
IN-738 Buckets
GTD-111 is also superior to that of equiaxed, IN-738 has been the stage 1 bucket material on
showing an advantage of more than 33%. all models built between 1971 and 1984, when
GTD-111 was introduced. In addition, IN-738
Equiaxed Buckets — GTD-111 has been used in more recent years as the
GTD-111, the basic alloy used for both DS and stage 2 bucket material in the three-stage
equiaxed applications, was developed and MS6001, MS7001 and MS9001 models. IN-738
patented by GE in the mid-1970s. It possesses is notable as being one of a very small class of
about a 35°F/20°C improvement in rupture modern superalloys that has an outstanding
strength in the equiaxed form, compared to combination of elevated temperature strength
IN-738. GTD-111 is also superior to IN-738 in and hot corrosion resistance. The balance of
low-cycle fatigue strength. At the same time, these two properties was optimal for heavy-duty
GTD-111 has corrosion resistance comparable gas turbine applications. It was specifically
to IN-738, the acknowledged corrosion stan- designed for application in a land-based gas
dard for the industry. turbine, as opposed to aircraft use. IN-738 was
The design of this alloy was unique in that it the first cast bucket material used by GE in the
utilized phase stability and other predictive heavy-duty gas turbines that had not seen prior
techniques to balance the levels of critical ele- service in aircraft gas turbine applications.
ments (Cr, Mo, Co, Al, W and Ta), thereby IN-738 was first developed by the International
maintaining the hot corrosion resistance of IN- Nickel Company, but its chemistry was subse-
738 at higher strength levels without compro- quently modified by GE to improve its castabil-
mising phase stability. The same principles that ity. This, together with considerable work on
were used to enhance the castability of IN-738 modifying the casting techniques, them-selves,
were also incorporated into GTD-111. enabled the commercial adoption of an alloy
A substantial effort was made to "tune in" that otherwise would have been classed as
GTD-111 so that it could be used to make high nearly impossible to cast in large sizes. This
quality investment cast buckets. During this work enabled the successful application of IN-
phase of the process/alloy development, a 738 over the past 20 years in GE gas turbines.
large number of trial parts were made, repre- Indeed, this alloy is now used throughout the
senting the span of bucket sizes envisioned. At entire heavy-duty gas turbine industry.
first, trials were confined to stage 1 parts, but
more recently this has expanded, and GTD- U-500 Buckets
111 is now being used in the larger, latter-stage Many of GE’s stage 3 gas turbine buckets are
buckets. During all of these producibility trials, currently made of U-500, an alloy that was
buckets were made, non-destructively tested, used for stage 1 buckets in the mid-1960s. Like
and many were cut up to determine proper- IN-738 and GTD-111, this alloy is a precipita-
ties. These evaluations provided the feedback tion-hardened (gamma prime), nickel-base
alloy. It is currently being applied to the latter achieved, thus providing a corresponding
stages of buckets in selected gas turbine mod- increase in high-temperature strength. The
els. transverse creep and fatigue strength is
increased, compared to equiaxed or DS struc-
Future Buckets
tures. GE Aircraft Engines has been applying
With the introduction of DS GTD-111, a com-
single-crystal bucket technology for more than
mercial reality, development efforts are now
10 years in flight engines. The advantage of sin-
focusing on single-crystal processing and
gle-crystal alloys compared to equiaxed and DS
advanced DS alloy development. Single-crystal
alloys in low-cycle fatigue (LCF) is shown in
airfoils offer the potential to further improve
Figure 8. GE is currently evaluating and Rainbow
component high-temperature material
rotor testing some of these single-crystal alloys
strength and, by control of crystal orientation,
can provide an optimum balance of properties. for application in our next generation gas tur-
bines.
In single-crystal material, all grain boundaries
are eliminated from the material structure and The continuing and projected temperature
a single crystal with controlled orientation is capability improvements in bucket material
produced in an airfoil shape. By eliminating all capabilities are illustrated in Figure 9. Together
grain boundaries and the associated grain with improved coatings, these new bucket
boundary strengthening additives, a substantial materials will provide continued growth capa-
increase in the melting point of the alloy can be bility for GE gas turbines in the years to come.
50
40
Relative Life
30
20
10
0
IN738 GTD-111 GTD-111 2nd Gen 1st Gen 2nd Gen
Equiaxed Equiaxed DS DS SC SC
Figure 8. Bucket alloys — LCF life
Temp
1650 900
Single
Crystal
880
1600
1500 820
1980 1985 1990 1995 2000
Year of Introduction
Figure 9. Continuing improvements in bucket materials capability
60
ion
Coating Life Requirements
dat
Life = Thousand Hours
50
40
Oxi
IN738 + PtAl Coating ue
30
atig
rmal F
20 IN738 Uncoated
The
10
U700 Uncoated
0
0.5 1.0 1.5 2.0
Hot Corrosion
Equivalent Sodium (Fuel, Air, Water Mix), ppm
50% 90%
Percentiles Nat. Gas: 50% 90%
for Commonly True Distillates:
Used Fuels Treated Ash Forming: 1970 1980 1990 2000
50% 90%
Year
Figure 10. Effect of sodium on bucket Figure 11. Bucket coating requirements
corrosion life and coating evolution
High-temperature hot corrosion has been and a denuded zone of base metal. Low-tem-
known since the 1950s. It is an extremely rapid perature corrosion characteristically shows no
form of oxidation that takes place at tempera- denuded zone, no intergranular attack, and a
tures between 1500°F/816°C and layered type of corrosion scale.
1700°F/927°C in the presence of sodium sul- The lines of defense against both types of cor-
fate (Na2SO4). Sodium sulfate is generated in rosion are similar. First, reduce the contami-
the combustion process as a result of the reac- nants. Second, use materials that are as corro-
tion between sodium, sulfur and oxygen. sion-resistant as possible. Third, apply coatings
Sulfur is present as a natural contaminant in to improve the corrosion resistance of the
the fuel. bucket alloy.
Low-temperature hot corrosion was recognized
as a separate mechanism of corrosion attack in
High-Temperature Oxidation
the mid-1970s. This attack can be very aggres- Metal oxidation occurs when oxygen atoms
sive if the conditions are right. It takes place at combine with metal atoms to form oxide
temperatures in the 1100°F/593°C to scales. The higher the temperature, the more
1400°F/760°C range and requires a significant rapidly this process takes place, creating the
partial pressure of SO2. It is caused by low potential for failure of the component if too
melting eutectic compounds resulting from much of the substrate material is consumed in
the combination of sodium sulfate and some the formation of these oxides. Figure 14a shows
of the alloy constituents such as nickel and the microstructure of a coated bucket that has
cobalt. It is, in fact, somewhat analogous to the seen about 30,000 hours of service. At the tem-
type of corrosion called Fireside Corrosion in peratures seen in this region of the airfoil, no
coal-fired boilers. significant oxidation attack of the coating can
be seen.
The two types of hot corrosion cause different
types of attack, as shown in Figures 12 and 13. By contrast, Figure 14b shows the microstruc-
These are metallographic cross sections of cor- ture of the same type of coating, which has
roded material. High-temperature corrosion been severely attacked after about the same
features intergranular attack, sulfide particles length of service. At the higher temperatures,
Sulfide Spikes
Low-Temp.
Corrosion Scale
25µ
25µ
Figure 12. Hot corrosion (high-temperature type) Figure 13. Hot corrosion (low-temperature type)
Sound Sound
Substrate Substrate
Coating Coating
Figure 14a. Photomicrograph showing sound Figure 14b. Photomicrograph of a coating on a bucket
microstructure of a coated bucket that material showing internal oxidation of
has been in service coating (dark particles)
which must have been present in the Figure 14b However, many high-strength superalloys in
case, insufficient aluminum was available in use today cannot form sufficient protective
the coating to maintain a protective oxide at scales because the compositional requirements
the surface, and oxygen was able to diffuse for achieving other properties, such as high
into the interior of the coating structure where strength and metallurgical stability, do not
it formed discrete, discontinuous, aluminum allow for the optimization of oxidation/corro-
oxide particles. This phenomenon is known as sion resistance in the superalloy itself.
internal oxidation. Such a situation quickly Therefore, most of today’s superalloys must
depletes the coating of its available aluminum, receive their oxidation protection from special-
rendering it non-protective. ly engineered coatings.
able today have lives that are 10 to 20 times been developed and applied to the stage 1
longer than the first-generation coatings under buckets of the higher firing temperature
a wider diversity of corrosion and oxidizing machines, such as the 7FA and 9FA machines.
conditions. This coating possesses even greater high-tem-
perature oxidation capability than the GT-29
GE has used two basic classes of coatings during
IN-PLUS™.
the past 25 years. The first class used was a diffu-
sion-style coating called platinum aluminide A comparison of stage 1 bucket coatings is
(PtAl). The second class is an overlay-style coat- shown in Figure 15, while a more detailed
ing such as PLASMAGUARD™ GT-29 IN- description of each is in the following sections.
PLUS™.
Platinum-Aluminide Coatings
The development of each of these coating sys- All stage 1 buckets have been coated since the
tems was in response to field needs. The plat- late 1970s. Up until mid-1983, the coating used
inum aluminide was the original heavy-duty by GE on most stage 1 buckets was a platinum-
coating and addressed corrosion problems aluminum (PtAl) diffusion coating. This coat-
seen by a large segment of the fleet in the
ing was selected over the straight aluminide
1960s. It doubled the corrosion life of the
coatings because it provided superior corro-
uncoated IN-738 buckets of that time. The
sion resistance both in burner test rigs and in
PLASMAGUARD™ GT-29 coating improved
field trials. The platinum-aluminum coating is
that corrosion resistance by a further 50%.
applied by electroplating a thin (0.00025
That same high level of hot-corrosion resist-
inch/0.006 mm) layer of platinum uniformly
ance is kept in the more recent PLASMA-
onto the bucket air-foil surface, followed by a
GUARD™ GT-29 PLUS, which also has sub-
pack diffusion step to deposit aluminum. This
stantially more oxidation resistance, as
results in a nickel-aluminide coating with plat-
required by the more advanced machines.
PLASMAGUARD™ GT-29 IN-PLUS™ is a two- inum in solid solution or present as a PtAl2
layer coating, with the top layer also applied to phase near the surface.
the internal surface of the bucket. Most recent- The platinum in the coating increases the
ly, GT-33 IN-COAT™ and IN-PLUS™ have activity of the aluminum in the coating,
GT-29
GT-29 PLUS
GT-29 IN-PLUSTM
GT-33 IN-PLUSTM
enabling a very protective and adherent Al2O3 IN-PLUS™ and GT-33 IN-PLUS™ for cooled
scale to form on the surface. or hollow vaned buckets.
A Rainbow example of comparative corrosion PLASMAGUARD™ coatings are examples of
on PtAl-coated and uncoated IN-738 buckets, overlay coatings and differ from diffusion coat-
run side-by-side in the same machine under ings, such as the platinum-aluminum coatings,
corrosive conditions, is shown in Figure 16. The in one major respect. At least one of the major
two buckets were removed for interim evalua- constituents, (generally nickel) in a diffusion
tion after 25,000 hours of service. This unit coating is supplied by the base metal. An over-
burned sour natural gas containing about lay coating, on the other hand, has all the con-
3.5% sulfur and was located in a region where stituents supplied by the coating itself. The
the soil surrounding the site contained up to advantage of overlay coatings is that more var-
3% sodium. ied corrosion resistant compositions can be
The uncoated IN-738 bucket has penetration applied since the composition is not limited by
extending 0.010 to 0.015 inches (0.25 to 0.4 the base metal composition, nor is thickness
mm) into the base metal over most of the limited by process considerations.
bucket surface. The coated bucket generally PLASMAGUARD™ coatings are applied by the
shows no evidence of base metal hot corrosion Vacuum Plasma Spray (VPS) process in equip-
attack, although some of the bucket areas ment especially designed to apply this coating
showed coating thinning. Only at some very in a uniform and controlled manner to GE
small locations on the leading edge of the buckets. In this process, powder particles of
coated bucket was the coating breached and the desired composition are accelerated
then to only a depth of 0.001 to 0.002 inches through a plasma jet to velocities higher than
(0.025 to 0.05 mm). those achievable through atmospheric plasma
spray methods. (See Figure 17.) The solidifica-
PLASMAGUARD™ Coatings tion of the powder onto the airfoil results in a
The latest GE-developed and patented much stronger coating bond than can be
PLASMAGUARD™ coatings are now GE’s stan- achieved by using conventional atmospheric
dard stage 1 bucket coatings — GT-29 PLUS™ plasma spray deposition because of the higher
and GT-33 PLUS™ for solid buckets; GT-29 particle speeds and the cleaner, hotter sub-
Figure 16. Stage 1 turbine buckets: coated and Figure 17. PLASMAGUARD™ GT-20 coated shroud
uncoated IN-738; 25,000 service hours
25µ
Figure 18. VPS production facility Figure 19. VPS coating after more than 40,000 hours
turbine exposure – pressure face
resistance to cracking than the GT-29 composi- PLASMAGUARD™ GT-20 is being used to
tion. This coating may also be used with an coat the surface of this high temperature,
outer layer enriched with aluminum to provide inner shroud component. (See Figure 20.) This
maximum long-term life. PLASMAGUARD™ coating was developed and has been used
GT-33 is currently being introduced in the F extensively by GE Aircraft Engines on its flight
class machines. engine shrouds. It provides an extremely oxi-
dation-resistant surface and a rub-tolerant
Coatings for bucket refurbishment have also
coating in the event that the bucket blade tips
been introduced recently. These coatings,
rub against the stationary shroud.
known as EXTEND-COAT™, are based upon
the GT-29 and GT-33 PLASMAGUARD™ com- Future Coatings
positions and were developed to be applied to Coating development work is continuing at
serviced hardware. Several GE Service Centers GE, aiming at further improvements to the
have been qualified to apply these coatings for oxidation- resistance and thermal fatigue resist-
the service market. ance of high-temperature bucket coatings. In
Low-Temperature Coatings addition to these environmentally resistant
coating development efforts, work is also
Low-temperature coatings find their greatest
underway to develop advanced thermal barrier
need in latter stage buckets and in stage 1
coatings (TBCs) for application to stationary
buckets of machines that run a substantial por-
and rotating gas path components. By careful
tion of their duty cycle at part load.
process control, the structure of these TBCs
For instance, the stage 3 buckets of the 7FA may be made more resistant to thermal fatigue
and 9FA machines are currently coated with a and their lives greatly extended. Rainbow rotor
diffused chromide coating which, although not testing of some of these coatings is currently in
suitable for higher temperature stages, will progress.
impart substantial protection against both cor-
rosion and oxidation at the lower tempera- Combustion Hardware
tures of this part. In addition, a PLASMA- The combustion system is a multiple-chamber
GUARD™ GT-43 coating composition has assembly composed of three basic parts: the
been developed, after an extensive laboratory
corrosion rig and mechanical testing program,
for use in severe low temperature corrosion
applications. This GE-patented coating, also
applied by the same VPS process, has shown
excellent performance in Rainbow rotors, con-
firming its laboratory corrosion resistance.
Shroud Coatings
New gas turbine models such as the 6FA,
7FA and 9FA operate at considerably higher
temperatures than previous heavy-duty gas
turbines. Therefore, to provide a durable Figure 20. 7FA PLASMAGUARD™ GT-20 coated
stage 1 bucket stationary shroud component, shroud
100 7.0
90 6.0 CHARACTERISTICS
80 5.0
70 N263 • 15—25 Mil (380—640 Micron) Thickness
Kg/cm2 x 10-3
60 4.0
Stress KSI
Figure 21. Rupture comparison, N-263 vs. Figure 22. Thermal barrier coatings
Hallestoy-X vs. 309SS
applied over the bond coat. The total thickness In the early 1980s, a new material, Nimonic
used is 0.015 to 0.025 inch (0.4 to 0.6 mm). 263, was introduced into service for transition
Characteristics and advantages of the TBCs are pieces on the MS7001 and MS9001 models.
shown in Figure 22, and the microstructure and This material is a precipitation-strengthened,
a coated liner are shown in Figure 23. The pri- nickel-base alloy with higher strength capabili-
mary benefit of the TBCs is to provide an insu- ty than Hastelloy-X. It was extensively tested in
lating layer that reduces the underlying base the Gas Turbine Combustion Development
material temperature and mitigates the effects Laboratory and successfully tested in Rainbow
of hot streaking or uneven gas temperature combustion hardware. Nimonic 263 transition
distributions. These coatings are now standard pieces have accumulated more than 25,000
on a number of GE gas turbines and have hours of successful experience in MS7001
demonstrated excellent performance in pro- machines. The Nimonic 263 material is being
duction machines. phased into the higher firing temperature gas
turbine models and will be used in future
Transition Pieces
uprated machines.
Although less complicated than the liners, the
Since the early 1980s, TBCs have been applied
transition pieces have probably been more
to the transition pieces of the higher firing
challenging from a materials/processes stand-
temperature gas turbine models and to uprat-
point. Therefore, new materials have tended to
ed machines. Field experience over thousands
be first introduced on the transition piece.
of hours of service has demonstrated good
From a design standpoint, significant improve-
durability for this coating on transition pieces.
ments have been made on advanced models
A recent improvement has also been made to
through the use of heavier walls, single-piece
aft ends, ribs, floating seal arrangements, selec- increase the wear resistance of some transition
tive cooling, etc. These design changes have pieces in the aft end or picture frame area.
been matched by material improvements. Cobalt-base hard coatings applied by thermal
Initial 1950s transition pieces were made from spray have been tested in field machines and
AISI 309 stainless steel. In the early 1960s, the best spray has been shown to improve the
nickel base alloys Hastelloy-X and RA-333 were wear life of sealing components by more than
used in the more limiting parts. These alloys four times. The selected coating, called
became standard for transition pieces by 1970. Extendor™, is available for many of the cur-
rent gas turbine models to extend the wear life
of these components. This improvement in
Top Coat
transition piece seal wear is now also being
Bond Coat incorporated into many of the new production
machines.
formed to verify that all design requirements centrations of alloying elements than Alloy
are met. 718, and is there-fore possible to produce in
Each turbine wheel is spin-tested prior to its the very large ingot sizes needed for the large
installation into a rotor in a facility such as the 7FA and 9FA wheel and spacer forgings. (See
one shown in Figure 24. Alloy steel wheels, Figure 27.)
because of the brittle-to-ductile transition tem- Cr-Mo-V Alloy
perature phenomenon, are first spun cold to
Turbine wheels and spacers of most GE single
verify the absence of critical size defects. All
shaft heavy-duty gas turbines are made of
wheels, including the cold spun steel wheels,
1% Cr - 1.25% Mo - 0.25% V steel. This alloy is
are hot spun at bore stresses slightly above
used in the quenched and tempered condition
their yield strength to induce residual com-
to enhance bore toughness. Stress rupture
pressive stresses in the bore region. These spin-
strength of the dovetail region (periphery) is
ning operations, in addition to the stringent,
controlled by providing extra stock at the
non-destructive testing performed both before
periphery to produce a slower cooling rate
and after spin testing, provide maximum assur-
during quenching. The stress rupture proper-
ance against brittle fracture in service.
Stress
TURBINE WHEEL ALLOYS 200 14.0
6.0 IN-706
60 A-286
4.0
This nickel-based, precipitation-hardened alloy
KSI
M-152
is the newest to be used in turbine wheel appli- 2.0 Cr-Mo-V
20
cation. It is the 7FA, 9FA, 6FA and 9EC turbine
0 0
wheel and spacer alloy, and it offers a very sig- 450 500 550 600
Temp. °C
nificant increase in stress rupture and tensile 100,000 Hr.
Life 800 900 1000 1100 1200
yield strength compared to the other wheel
°F
alloys. (See Figures 25 and 26.) This alloy is simi- Figure 25. Stress rupture comparison (turbine
lar to Alloy 718, an alloy that has been used for wheel alloys)
wheels in aircraft turbines for more than 20
years. Alloy 706 contains somewhat lower con-
140 10.0
IN-706
Stress kg/cm2 x 10-3
120 A-286
8.0
Stress KSI
0.2%
Yield M-152
100
Strength CrMoV
6.0
80
60
4.0
0 100 200 300 400 500
Temp.°C
Figure 24. Spin test facility (Greenville plant) Figure 26. Tensile yield strength comparison
(turbine wheel alloys)
forging flow studies, the necessary sub-scale 403. Superior corrosion resistance is also
wheel forging experiments and the extensive achieved due to its higher concentration of
mechanical and physical property determina- chromium and molybdenum. Compressor cor-
tions necessary to bring into production a new rosion results from moisture containing salts
wheel material. and acids collecting on the blading. During
This development effort has resulted in the operation, moisture can be present because of
production of the largest ingots ever made and rain, use of evaporative coolers or condensa-
forged into high quality qualification turbine tion resulting from humid air being accelerat-
wheel and spacer forgings. Concurrent with ed at the compressor inlet. Moisture may be
the process development effort was an effort to present in the compressor during operation
develop new non-destructive techniques to up to between stage 5 and stage 8, where it
inspect these turbine forgings to greater levels usually becomes warm enough to prevent con-
of sensitivity than ever before possible. These densation. When the turbine is not in opera-
new ultrasonic inspection techniques are being tion, the compressor can still become wet if
applied to all the Alloy 706 and the Alloy 718 metal temperatures are below the local dew
turbine forgings to ensure an even greater point. (This can happen to units stored in
level of confidence in these high strength forg- humid environments.) The chemistry of this
ings. moisture deposit on the blading determines
the severity of the corrosion phenomenon.
Additional development efforts continue to
improve the current processing of other forg- In the early 1960s, GE first experienced corro-
ings by working with our suppliers on the fur- sion pitting on bare 403 in oil platform appli-
ther optimization of properties and forging cations when several machines developed pits
quality. In-process, non-destructive evaluation and failed compressor blades. Generally, the
of all rotor components continues to be service time on these machines ranged from
emphasized as a critical aspect to produce 20,000 to 60,000 hours. As a result of this expe-
quality forgings. rience, GE adopted NiCd coating for use in
selected applications, and later for all compres-
Compressor Blades sor blades in the "wet" stages (normally up to
Compressor blading is variously made by forg- stage 8). However, because of recent, more
ing, extrusion or machining. All production stringent EPA requirements, this coating has
blades, until recently, have been made from now been replaced by a new GE developed
Type 403 or 403 Cb (both 12 Cr) stainless and patented coating called GECC-1. This new
steels. During the 1980s, a new compressor aluminum slurry coating has a protective
blade material, GTD-450, a precipitation hard- ceramic top layer that provides improved ero-
ened, martensitic stainless steel, was intro- sion resistance. (See Figure 28.) This coating has
duced into production for advanced and accumulated more than 100,000 hours of field
uprated machines, as shown in Table 1. This testing and has shown to be equal to or better
material provides increased tensile strength than conventional aluminum slurry coatings in
without sacrificing stress corrosion resistance. corrosion protection and substantially better in
Substantial increases in the high-cycle fatigue erosion resistance. This coating has been
and corrosion fatigue strength are also applied by GE Service Shops as a refurbish-
achieved with this material, compared to Type ment coating for several years and is now
being applied to all new units. All IGVs and ting. These test results, shown in Figure 29,
the first three stages of rotating and stationary indicate that uncoated GTD-450 without a
airfoils in the compressor will be made from coating is equivalent or better than Al or NiCd
GTD-450; the next five stages will be made of coatings for acidic corrosion resistance. Field
GTD-450 for the F class machines and GECC-1 experience of more than 48,000 hours has
coated AISI 403 or 403 Cb for the other confirmed the excellent corrosion resistance of
machines. The rest of the blading will be AISI uncoated GTD-450. These tests have also
403 or 403 Cb uncoated. This change will pro- shown that conventional aluminum slurry coat-
vide GE’s machines with better corrosion and ings can suffer erosion damage and leave sig-
erosion protection and eliminate cadmium nificant areas of the blading unprotected.
from the environment. Therefore, in machines where erosion may be
GTD-450 is a precipitation-hardened, marten- experienced, GECC-1 on 12 chromium blades,
sitic stainless steel with excellent aqueous cor- or uncoated GTD-450, is recommended. The
rosion resistance. Laboratory tests have shown GTD-450 material should not be used coated,
that GTD-450, in very acidic salt environments as coating will decrease fatigue life.
(pH~4), possesses excellent resistance to pit-
Casings
For all models except the F-technology
GTD-450 Bare machines, the entire "tube" surrounding the
gas turbine rotor is composed of a series of
Al Slurry Coatings
cast iron castings bolted together end-to-end.
NiCd + Topcoats The castings (inlet and compressor) at the for-
ward end of the machines are made of gray
NiCd iron, while those at the aft end (discharge and
turbine shell) are generally made of ductile
Bare
iron or, in some, steel castings or fabrications.
0 2 4 6 8 10 The excellent castability and machinability
Worst Best offered by cast iron makes it the obvious
Figure 29. Acidic laboratory tests choice for these somewhat complex parts that
have close tolerances. Cast iron is less prone to from sand castings. Cast iron is again used
hot tears and shrinkage problems than cast where possible; however, where higher temper-
steel. Experience has also shown it to provide a ature or planned welding is encountered, steel
higher degree of dimensional stability during is employed. For example, Cr-Mo-V has been
shop processing. used for support rings where temperatures
Although stress is important in determining reach 1000°F/538°C, and carbon steel has
which of the two types of cast iron (gray or been used for bearing housings requiring weld
ductile) is used in the castings, operating tem- fabrication.
perature is of prime importance. Gray iron is Quality is a key factor in the successful opera-
generally limited to applications where temper- tion of any part, and sand castings are no
atures do not exceed 450°F/239°C, ductile exception. From the conceptual stage, quality
iron to applications no greater than is built into these parts. Foundry personnel are
650°F/343°C. In the case of gray iron, GE uses called in early in the design stage to provide
a type that has a minimum tensile strength of the best possible castability consistent with
30 ksi (2.1 kg/cm2 x 10-3), similar to ASTM- functional requirements. Before any casting is
A48, Class 30. Ductile iron, on the other hand, granted production approval, a process must
is a ferritic type [60 ksi (4.2 kg/cm2 x 10–3) be found that produces three consecutive cast-
TS, 40 ksi (2.8 kg/cm2 x 10–3) YS, 18% E1], ings meeting rigid X-ray inspection require-
similar to ASTM-A395. The 7FA and 9F ments. Once such a process is found, it is pre-
machines utilize ductile iron for the inlet and cisely documented and must be followed for
compressor casing and a fabricated CrMo steel all subsequent production. Recently, a sonic
combustion wrapper and turbine shell. More screening procedure was developed to supple-
recently, cast 2 1/4 Cr - 1Mo steel is being intro- ment X-ray inspection. It was designed to
duced into the F-technology machines for the reduce inspection time and increase coverage
combustion wrapper and turbine shells. while maintaining strict standards of casting
integrity.
Future Materials
In addition to the X-ray/sonic monitoring of
Advances in ductile iron have been made in casting visual examinations, magnetic particle
laboratory trial castings that will enable this inspection and, in the case of bearing hous-
material to be extended to higher temperature ings, leak tests, are always employed. All these
applications. These trial heats have shown the combine to provide a very comprehensive
capability to extend the useful temperature of quality check on sand cast components.
this material by 100°F/56°C. This development
program is now in the Rainbow field trial Inlet and Exhaust Systems
phase and will most likely find application in
advanced and uprated GE gas turbines.
Inlet Systems
The inlet system environment is ambient air
Additional Sand Castings with low velocity air flow over interior surfaces.
In addition to the casings, several other large Materials of construction are generally low car-
components, such as bearing housings, inner bon steel, including the inter baffles used over
barrels, support rings and diaphragms in the acoustic material to reduce the noise level. In
stator section of the turbine, are produced selected marine environments, a corrosion-
Summary
The purpose of this paper has been to describe some of the materials currently being used in GE
gas turbines and to verify our commitment to continued GE leadership in material and process
development. The activities described in this paper are by no means complete. Major materials
development work is underway at GE to provide a continuous stream of new and improved materi-
als for gas turbine application to meet our customers’ needs for the most efficient gas turbines.
GE’s intent is to provide the materials necessary for the advancement of turbine firing temperatures
while maintaining the high levels of unit reliability, availability and maintainability.
List of Figures
Figure 1. Investment cast buckets
Figure 2. Investment cast nozzles
Figure 3. Directionally solidified buckets
Figure 4. Firing temperature trend and bucket material capability
Figure 5. Advanced air cooling technology
Figure 6. Stress rupture comparison — bucket and nozzle materials
Figure 7. Directionally solidified GT-111 vs. equiaxed
Figure 8. Bucket alloys — LCF life
Figure 9. Continuing improvements in bucket materials capability
Figure 10. Effect of sodium on bucket corrosion life
Figure 11. Bucket coating requirements and coating evolution
Figure 12. Hot corrosion (high-temperature type)
Figure 13. Hot corrosion (low-temperature type)
Figure 14a. Photomicrograph showing sound microstructure of a coated bucket that has been in
service
Figure 14b. Photomicrograph of a coating on a bucket material showing internal oxidation of coat-
ing (dark particles)
Figure 15. Comparative resistance in types of coatings
Figure 16. Stage 1 turbine buckets: coated and uncoated IN-738; 25,000 service hours
Figure 17. PLASMAGUARD™ GT-20 coated shroud
Figure 18. VPS production facility
Figure 19. VPS coating after more than 40,000 hours turbine exposure — pressure face
Figure 20. 7FA PLASMAGUARD™ GT-20 coated shroud
Figure 21. Rupture comparison, N-263 vs. Hallestoy-X vs. 309SS
Figure 22. Thermal barrier coatings
Figure 23. Thermal barrier coated liner, Hallestoy-X vs. 309SS
Figure 24. Spin test facility (Greenville plant)
Figure 25. Stress rupture comparison (turbine wheel alloys)
Figure 26. Tensile yield strength comparison (turbine wheel alloys)
Figure 27. 7FA IN-706 turbine forging
Figure 28. GECC-1 compressor blade coating
Figure 29. Acidic laboratory tests
List of Tables
Table 1 High-temperature alloys