Aerospace Science and Technology 33 (2014) 100106

Contents lists available at ScienceDirect

Aerospace Science and Technology

www.elsevier.com/locate/aescte

Effect of low-temperature overload on fatigue crack growth retardation

and prediction of post overload fatigue life
Vaneshwar Kr. Sahu a , J.K.S. Anil Kumar a , J.R. Mohanty c , B.B. Verma b , P.K. Ray a,
a
b
c

Department of Mechanical Engineering, National Institute of Technology, Rourkela 769008, India

Department of Metallurgical and Materials Engineering, National Institute of Technology, Rourkela 769008, India
Department of Mechanical Engineering, Veer Surendra Sai University of Technology, Burla 768018, India

a r t i c l e

i n f o

Article history:
Received 19 June 2013
Received in revised form 11 January 2014
Accepted 14 January 2014
Available online 20 January 2014
Keywords:
Crack growth model
Fatigue crack growth
Overload cycle
Retarded crack length
Retardation parameters

a b s t r a c t
In the present work an attempt has been made to study the simultaneous effect of overload and
low temperature on fatigue crack growth behavior. It is known that overload application retards a
propagating fatigue crack. It has been observed in this study that application of the overload at low
temperature further enhances the magnitude of retardation. Various factors affecting retardation have
been analyzed and crack growth behavior is predicted using a modied exponential model. It is observed
that the proposed model estimates the crack growth rate and life accurately. The theoretically estimated
retardation parameters have been also found to be in good agreement with their experimental values.
2014 Elsevier Masson SAS. All rights reserved.

1. Introduction
Manufacturing, processing and fabrication introduce some defects and aws in engineering materials. The application of cyclic
and uctuating loads may initiate fatigue crack and subsequent
crack extension from these stress concentration sites. Attainment
of a critical length makes the component unreliable and may cause
complete failure of the structure. Estimation of crack growth behavior under such loading conditions is essential for stability and
safety of structure.
The loading condition, operating temperature and material
properties play important role in fatigue initiation and crack
growth behavior. It is known that introduction of spike or band
overload retards a growing fatigue crack resulting in enhancement
of the life of the structure and component. Aircrafts, ships, offshore structures etc. experience such load interactions during their
service. The introduction of overload cycle/s has been studied extensively and it is known to retard a growing fatigue crack [2,5,
14,26,30,31]. The magnitude of overload induced retardation is a
function of overload ratio, position of application of the overload,
material properties and the operating temperature.
Wheeler [30] proposed that retardation is achieved due to the
compressive residual stress eld due to the application of overload

Corresponding author. Tel.: +91 661 2462518.

E-mail addresses: pkray@nitrkl.ac.in, prabal_kray@yahoo.com (P.K. Ray).

1270-9638/$ see front matter 2014 Elsevier Masson SAS. All rights reserved.
http://dx.doi.org/10.1016/j.ast.2014.01.005

cycle/s, and modied the ParisErdogan relation [20] by introducda

ing a crack retardation parameter C p , dN
= C p ( K )n [31].
Later, Ray et al. [25] introduced the concept of a plastic zone
correction factor. It is argued that the presence of a monotonic
compressive residual stress eld ahead of the crack tip reduces the
size of the plane stress cyclic plastic zone, embedded in the monotonic plastic zone.
Though Wheelers model is successful in predicting the crack
growth behavior subjected to single overload cycle, it fails to predict the same under a spectrum loading condition where loading
sequence consists of underload and overload spikes. It has long
been recognized that the transient crack growth behavior following
the application of overload is often controlled by several concurrent mechanistic processes [29]. The plasticity induced crack closure, crack tip blunting/deection and residual compressive stress
eld are prominent retardation mechanisms [27]. More recently
Pommier [23] has argued that Bauschinger effect may cause reversed plasticity within the region of overload plastic zone which
reduces the magnitude of overload retardation.
A few models are available in literature to predict the fatigue
life of components subjected to overload cycle [2,5,14,26,30,31].
The earliest of these are based on calculation of the yield zone
ahead of the crack tip. Wheeler model [30] and Willenborg model
[31], for example, fall in this category. Elber [6], Budiansky and
Hutchinson [4] had proposed their models based on crack closure mechanism while Newman proposed a model based on strip

V.K. Sahu et al. / Aerospace Science and Technology 33 (2014) 100106

101

Nomenclature
a
ai
aj
aol
ad
aPd

aEd
A, B , C 
B
C
da/dN
E

ys
f (g)
K
KC

crack length . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
crack length corresponding to the ith step . . . . . .
crack length corresponding to the jth step . . . . . .
crack length at overload . . . . . . . . . . . . . . . . . . . . . . . . . . .
retarded crack length . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
retarded (predicted) crack length . . . . . . . . . . . . . . . . .

mm
mm
mm
mm
mm
mm

retarded (experimental) crack length . . . . . . . . . . . . . mm

& D  material constants in the Exponential Model
plate thickness . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . mm
constant in the Paris equation
crack growth rate . . . . . . . . . . . . . . . . . . . . . . . . . . . mm/cycle
modulus of elasticity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . MPa
yield strength . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . MPa
geometrical factor

stress intensity factor . . . . . . . . . . . . . . . . . . . . . . . . . MPa m

plane stress fracture toughness . . . . . . . . . . . . . . MPa m

K IC
K max
B
K max
K ol
K
 K eff
l
m
mi j
n
N
Ni
Nj
Nd
N dP

plane strain fracture toughness . . . . . . . . . . . . . . MPa m

maximum stress intensity factor . . . . . . . . . . . . . MPa m
maximum (base line) stress intensity factor . MPa m
stress intensity factor at overload . . . . . . . . . . . . MPa m
stress intensity factor range . . . . . . . . . . . . . . . . . . MPa m
effective stress intensity factor range . . . . . . . . MPa m
dimensionless factor in the Exponential Model
specic growth rate
specic growth rate corresponding to the interval i j
constant in the Paris equation
number of cycles or fatigue life
number of cycles corresponding to the ith step
number of cycles corresponding to the jth step
number of delay cycle
number of delay cycle (predicted)

N dE

number of delay cycles (experimental)

Table 1
Chemical composition (in wt%).
Al

Cu

Mg

Mn

Fe

Si

Zn

Cr

Others

90.794.7

3.84.9

1.21.8

0.30.9

0.5

0.5

0.25

0.1

0.15

Table 2
Mechanical properties.
Tensile
strength (ut )
(MPa)

Yield
strength (ys )
(MPa)

Modulus of
elasticity (E)
(MPa)

Poissons
ratio ( )

Plane strain fracture

toughness (K IC )

(MPa m)

Plane stress fracture

toughness (K C )

(MPa m)

Elongation

469

324

73,100

0.33

37.0

95.31

19% over
12.7 mm
GL

yield [19]. In addition, there are some retardation models [1,12,24]

based on damage accumulation and strain energy release rate. In
recent years Mohanty et al. [1518] have proposed an exponential
model based on specic growth rate and loading history.
Working environment and working temperature are among the
various factors that inuence the fatigue life of a material. Effect
of working temperature on fatigue life was rst initiated by Zambrow and Fontana [32], who dealt with measurements on magnesium and aluminum alloys and on various stainless steels and
observed that the fatigue lives increased with lowering of temperature in contrast to the principle of linear elastic fracture mechanics. It was thought that such increase in fatigue life was due
to their increased fracture strength. During the tests conducted by
Kwon et al. [11], it was observed that oxidizing environment increased cyclic strain hardening, ultimately increasing fatigue life.
Presence of vacuum also causes increased fatigue life of a component than that in air [13]. It was further observed that [7,10,
21,28] fatigue life of most FCC materials increased with decrease
in temperature whereas, for BCC materials it increased up to DBT
after which it became even worse than that at room temperature.
It is well established that temperature affects the mechanical
properties of the metallic materials. In the present investigation
combined effect of both monotonic overload and low temperature
at the point of application of overload has been studied. Aerospace
structures are commonly exposed to this situation during their services. Therefore, the present study is carried out for commercially
important 2024-T3 Al-alloy. The life estimation and crack growth
prediction are important for such structures. An exponential model
[16] has been used and attempted for these predictions.

2. Experimental procedure
The present study was conducted on 2024-T3 Al-alloy whose
chemical composition and mechanical properties are given in Tables 1 and 2 respectively. Single-edge notched tension (SENT) specimens having a thickness of 6.5 mm were used for conducting
the fatigue tests. The specimens were made in the longitudinal
transverse (LT) direction. The detail geometry of the specimens is
presented in Fig. 1.
The experiments were performed in a servo-hydraulic dynamic
testing machine, Instron-8502, having a load capacity of 250 kN,
interfaced to a computer for machine control and data acquisition.
All the fatigue tests were conducted in air and at room temperature except during overloading. The test specimens were fatigue
precracked under mode-I loading to an a/ w ratio of 0.3 and were
subjected to constant load fatigue test (i.e. progressive increase in
 K with crack extension) maintaining a load ratio, R = 0.1. Sinusoidal loads were applied at a frequency of 6 Hz. The crack growth
was monitored with the help of a COD gauge mounted on the face
of the machined notch. The following equation is used to determine stress intensity factor, K [3].

K = f (g)

wB

(1)

where,

f ( g ) = 1.12 0.231(a/ w ) + 10.55(a/ w )2 21.72(a/ w )3

+ 30.39(a/ w )4

(2)

The fatigue crack was allowed to grow up to a/ w ratio of 0.4

and subsequently subjected to single spike overload cycle at a

102

V.K. Sahu et al. / Aerospace Science and Technology 33 (2014) 100106

Table 3
Load scenarios of the tested specimen.
F max
(kN)

F min
(kN)

F max(ol)
(kN)

R ol

ai
(mm)

aol
(mm)

af
(mm)

7.305

0.7305

15.341

2.10

17.75

20.40

32.40

chamber walls. Temperature was measured with a sixteen channel

RTD indicator with a temperature range of 120 C to 400 C.
3. Determination of fatigue crack growth rate
The experimental crack growth data are usually scattered and
need smoothening for analysis. Various methods have been proposed to minimize the scatter and present the crack growth for
life prediction. One such method proposed by Mohanty et al. [18],
has been adopted in this work for data smoothening and determination of the crack growth rate which is described below.
It has been established [18] that the experimental aN data can
be well tted by an exponential equation of the form:

a j = a i emi j ( N j N i )

(3)

where,
ai and a j = crack lengths in ith step and jth step in mm respectively,
N i and N j = Nos. of cycles in ith step and jth step respectively,
mi j = specic growth rate in the interval i j,
i = No. of experimental steps, and j = i + 1.
Fig. 1. Single edge notched specimen geometry.

loading rate of 8 kN/min. The specimens were subjected to mode-I

K
overloads at an overload ratio (R ol = Bol ) of 2.10 and at differK max

B
ent temperatures of +25, 0, 28, 43, 53, 83 C, where, K max

is the maximum stress intensity factor for base line test. This is
followed by fatigue cycling under constant amplitude loading condition. Subsequently all the specimens were subjected to fatigue
test in mode-I and at constant load condition. The load scenarios
of the tests are presented in Table 3.
Low temperature was attained by passing liquid nitrogen
through a cooling chamber surrounding the specimen. The cooling
chamber consisted of two inlet ports in lower end and two outlet
ports in upper end connected by blind holes as shown in Figs. 2(a)
and 2(b). A gap of 4 mm was maintained on either sides of the
specimen in order to avoid direct physical contact with cooling

(a)

In the above equation the exponent, known as specic growth

rate, mi j is an important parameter which can be obtained by taking logarithm of Eq. (3) as follows:
a

mi j =

ln( a j )
i

(4)

(N j N i )

The raw values of specic growth rate (mi j ) from experimental

aN data are calculated using the above equation. These are then
tted with corresponding crack lengths by a polynomial curvet which gives a 3rd degree polynomial equation of m vs. a. To
get a better result, crack lengths (modied) at small increments
(0.005 mm) are obtained in Excel sheet keeping the initial and nal
values (recorded from fatigue test) intact. Using the above polynomial equation, the new (smoothened) values of mi j are obtained
which can be subsequently used to get the smoothened values of
the number of cycles as per the following equation:

(b)
Fig. 2. (a) Photograph of cooling chamber. (b) Schematic diagram of cooling chamber.

V.K. Sahu et al. / Aerospace Science and Technology 33 (2014) 100106

103

Table 4
Curve tting constants for R ol = 2.10.
A

B

C

D

103298.41 106

48410.452 106

5340.184 106

173.798 106

Table 5
Experimental and predicted results of the tested specimen at 0 C.
aPd
(mm)

aEd
(mm)

% error
in aPd

N dP 103
cycle

N dE 103
cycle

% error
in N dP

2.244

2.133

5.19

15.817

15.510

1.98

83 C). Subsequently, the model was validated by predicting the

fatigue life for the specimen overloaded at 0 C.

It has been already established [1517] that the specic growth
rate m is the controlling parameter of the proposed model which
has been correlated with the ratio of overload temperature to the
room temperature R t , two crack driving parameters K max , and  K ,
as well as the material properties K C , E and ys , as dened by the
equations,

m = A l3 + B l2 + C l + D 

Fig. 3. aN curves for different overload temperatures (OLR2.10).

(7)

where,


l=

K



K max

KC



KC

ys
E

1/4
(Rt)

(8)

Plane stress fracture toughness value was used in the expression as tests were conducted in that state. The plane stress fracture
toughness (K C ) is calculated from plane strain fracture toughness
(K IC ) using an empirical relation proposed by Irwin [9] as follows,

2
2
K C2 = K IC
1 + 1.4IC

(9)

where,

IC =

Fig. 4. da/dN K curves for different overload temperatures (OLR2.10).

Nj =

ln( a j )
i

mi j

+ Ni

(5)

Finally, the crack growth rates (da/dN) are calculated directly by

using the above calculated N values and modied a values as follows:

(a j ai )
=
dN
(N j N i )
da

1
B

K IC

ys

(10)

The calculated values of plane stress fracture toughness are presented in Table 2. The different m and l values for post overload
region are tted by a 3rd degree polynomial. The predicted values of m are obtained from Eq. (7) as per the procedure given
elsewhere [16]. The values of the constants A  , B  , C  and D  corresponding to the overload ratio 2.1 and the temperature range
+25 C to 83 C are tabulated in Table 4.
The predicted values of m for the specimen overloaded at 0 C
are calculated by putting the values of different constants from the
Table 4 in Eq. (7). Then the predicted values of N are calculated
using Eq. (5).
5. Results and discussion

(6)

The smoothened values of aN and da/dN K are presented

in Figs. 3 and 4 respectively which illustrate the combined effect
of both overload and low temperature at the point of application
of overload.
4. Modeling and prediction
The exponential model proposed by Mohanty et al. [16], suitably modied to take into account the overload temperature, was
adopted to develop the model using the experimental data from
ve different overloading temperatures (+25, 28, 43, 53,

The experimental results are presented in Figs. 3 and 4. Fig. 3

shows the variation of crack length (a) as a function of number
of stress cycles (N) for different overloading temperatures. It can
be seen that as the temperature of overload application decreases
the magnitude of retardation increases for the same overload ratio
(OLR = 2.10). Fig. 4 presents da/dN K plots for different overloading temperatures. The predicted and experimentally obtained
retardation parameters, ad and N d , along with percentage errors
are presented in Table 5. It can be observed from the above cited
gures and table that retardation caused due to the application of
overload at low temperature is more than that at room temperature and the magnitude of retardation increases as the temperature
of overload is decreased.

104

V.K. Sahu et al. / Aerospace Science and Technology 33 (2014) 100106

Fig. 5. Comparison of predicted and experimental number of cycle (overload at 0 C).

Fig. 7. Retarded crack length (ad ) for overload at 0 C.

Fig. 6. Comparison of predicted and experimental crack growth rate (overload at

0 C).

The proposed model is validated by comparing the predicted

data with experimental one at 0 C. These are presented through
Figs. 58. Figs. 5 and 6 show the variation of crack lengths (predicted and experimental) with number of cycles and the variation
of crack growth rates with stress intensity factor range respectively.
It can be seen from these gures that the predicted results match
reasonably well with the experimental data. The retarded crack
lengths (ad ) and delay cycles (N d ) have been presented in Figs. 7
and 8 respectively which show the magnitude of retardation at the
0 C overload. Variation of delay cycles with overload temperatures
has been illustrated in Fig. 9.
The tear zones produced on the application of overload at room
temperature and different low temperatures (0, 28, 43, 53,
83 C) were observed under SEM. It revealed that specimens
overloaded at room temperature exhibited microvoid coalescence
(dimpled fracture). This is indicative of ductile crack extension during that stage. Examination of specimens overloaded at low temperatures revealed existence of a large number of secondary cracks.
This can be attributed to the loss of ductility at low temperatures.
An increase in the density of secondary cracks with falling temperature (0 to 80 C) can be visualized in Fig. 10.
According to Griths energy criterion, a crack propagates if
the decrease in strain energy is sucient to form new fracture
surfaces. The formation of secondary cracks during the course of
crack growth reduces the available strain energy required for the
primary crack extension. This causes a decrease in crack growth
(increased retardation).
It is known that in case of cyclic stress condition reversed plasticity is introduced within the monotonic overload plastic zone. It

Fig. 8. Delay cycle (N d ) for overload at 0 C.

Fig. 9. Variation of delay cycles with overload temperature.

has been observed that reversed plastic zone produced in the overload region due to Bauschinger effect is responsible for decrease in
the amount of retardation due to application of an overload [8,
22]. The reversed plasticity reduces the effectiveness of retardation caused by monotonic plastic zone [30]. The overload induced
monotonic plastic zone is dictated by K ol whereas  K is responsible for inducing reversed plasticity. Application of overload at
low temperature decreases the size of overload-induced plastic
zone, responsible for retardation, due to increased value of yield

V.K. Sahu et al. / Aerospace Science and Technology 33 (2014) 100106

105

Fig. 10. Comparison of density of secondary cracks at different conditions of test temperatures. (a) Overloaded at room temperature. (b) Overloaded at 28 C. (c) Overloaded
at 53 C. (d) Overloaded at 83 C.

3. Monotonic and cyclic plastic zones and the size of elastic enclave.
All the above factors make the situation complex and the contribution of each component is dicult to quantify. The delay cycles shown in Fig. 9 reveals the combined effect of all the above
mentioned factors.
6. Conclusions

Fig. 11. Monotonic plastic zone size and respective elastic enclave.

strength. Due to this decrease in monotonic plastic zone size the

magnitude of retardation may decrease, whereas the decrease in
reversed plasticity may increase extent of retardation. The variation
of monotonic plastic zone size and the respective complimentary
elastic zone (elastic enclave) with the overload temperature are indicated in Fig. 11.
Thus, due to low temperature overload, the net behavior of the
crack growth may depend on the following factors:
1. Formation of secondary cracks in the stretch zone ahead of the
crack tip.
2. Variation of yield stress and strain hardening behavior with
temperature.

The simultaneous effect of overload and low temperature on fatigue crack growth behavior has been studied and the crack growth
has been predicted by using the exponential model suitably modied to take into account the overload temperature. The model has
been developed by using the crack growth data of 25, 28, 43,
53, 83 C overload temperatures and validated by predicting
the fatigue life for the specimen overloaded at 0 C. Various factors affecting retardation have been analyzed and the results are
concluded as,
1. The application of an overload retards the crack growth rate
resulting in enhanced fatigue life.
2. Overload at low temperature enhances the retardation effect
more than overloading at normal (room) temperature.
3. The presence of a large number of secondary cracks formed in
the overload affected tear zone due to low temperature environment is responsible for enhancing the retardation effect by
reducing the available  K for primary crack extension.
4. The enhanced crack extension with lowering of temperature
of overload cycle can also be explained with the help of
Bauschinger effect.
5. The relative increase in magnitude of retardation decreases
with decrease in temperature. This may be explained on the
basis of decrease in the monotonic plastic zone size as well as

106

V.K. Sahu et al. / Aerospace Science and Technology 33 (2014) 100106

reverse plasticity on lowering of temperature during overload

application.
6. Exponential model of the form a j = ai emi j ( N j N i ) can be effectively applied to predict fatigue life under low temperature
overloading conditions (room temperature to 83 C) and to
determine retardation parameters ad and N d with reasonable
accuracy.
References
[1] V.V. Bolotin, I.L. Belousov, Early fatigue crack growth as the damage accumulation process, Probabilistic Engg. Mechanics 16 (4) (2001) 279287.
[2] V.V. Bolotin, V.L. Lebedev, Analytical model of fatigue crack growth retardation
due to overloading, Int. J. Solids Struct. 33 (9) (1996) 12291242.
[3] W.F. Brown, J.E. Srawley, Plane strain crack toughness testing of high strength
metallic materials, ASTM Spec. Tech. Publ. 401 (1966) 1.
[4] B. Budiansky, J.W. Hutchinson, Analysis of closure in fatigue crack growth,
Trans. ASME, J. Appl. Mech. 45 (2) (1978) 267276.
[5] A.U. De Koning, A simple crack closure model for prediction of fatigue crack
growth rates under variable amplitude loading, ASTM Spec. Tech. Publ. 743
(1981) 6385.
[6] W. Elber, The signicance of fatigue crack closure, in: Damage Tolerance in
Aircraft Structures, ASTM Spec. Tech. Publ. 486 (1971) 230242.
[7] N.M. Grinberg, L.F. Yakovenko, Sbornik Fizicheskie Protsessy Plasticheskoi Deformatsiipri Nizkikh Temperaturakh, Naukova Dumka, Kiev, 1974, p. 292.
[8] R. Hamam, S. Pommier, F. Bumbieler, Variable amplitude fatigue crack growth,
experimental results and modeling, Int. J. Fatigue 29 (911) (2007) 16341646.
[9] G.R. Irwin, NRL Report 6598, Nov. 21, 1967.
[10] A. Kerro, G. Montalenti, Philos. Mag. 24 (1971) 619.
[11] I.B. Kwon, M.E. Fine, J. Weertman, Fatigue damage in copper single crystals at
room and cryogenic temperatures, Acta Metall. 37 (11) (1989) 29372946.
[12] B.L. Lee, K.S. Kim, K.M. Nam, Fatigue analysis under variable amplitude loading
using an energy parameter, Int. J. Fatigue 25 (7) (2003) 621631.
[13] A.J. Mcevily, Dislocation substructures at fatigue crack tips of 304 stainless steel
cycled in air or vacuum, Scr. Mater. 35 (1996) 761765.
[14] J.R. Mohanty, B.B. Verma, P.K. Ray, Evaluation of overload-induced fatigue crack
growth retardation parameters using an exponential model, Eng. Fract. Mech.
75 (2008) 39413951.
[15] J.R. Mohanty, B.B. Verma, P.K. Ray, Prediction of fatigue crack growth and residual life using an exponential model: Part I. Constant amplitude loading, Int. J.
Fatigue 31 (2009) 418424.

[16] J.R. Mohanty, B.B. Verma, P.K. Ray, Prediction of fatigue crack growth and residual life using an exponential model: Part II. Mode-I overload induced retardation, Int. J. Fatigue 31 (2009) 425432.
[17] J.R. Mohanty, B.B. Verma, P.K. Ray, Prediction of fatigue life with interspersed
mode-I and mixed mode (I & II) overloads by an exponential model: Extensions
and Improvements, Eng. Fract. Mech. 76 (2009) 454468.
[18] J.R. Mohanty, B.B. Verma, P.K. Ray, Determination of fatigue crack growth rate
from experimental data: A new approach, Int. J. Microstruct. Mater. Prop. 5 (1)
(2010) 7987.
[19] J.C. Newman Jr., Crack-closure model for predicting fatigue crack growth under
aircraft spectrum loading, ASTM Spec. Tech. Publ. 748 (1981) 5384.
[20] P.C. Paris, F. Erdogan, A critical analysis of crack propagation laws, Trans. ASME,
J. Basic Eng. 85 (1963) 528534.
[21] J. Polak, M. Klesnil, The dynamics of cyclic plastic deformation and fatigue
life of low carbon steel at low temperatures, Mater. Sci. Eng. 26 (2) (1976)
157166.
[22] S. Pommier, Plane strain crack closure and cyclic hardening, Eng. Fract. Mech.
69 (2002) 2544.
[23] S. Pommier, P. Bompard, Bauschinger effect of alloys and plasticity-induced
crack closure: a nite element analysis, Fatigue Fract. Eng. Mater. Struct. (2000)
129139.
[24] M.A. Pompetzki, T.H. Topper, D.L. Du Quesnay, M.T. Yu, Effect of compressive underloads and tensile overloads on fatigue damage accumulation in
2024-T351 aluminum, J. Test. Eval. 18 (1) (1990) 5361.
[25] P.K. Ray, B.B. Verma, A study on spot heating induced fatigue crack growth
retardation, Fatigue Fract. Eng. Mater. Struct. 28 (2005) 579585.
[26] J. Schijive, Fatigue crack propagation in light alloy steel material and structures, in: Advances in Aeronautical Sciences, Pergamon Press, New York, 1961,
pp. 387408.
[27] S. Suresh, Fatigue of Materials, Cambridge University Press, p. 318.
[28] B.I. Verkin, I.M. Lyubarskii, N.M. Grinberg, L.F. Yakovenko, Sbornik Kosmicheskie
Issledovaniyana Ukraine, vol. 1, Naukova Dumka, Kiev, 1973, p. 14.
[29] R.P. Wei, T.T. Shih, Delay in fatigue crack growth, Int. J. Fract. 10 (1) (1974)
7785.
[30] O.E. Wheeler, Spectrum loading and crack growth, J. Basic Eng. 94 (1972)
181186.
[31] J. Willenborg, R.M. Engle, H.A. Wood, A crack growth retardation model using
an effective stress intensity concept, 1971, AFFDL, TM-71-1-FBR.
[32] J.L. Zambrow, M.G. Fontana, Mechanical Properties, including fatigue of aircraft alloys at very low temperatures, Trans. Am. Soc. Met. 41 (1949)
480518.