Accepted Manuscript

Utilizing Digital Image Correlation to Determine Stress Intensity Factors

Mohammadreza Yadegari. Dehnavi, Seyedmeysam Khaleghian, Anahita Emami,

Mehran Tehrani, Nasser Soltani

PII: S0142-9418(14)00091-9
DOI: 10.1016/j.polymertesting.2014.04.005
Reference: POTE 4226

To appear in: Polymer Testing

Received Date: 23 February 2014

Revised Date: 8 April 2014
Accepted Date: 11 April 2014

Please cite this article as: M.Y. Dehnavi, S. Khaleghian, A. Emami, M. Tehrani, N. Soltani, Utilizing
Digital Image Correlation to Determine Stress Intensity Factors, Polymer Testing (2014), doi: 10.1016/
j.polymertesting.2014.04.005.

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to
our customers we are providing this early version of the manuscript. The manuscript will undergo
copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please
note that during the production process errors may be discovered which could affect the content, and all
legal disclaimers that apply to the journal pertain.
 ACCEPTED MANUSCRIPT
Test Method

Utilizing Digital Image Correlation to Determine Stress Intensity Factors

Mohammadreza Yadegari. Dehnavia, Seyedmeysam Khaleghianb, Anahita Emamic, Mehran

Tehranid,*,and Nasser Soltania

PT
a
School of Mechanical Engineering, University of Tehran, Tehran, Iran
b
Department of Engineering Science and Mechanics, Virginia Polytechnic Institute and State University,
Blacksburg, VA 24061, USA

RI
c
Department of Mechanical Engineering, Purdue School of Engineering and Technology, Indiana University-
Purdue University, Indianapolis, IN 46202, USA
d
Mechanical Engineering Department, University of New Mexico, Albuquerque, NM 87131, USA

C
*
Corresponding author Tel: +1-505-2776298, Fax: +1-505-2771571
E-mail address: [email protected]

US
Postal address: University of New Mexico, Mechanical Engineering Department, 1 University of New Mexico,
MSC01 1150, Albuquerque, NM 87131

Abstract
AN
There is a major concern with aging and fretting wear in the transportation industry.
M

The presence of an edge crack in structures can be tolerated if its evolution is predicted using

fracture mechanics. In this study, the digital image correlation (DIC) technique was employed
D

to evaluate the in-plane stress intensity factors (SIFs), KI and KII, for the contact between a
TE

half plane with an edge crack and an asymmetric tilted wedge. DIC requires two digital
EP

images of the same sample at different stages of loading to measure the corresponding

displacements and strain fields. A computer code was developed to process the captured
C

images and determine the correlation coefficients. Fracture mechanics theory was then
AC

applied to calculate the SIFs from displacement fields obtained via DIC analysis. The effect

of the wedge tip angle on the SIFs was studied using DIC. To highlight the advantages of

DIC, a photoelasticity method was also employed to determine SIFs for the identical

problem. The SIF values from the two techniques were compared and a good agreement was

observed. In summary, it was shown that although the DIC method requires simple setup and

minimal sample preparation relative to the alternatives, it provides reliable results.

1
 ACCEPTED MANUSCRIPT
Keywords: digital image correlation, stress analysis, fracture mechanics, photoelasticity

1. Introduction

Stress concentrations at discontinuities such as cracks may lead to early structural failure

in railroads. The crack growth could be controlled if its fracture characteristics are

PT
understood. The problem becomes even more complicated when the edge cracks propagate in

mixed mode conditions [1]. Modeling and simulation of railroad contacts can be used as a

RI
useful tool to understand their behavior. Railroad contacts can be simulated as a contact

C
problem between a tilted wedge and a half-plane embodying an edge crack [2].

US
Digital image correlation (DIC) has become a very popular choice of experimental optical

method [3-5]. DIC test setup is less complex compared to other methods such as
AN
shearography, moiré and caustics [6]. In addition, its vast applications in stress and strain

analyses have made DIC a practical and valuable technique. In contrast to other optical
M

methods, such as photoelasticity, DIC is applicable to both opaque and transparent materials,
D

and its image processing does not entail fringe pattern analysis and phase-unwrapping [7].
TE

The analysis of stresses transferred within various contact components is a critical

mechanical design criterion. Therefore, a great number of investigations of contact problems

EP

in different geometries have been carried out [8, 9]. This design problem becomes escalated
C

when one of the touching components becomes vulnerable due to the presence of
AC

discontinuities and cracks. The distribution of contact forces and displacement discontinuity

of cracks in contact problems, as well as the effects of frictional forces on stress intensity

factors (SIFs) have been greatly inspected [8-11]. Nonetheless, there are few published

studies on the effect of wedge tip angle on the SIFs for the contact problem.

The DIC method was first used at the University of South Carolina to measure displacement

fields of loaded components by Peters and Ranson [12]. A few years later, Sutton et al. [13]

2
 ACCEPTED MANUSCRIPT
improved the DIC technique and demonstrated its precision in obtaining strain fields. Since

then, many researchers have been working on this field and further developed this method

[14, 15]. Recently, Réthoré et al. [16, 17] developed another form of DIC method for crack

analysis. Moreover, Sutton et al. [18] used a 3D DIC setup to demonstrate the effect of out-

of-plane displacements on 2D DIC measurements.

PT
Due to the sensitivity and precision of the DIC method, it has been widely used for

RI
different engineering problems [19]. Determining the SIFs is one of the most important

applications of this method. SIFs could be determined by using displacement fields along

C
with the finite element or finite deference method [20]. Yoneyama et al. [21] and Ju et al.

US
[22] employed this method for the evaluation of mixed mode SIFs. Furthermore, DIC

technique was employed to analyze 3D x-ray computed micro-tomographic images cracks to

AN
estimate SIFs [23].
M

In this study, effects of wedge tip angle and edge crack angle on the SIFs of a half

plane under three different contact loads were studied. The results were also compared with
D

those using photoelasticity and good agreement between the obtained SIFs was observed.
TE

2. Experimental details
EP

2.1. Sample preparation

C

Two polycarbonate plates of 240×60×4 mm were used. According to the provider, the
AC

polycarbonate had an elastic modulus of 2.4 GPa, a Poisson’s ratio of 0.38, and a friction

coefficient of 0.4. Crack lengths of 5 mm with a 0.2 mm width were machined into the

specimens employing a 0.2-mm blade. The plate edge cracks were inclined at 60° and 90°

with respect to the horizontal direction.

Four asymmetric tilted wedges made of CK45 steel (Mat.No. 1.1191, DIN Ck45,

AISI 1045) were used. All of the wedges had a total tilt angle of 165° at their apexes (θ1 + θ2)

3
 ACCEPTED MANUSCRIPT
with different side angles (θ1 and θ2 as shown in Fig. 1). Their side angle values are presented

in Table 1.

Fig. 1

Table 1 The geometry of the wedges used in the experiments

PT
Wedge label θ1 θ2

RI
W1 87 78

W2 86 79

C
US
W3 85 80

W4 84 81
AN
The difference in the wedge side angles changes the distribution of the components of

contact force in the horizontal and vertical directions. The horizontal force applied to the
M

wedge amplifies as the difference between wedge side angles increases. Wedges in Table 1
D

are sorted from W1 to W4, according to the magnitude of the horizontal force that they exert
TE

on the crack for a certain applied force. The distance between the contact point and the crack

opening was 4 mm in all cases and the applied load to the wedge for each specimen was
EP

varied as 188, 255, and 305N. A schematic representation of the experimental model is

shown in Fig. 2.
C
AC

Fig. 2

2.2. DIC Method

The experiment setup is shown in Fig. 3. The load cell capacity was 200kg.F. An art-

cam 320p CCD camera equipped with a Fujian lens was used to acquire the images and a

standard white light source was utilized to illuminate the sample surfaces.. The captured

4
 ACCEPTED MANUSCRIPT
images had a spatial resolution of 2288×1700 pixels. This setup was mounted on a vibration

isolation table.

Fig. 3

PT
A random speckle pattern is needed for the correlation process. Specimen surfaces

RI
were sprayed with white and, subsequently, black paints to create this speckle pattern.

Specimens were secured in the load cell and a series of images were acquired before and after

C
the loading. A pair of images for each experiment was considered and analyzed in

US
comparison in order to study their correlation and obtain the displacement fields. The

correlation equation can be defined as [24]:

AN
i=n j=n

∑ ∑ (I ( x p + i, y p + j ) − I D ( x p + i + u p , y p + j + v p ) )
2
M

R
i = − n j =− n
C ( x, y , u , v ) = i=n j=n (1)
∑ ∑ (I ( x p + i, y p + j ) )
2
R
i =− n j =− n
D
TE

where n=0.5x(m-1), m is the square subset size of the image in pixels, which is always an odd

number. In this study, a subset size of 21×21 pixels was considered. In Eq. (1), xp and yp are
EP

the coordinates of a pixel location in the reference image, IR and ID represent the light

intensity before and after the loading, respectively, and C is the correlation coefficient.
C

Considering the elastic stress field solution, a singularity exists at the tip of the crack.
AC

In practice, this region undergoes plastic deformation and the stress values are finite, and thus

linear elastic fracture mechanics (LEFM) is not valid in this area. This region is known as the

crack tip plastic zone denoted by rp [25]. As a result, the region of interest to extract the

experimental displacements data (u and v) must be selected from outside of this plastic zone.

Irwin [26] has suggested a correction solution for measurement of the length of the crack tip

plastic zone following the plane stress condition:

5
 ACCEPTED MANUSCRIPT
2
1K 
rp =  I  (2)
π  σ y 

and for the plane strain condition:

2
1  KI 
rp =   (3)
3π σy 

PT
In the current study, the area of interest in the acquired images was chosen with a 21

RI
pixels distance from the crack edges on each side of the crack, and the DIC analysis was

carried out on 21-pixel subsets with a 3-pixel subset spacing. Coordinates of pixels were

C
extracted using an image processing code developed in MATLAB. Subsequently, SIFs were

US
calculated by least-squares fitting to the following equations [25-27]:
AN
n
∞
r2 n n n n
U =∑ {a n [(κ + + ( − 1) n ) cos θ − cos( − 2)θ ]
n =0 2 µ 2 2 2 2
n n n n
− b n [(κ + − ( − 1) n ) sin θ − sin( − 2)θ ]} + α r cos θ + U 0
M

2 2 2 2
n (4)
∞
r2 n n n n
V =∑ {a n [(κ − − ( − 1) n ) sin θ + sin( − 2)θ ]
D

n =0 2 µ 2 2 2 2
TE

n n n n
+ b n [(κ − + ( − 1) n ) cos θ + cos( − 2)θ ]} + α r sin θ +V 0
2 2 2 2
EP

where U and V are the displacement components in x and y directions, respectively. µ is the

shear modulus equal to E/2(1+ν), E is elastic modulus, ν is the Poisson’s ratio, κ is (3-
C

ν)/(1+ν) for plane stress and 3-4ν for plane strain conditions. In Eq. (4), r and θ represent the
AC

polar coordinates around the crack tip with the origin of coordinate system located on the

crack tip and r-axis pointing in the direction of crack length. The constants U0, V0, and α are

associated with rigid body displacements and rotations. In Eq. (4), a1 is equal to ⁄√2 and

b1 is equal to - ⁄√2 .

6
 ACCEPTED MANUSCRIPT
Besides U0, V0, and αr terms, at least four more terms of the William’s equation, Eq. (4), are

required in order to obtain an accurate solution for a finite size plate [27]. The necessity of

using the higher order terms in Eq. (4) is primarily to omit the effect of boundary conditions

on the results [28]. In this study, n was varied from 0 to 4 in Eq. (4) in order to calculate the

SIFs; this was the least number of terms to produce a reliable result.

PT
Another factor that significantly influences the accuracy of this method is the ability to

RI
precisely detect the crack tip location. Therefore, the analysis for detecting the crack tip was

run repeatedly for different crack tip locations around an initial rough guess coordinate, and

C
the summation of errors between all data points was calculated. Following [20], the location

US
that gave the minimum error was assumed to be the crack tip location. For this purpose, a

MATLAB code was developed to automatically perform the iterations and find the crack tip
AN
location and SIFs.
M

2.3. Photoelasticity Method

Photoelastic data were obtained concurrently with the DIC image acquisition.
D

Specimens were placed in a polariscope consisting of a monochromatic red light source, a

TE

polarizer, quarter-wave plates and an analyzer. A 7-mega pixel color camera was also utilized

to capture images of the photoelastic fringes that were formed in the plates under loading.
EP

Since the widths of isochromatic fringes in the photoelastic images were non-zero and the
C

exact positions of dark fringe paths were not discernible, all the acquired images were post-
AC

processed and sharpened with the aid of an image processing computer code. The developed

code enabled the detection of the exact locations of pixels on the isochromatic fringe where

the light intensity was a minimum. A sample of an image before and after the processing is

shown in Fig. 4.a and Fig. 4.b, respectively.

Fig. 4.a and b

7
 ACCEPTED MANUSCRIPT

After the image processing was performed, each image was loaded to another

computer code to identify the actual size of the model in the image and determine the actual

polar coordinates of selected pixels in the image. The coordinate system was set so that its

origin was located on the crack tip and r-axis in the direction of crack length, and positive θ

PT
values were measured counter-clockwise. It should be noted that the last black pixel of the

RI
crack line was selected as the crack tip. Subsequently, 20 arbitrary pixels from the

isochromatic fringes were selected in each image and their coordinates were extracted. In the

C
authors’ previous work, the use of twenty points was shown to minimize the errors in

US
calculation of the SIFs from the photoelastic data [2].

LEFM gives the following equations for calculating the stress field around a crack tip
AN
in the polar coordinate system [25]:
M

∞
n n n n n
σ x = ∑ r ( n / 2)−1{an [(2 + + (−1) n ) cos( − 1)θ − ( − 1) cos( − 3)θ ]
n =1 2 2 2 2 2
D

n n n n
− b n [(2 + − (−1) n ) sin( − 1)θ − ( − 1) sin( − 3)θ ]}
2 2 2 2
TE

∞
n n n n n
σ y = ∑ r ( n /2) −1{an [(2 − − (−1) n ) cos( − 1)θ + ( − 1) cos( − 3)θ ]
n =1 2 2 2 2 2
(5)
n n n n
− b n [(2 − + (−1) n ) sin( − 1)θ + ( − 1) sin( − 3)θ ]}
EP

2 2 2 2
∞
n n n n n
τ xy = ∑ r ( n /2)−1{an [( − 1)sin( − 3)θ − ( + (−1) n ) sin( − 1)θ ]
n =1 2 2 2 2 2
C

n n n n
− b n [( − 1) cos( − 3)θ − ( − (−1)n ) cos( − 1)θ ]}
AC

2 2 2 2

where r and θ are coordinates in the vicinity of the crack tip, a1 equals to ⁄√2 and b1

equals to - ⁄√2 . The value of n for this study was varied from 1 to 5 [29]. Moreover, the

stress-optic law for a plane stress case is defined as [30]:

Nf σ
τ max = (6)
2h

8
 ACCEPTED MANUSCRIPT
where τmax is the maximum shear stress, N is the relative retardation in terms of a complete

cycle of retardation (isochromatic fringe order) and fσ is the material fringe value, which was

acquired from a calibration test [2]. The maximum in-plane shear stress can be defined as:

σ r −σθ
(τ m ax ) 2 = ( ) 2 + (τ r θ ) 2 (7)
2

PT
Combining Eq. (2.6) and Eq. (2.7) leads to:

RI
Nf σ 2
( ) = (σ r − σ θ ) 2 + 4(τ r θ ) 2 (8)
h

C
The over-determined least square estimation method was employed to calculate the

US
SIFs using the combination of Eq. (1) and Eq. (5). AN
4. Results and discussion
M

Figs. 5 and 6 represent the contour maps for the contact of the half plane with wedges
D

labeled W1 and W4, respectively. A vertical force of 255 N was applied by the wedges to the
TE

specimens.
EP

Fig. 5.a and b

C
AC

Fig. 6.a and b

9
 ACCEPTED MANUSCRIPT
The calculated SIFs along with the relative difference between the SIFs obtained from

DIC and photoelasticity methods are presented in Tables 2 and 3. The half plane with a 90°

edge crack experienced a mixed mode fracture and retained both non-zero in-plane SIFs, KI

and KII. However, the half plane embodying the 60° edge crack, only experienced a second

mode fracture for all the studied cases; the calculated KI values were negative, and hence

PT
were considered to be zero. This is due to the net effect of the horizontal and vertical

RI
components of the applied forces and their corresponding displacements, where they caused a

closing mode of the crack.

C
Table 2. Half plane with a 60° edge crack

KII (KPa.m0.5)
US
AN
Photoelasticity Relative
Wedge Force (N) DIC Method
Method Difference (%)
M

188 150.7 148.3 1.63

W1 255 151.3 148.6 1.85
305 166.6 163.0 2.23
188 117.7 115.7 1.66
D

W2 255 135.2 132.4 2.12

305 154.2 150.6 2.42
TE

188 111.5 109.6 1.75

W3 255 113.8 111.2 2.32
305 151.9 148.4 2.38
EP

188 103.1 101.3 1.79

W4 255 111.3 108.7 2.31
305 125.2 122.1 2.56
C
AC

Table 3. Half plane with a 90° edge crack

KI (KPa.m0.5) KII (KPa.m0.5)

DIC Photoelasticity Relative DIC Photoelasticity Relative
Force
Wedge Method Method Difference Method Method Difference
(N)
(%) (%)
188 482.6 475.7 1.47 249.5 246.3 1.32
W1 255 602.7 592.7 1.69 316.0 311.0 1.58
305 838.1 819.9 2.22 441.7 434.2 1.73
W2 188 442.1 436.3 1.34 244.4 241.8 1.07

10
 ACCEPTED MANUSCRIPT
255 529.3 520.8 1.63 309.7 305.3 1.44
305 647.8 634.9 2.03 440.0 432.1 1.82
188 354.5 350.0 1.29 223.3 220.8 1.12
W3 255 455.1 448.7 1.42 288.1 284.7 1.18
305 623.6 610.3 2.19 425.4 417.4 1.92
188 305.4 301.8 1.18 184.5 182.7 0.96
W4 255 420.2 414.0 1.5 266.7 263.6 1.2
305 549.5 538.0 2.14 375.8 368.9 1.87

PT
The relative differences between the results of DIC and photoelasticity methods are

RI
less than 3% for all cases. This relative difference was calculated as the difference between

C
the results of the two methods for any given configuration divided by the corresponding value

US
obtained from the photoelasticity method. The photoelasticity method was selected as the

base for comparison as it exhibited good agreement with the finite element results in our
AN
previous study [2]. As shown in Fig. 7, this relative difference was less than 2.2 % for all the

cases and was raised with increase of the vertical force.

M

Fig. 7
D
TE

The relationships between the vertical force and the relative difference between KII

values obtained from the DIC and photoelasticity methods are shown in Figs. 8.a and 8.b.
EP

The maximum relative difference in the results is less than 3%, which makes both methods

effective for evaluation of the stress intensity factors. This error increases with increase of the
C

applied vertical load. This can be attributed to the errors associated with the image data
AC

processing of the DIC method at larger displacements, as well as the augmented

nonlinearities due to the larger loads and displacements. Where the magnitude of the vertical

force is small, due to the lack of detectable isochromatic fringes near the crack tip, the

photoelasticity becomes ineffective and DIC is the method of preference. Moreover, there is

no clear correlation between the magnitude of the horizontal force and the relative difference

between the DIC and photoelasticity results in all the studied cases.

11
 ACCEPTED MANUSCRIPT
Fig. 8.a

Fig. 8.b

As indicated in Fig. 8, the relative difference between results of the DIC and

PT
photoelasticity methods for the half plane with 90° edge crack were slightly lower than that

of the half plane with 60° edge crack. Thus, the DIC SIFs obtained for the 90° edge crack are

RI
more reliable compared to the 60° case. We speculate that these deviations are caused by the

C
image processing algorithms used in the DIC analysis.

The relationship between the second mode SIF (KII) of the half plane with 60° edge

US
crack and the vertical force applied to the four wedges is shown in Fig. 9.
AN
As anticipated, the increase of the vertical force resulted in a higher value of KII due to

the increase in the in-plane displacements. Moreover, the magnitude of the horizontal force
M

affected the KII values in the same manner. The increase of the horizontal force induces

sliding mode (mode II) displacements on the crack. As mentioned earlier, the horizontal force
D

applied by the wedges has a direct relation with the difference between tip side angles and,
TE

therefore, the horizontal force of the wedges increases as W1, W2, W3, and W4.
EP

The effect of the vertical force on KI and KII in the half plane with 90° edge crack for

the four wedges is shown in Figs. 10 and 11, respectively.

C
AC

Fig. 9

Fig.10

As displayed in Fig. 10, the increase of vertical force affects the opening mode of the

crack and elevates the value of KI. In contrast to the 60° edge crack, the 90° one tended to

12
 ACCEPTED MANUSCRIPT
rupture under the horizontal force. As a consequence, KI reached its highest values when the

W1 wedge (the largest horizontal force) was used. Similarly, KI decreased when the horizontal

force was reduced using W2, W3, and W4 wedges.

Fig. 11

PT
RI
As presented in Fig. 11, the increase of the vertical force leads to a larger in-plane

displacement and, consequently, higher values of KII are achieved. Likewise, the increase of

C
the horizontal force instigates higher in-plane shear SIF (KII). Therefore, while the half plane

US
was in contact with the W1 wedge, KII reached its maximum value. KII exhibited its minimum

value when the W4 wedge was used. The KII values for the 90° crack were much larger than
AN
that of the 60° crack. This is attributed to the opposing shearing displacements induced by the
M

vertical and horizontal components of the applied force on the 90° crack.

Finally, the DIC method gave higher SIF values for all the cases, and thus DIC provided an
D

overestimated design criterion for both modes I and II fracture.

TE

5. Conclusions
EP

DIC method was utilized to evaluate SIFs in the contact problem between a half plane

with an edge crack and an asymmetric tilted wedge. The effect of wedge tip angles on SIF
C

values was studied for two different crack angles (60° and 90°) under three different force
AC

magnitudes. The results of DIC method were compared with the ones obtained from the

photoelasticity technique and revealed that the relative difference between the SIF values

obtained from the two methods was less than 3% for all the studied cases. This relative

difference increased as a larger vertical force was applied on the specimens. DIC method

could be a reliable alternative for the classical photoelasticity technique where the magnitude

of the applied force and, consequently, the number of isochromatic fringes decrease to such a

13
 ACCEPTED MANUSCRIPT
low number that necessary data for accurate determination of SIFs from photoelastic images

is not provided. Moreover, the DIC offers SIF evaluation for non-transparent samples and

employs a simpler setup compared to the photoelasticity method.

A comparison between the two half planes indicates that the half plane with 60° edge

crack was only under second mode fracture. On the other hand, the half plane with 90° edge

PT
crack experienced mixed mode fracture where its first mode SIF, KI, had a greater value than

RI
the second one, KII. The increase of the vertical force had the same effect on SIFs for both

half planes and raised the SIF values, but never led the 60° edge crack to undergo first mode

C
fracture; KI value remained zero. The increase of horizontal force similar to the increase of

US
vertical force had the same effect on SIFs. Thus, it can be concluded that, as the asymmetry

of in-contact wedge increases, the half plane will break under a smaller force.
AN
References
M

[1] M. Guagliano, M. Sangirardi, L. Vergani, Experimental analysis of surface cracks in rails under
rolling contact loading, Wear, 265 (2008) 1380-1386.
D

[2] S. Khaleghian, A. Emami, M. Tehrani, N. Soltani, Analysis of effective parameters for stress
intensity factors in the contact problem between an asymmetric wedge and a half-plane using an
TE

experimental method of photoelasticity, Materials & Design, 43 (2013) 447-453.

[3] M.R.Y. Dehnavi, I. Eshraghi, N. Soltani, Investigation of fracture parameters of edge V-notches in
a polymer material using digital image correlation, Polym Test, 32 (2013) 778-784.
[4] S.R. Heinz, J.S. Wiggins, Uniaxial compression analysis of glassy polymer networks using digital
EP

image correlation, Polym Test, 29 (2010) 925-932.

[5] X. Wang, S.P. Ma, Y.T. Zhao, Z.B. Zhou, P.W. Chen, Observation of damage evolution in polymer
bonded explosives using acoustic emission and digital image correlation, Polym Test, 30 (2011) 861-
C

866.
[6] K.H. Laermann, New Achievements and Perspectives of Optical Methods in Experimental Solid
AC

Mechanics, Opt Laser Eng, 22 (1995) 249-270.

[7] S. Yoneyama, Y. Morimoto, M. Takashi, Automatic evaluation of mixed-mode stress intensity
factors utilizing digital image correlation, Strain, 42 (2006) 21-29.
[8] A.N. Guz, V.V. Zozulya, Elastodynamic unilateral contact problems with friction for bodies with
cracks, Int Appl Mech+, 38 (2002) 895-932.
[9] A.N. Guz, V.V. Zozulya, Fracture dynamics with allowance for crack edge contact interaction, Int J
Nonlinear Sci, 2 (2001) 173-233.
[10] N.V. Zozulya, Contact problem for a plane crack under a normally incident antiplane shear wave,
Int Appl Mech+, 43 (2007) 586-590.
[11] V.V. Zozulya, Stress intensity factor in a contact problem for a plane crack under an antiplane
shear wave, Int Appl Mech+, 43 (2007) 1043-1047.

14
 ACCEPTED MANUSCRIPT
[12] W.H. Peters, W.F. Ranson, Digital Imaging Techniques in Experimental Stress-Analysis, Opt Eng,
21 (1982) 427-431.
[13] M.A. Sutton, W.J. Wolters, W.H. Peters, W.F. Ranson, S.R. McNeil, Determination of
Displacements Using an Improved Digital Correlation Method, Image and Vision Computing, 1 (1083)
133-139.
[14] H. Lu, P.D. Cary, Deformation measurements by digital image correlation: Implementation of a
second-order displacement gradient, Experimental Mechanics, 40 (2000) 393-400.
[15] Y. Wang, A.M. Cuitino, Full-field measurements of heterogeneous deformation patterns on
polymeric foams using digital image correlation, International Journal of Solids and Structures, 39

PT
(2002) 3777-3796.
[16] J. Rethore, F. Hild, S. Roux, Shear-band capturing using a multiscale extended digital image
correlation technique, Comput Method Appl M, 196 (2007) 5016-5030.
[17] J. Rethore, F. Hild, S. Roux, Extended digital image correlation with crack shape optimization, Int

RI
J Numer Meth Eng, 73 (2008) 248-272.
[18] M.A. Sutton, J.H. Yan, V. Tiwari, H.W. Schreier, J.J. Orteu, The effect of out-of-plane motion on
2D and 3D digital image correlation measurements, Opt Laser Eng, 46 (2008) 746-757.

C
[19] J.D. Krehbiel, J. Lambros, J.A. Viator, N.R. Sottos, Digital Image Correlation for Improved
Detection of Basal Cell Carcinoma, Experimental Mechanics, 50 (2010) 813-824.

US
[20] S.R. Mcneill, W.H. Peters, M.A. Sutton, Estimation of Stress Intensity Factor by Digital Image
Correlation, Engineering Fracture Mechanics, 28 (1987) 101-112.
[21] S. Yoneyama, T. Ogawa, Y. Kobayashi, Evaluating mixed-mode stress intensity factors from full-
AN
field displacement fields obtained by optical methods, Engineering Fracture Mechanics, 74 (2007)
1399-1412.
[22] S.H. Ju, S.H. Liu, K.W. Liu, Measurement of stress intensity factors by digital camera,
International Journal of Solids and Structures, 43 (2006) 1009-1022.
M

[23] S. Roux, J. Rethore, F. Hild, Digital image correlation and fracture: an advanced technique for
estimating stress intensity factors of 2D and 3D cracks, J Phys D Appl Phys, 42 (2009).
[24] G. Vendroux, W.G. Knauss, Submicron deformation field measurements: Part 2. Improved digital
D

image correlation, Experimental Mechanics, 38 (1998) 86-92.

[25] N.E. Dowling, H.A. Ernst, The Mechanics of Fracture and Fatigue - an Introduction - Spon,E,
TE

Spon,Fn, J Test Eval, 11 (1983) 228-228.

[26] G. Irwin, Plastic zone near a crack tip and fracture toughness, in: Proceedings of the Seventh
Sagamore Ordnance Material Conference, 1960, pp. 63–78.
[27] V. Vesely, L. Sestakova, S. Seitl, Influence of Boundary Conditions on Higher Order Terms of
EP

near-Crack-Tip Stress Field in a Wst Specimen, Key Eng Mater, 488-489 (2012) 399-402.
[28] F. Berto, P. Lazzarin, On Higher Order Terms in the Crack Tip Stress Field, Int J Fracture, 161
(2010) 221-226.
C

[29] M. Mahinfalah, L. Zackery, Photoelastic Determination of Mixed-Mode Stress Intensity Factors

for Sharp Reentrant Corners, Engineering Fracture Mechanics, 52 (1995) 639-645.
AC

[30] E.J.E.J. Hearn, Photoelasticity, Merrow Publishing Co. Ltd 1971.

Figure captions:

Fig. 1 Schematic of asymmetric wedge geometry

Fig. 2 Illustration of the contact experimental model
Fig. 3 Set-up of DIC experiment

15
 ACCEPTED MANUSCRIPT
Fig. 4.a) Before image processing and b) after image processing; This image is for a half
plane with 60° edge crack and wedge label W3 under a 305N force.
Fig. 5.a Contour map of the horizontal displacement field around the crack tip for the contact
between the half plane and wedge label W1
Fig. 5.b Contour map of the vertical displacement field around the crack tip for the contact
between the half plane and wedge label W1
Fig. 6.a Contour map of the horizontal displacement field around the crack tip for the contact
between the half plane and wedge label W4
Fig. 6.b Contour map of the vertical displacement field around the crack tip for the contact

PT
between the half plane and wedge label W4
Fig. 7 The relationship between the vertical force magnitude and relative difference between
KI values obtained from the DIC and photoelasticity methods for the half plane with 90° edge

RI
crack. W1 to W4 represent the wedge labels as explained in table 1.
Fig. 8.a The relationship between the vertical force magnitude and relative difference
between KII values obtained from the DIC and photoelasticity methods for the half plane with

C
90° edge crack. W1 to W4 represent the wedge labels as explained in table 1.
Fig. 8.b The relationship between the vertical force magnitude and relative difference

US
between KII values obtained from the DIC and photoelasticity methods for the half plane with
60° edge crack. W1 to W4 represent the wedge labels as explained in table 1.
Fig. 9 The relationship between the second mode SIF (KII ) of the 60° edge crack and the
vertical force applied on the four wedges with different tip side angles.
AN
Fig.10 The effect of vertical force on SIF (KI ) of half plane with 90° edge crack for four
wedges used in the experiments
Fig. 11 The effect of vertical force on SIF (KII )of half plane with 90° edge crack for four
M

wedges used in the experiments

D
TE
C EP
AC

16
 ACCEPTED MANUSCRIPT

PT
C RI
US
AN
M
D
TE
EP
C
AC
 ACCEPTED MANUSCRIPT

PT
C RI
US
AN
M
D
TE
EP
C
AC
 ACCEPTED MANUSCRIPT

PT
C RI
US
AN
M
D
TE
EP
C
AC
 ACCEPTED MANUSCRIPT

PT
C RI
US
AN
M
D
TE
EP
C
AC
 ACCEPTED MANUSCRIPT

PT
C RI
US
AN
M
D
TE
EP
C
AC
 ACCEPTED MANUSCRIPT

PT
C RI
US
AN
M
D
TE
EP
C
AC
 ACCEPTED MANUSCRIPT

PT
C RI
US
AN
M
D
TE
EP
C
AC
 ACCEPTED MANUSCRIPT

PT
C RI
US
AN
M
D
TE
EP
C
AC
 ACCEPTED MANUSCRIPT

PT
C RI
US
AN
M
D
TE
EP
C
AC
 ACCEPTED MANUSCRIPT

PT
C RI
US
AN
M
D
TE
EP
C
AC
 ACCEPTED MANUSCRIPT

PT
C RI
US
AN
M
D
TE
EP
C
AC
 ACCEPTED MANUSCRIPT

PT
C RI
US
AN
M
D
TE
EP
C
AC
 ACCEPTED MANUSCRIPT

PT
C RI
US
AN
M
D
TE
EP
C
AC
 ACCEPTED MANUSCRIPT

PT
C RI
US
AN
M
D
TE
EP
C
AC
 ACCEPTED MANUSCRIPT

PT
C RI
US
AN
M
D
TE
EP
C
AC