Virtual crack closure technique: History, approach

,
and applications
Ronald Krueger
National Institute of Aerospace, Hampton, Virginia 23666
rkrueger@nianet.org

An overview of the virtual crack closure technique is presented. The approach used is dis-
cussed, the history summarized, and insight into its applications provided. Equations for two-
dimensional quadrilateral finite elements with linear and quadratic shape functions are given.
Formulas for applying the technique in conjunction with three-dimensional solid elements as
well as plate/shell elements are also provided. Necessary modifications for the use of the
method with geometrically nonlinear finite element analysis and corrections required for ele-
ments at the crack tip with different lengths and widths are discussed. The problems associ-
ated with cracks or delaminations propagating between different materials are mentioned
briefly, as well as a strategy to minimize these problems. Due to an increased interest in using
a fracture mechanics–based approach to assess the damage tolerance of composite structures
in the design phase and during certification, the engineering problems selected as examples
and given as references focus on the application of the technique to components made of
composite materials. 关DOI: 10.1115/1.1595677兴
Keywords: Finite Element Analysis, Fracture Mechanics, Crack Closure Integral,
Composite Structures, Delamination, Interlaminar Fracture

1 INTRODUCTION a mathematical relationship between G c and G II /G T 关6兴.
One of the most common failure modes for composite struc- Failure is expected when, for a given mixed-mode ratio
tures is delamination 关1– 4兴. The remote loadings applied to G II /G T , the calculated total energy release rate, G T , ex-
composite components are typically resolved into interlami- ceeds the interlaminar fracture toughness, G c . Although sev-
nar tension and shear stresses at discontinuities that create eral specimens have also been suggested for the measure-
ment of the mode III interlaminar fracture toughness
mixed-mode I, II, and III delaminations. To characterize the
property 关15–18兴, an interaction criterion incorporating the
onset and growth of these delaminations the use of fracture scissoring shear has not yet been established. The virtual
mechanics has become common practice over the past two crack closure technique 共VCCT兲 关19–23兴 is widely used for
decades 关5–7兴. The total strain energy release rate, G T , the computing energy release rates based on results from con-
mode I component due to interlaminar tension, G I , the mode tinuum 共2D兲 and solid 共3D兲 finite element 共FE兲 analyses to
II component due to interlaminar sliding shear, G II , and the supply the mode separation required when using the mixed-
mode III component, G III , due to interlaminar scissoring mode fracture criterion.
shear, as shown in Fig. 1, need to be calculated. In order to Although the original publication on VCCT dates back a
predict delamination onset or growth for two-dimensional quarter century 关19兴, the virtual crack closure technique has
problems, these calculated G components are compared to not yet been implemented into any of the large commercial
interlaminar fracture toughness properties measured over a general purpose finite element codes such as MSC NASTRAN,
range from pure mode I loading to pure mode II loading ABAQUS, ANSYS, ASKA, PERMAS or SAMCEF. Currently
关8 –13兴. A quasistatic mixed-mode fracture criterion is deter- FRANC2D, developed by the Cornell Fracture Group 共CFG兲 at
mined by plotting the interlaminar fracture toughness, G c , Cornell University, appears to be the only publically avail-
versus the mixed-mode ratio, G I /G T , determined from data able, highly specialized finite element code that uses the vir-
generated using pure mode I Double Cantilever Beam 共DCB兲 tual crack closure technique 关24,25兴. The virtual crack clo-
(G II /G T ⫽0), pure mode II End Notched Flexure 共4ENF兲 sure technique has been used mainly by scientists in
(G II /G T ⫽1), and mixed-mode Mixed Mode Bending universities, research institutions, and government laborato-
共MMB兲 tests of varying ratios, as shown in Fig. 2 for IM7/ ries and is usually implemented in their own specialized
8552 关14兴. A curve fit of these data is performed to determine codes or used in postprocessing routines in conjunction with

Transmitted by Associate Editor JN Reddy

Appl Mech Rev vol 57, no 2, March 2004 109 © 2004 American Society of Mechanical Engineers

110 Krueger : Virtual crack closure technique: History, approach, and applications Appl Mech Rev vol 57, no 2, March 2004

general purpose finite element codes. Lately, an increased which the crack was extended virtually, yielding information
interest in using a fracture mechanics–based approach to as- on the most likely growth direction. Modifications of the
sess the damage tolerance of composite structures in the de- method have been suggested in the literature to allow the
sign phase and during certification has also renewed the in- mode separation for two-dimensional analysis 关38,39兴. An
terest in the virtual crack closure technique 关4,23兴. Efforts equivalent domain integral method that can be applied to
are underway to incorporate these approaches in the Com- both linear and nonlinear problems and additionally allows
posites Material MIL-17 Handbook.1 for mode separation was proposed in Refs. 关40– 45兴. The
The goal of the current paper is to give an overview of the methods above have been mentioned here briefly to comple-
virtual crack closure technique, discuss the approach used, ment the background information. A comprehensive over-
summarize the history, and provide insight into its applica- view of different methods used to compute energy release
tion. Equations for two-dimensional quadrilateral elements rates is given in Ref. 关46兴. Alternative approaches to compute
with linear and quadratic shape functions will be provided. the strain energy release rate based on results obtained from
Formulas for applying the technique in conjunction with finite element analysis have also been published recently
three-dimensional solid elements as well as plate/shell ele- 关47– 49兴.
ments will also be given. Necessary modifications for the use For delaminations in laminated composite materials
of the method with geometrically nonlinear finite element where the failure criterion is highly dependent on the mixed-
analysis and corrections required for elements at the crack tip mode ratio and propagation occurs in the laminate plane, the
with different lengths and widths will be discussed. The virtual crack closure technique 关19–22兴 has been most
problems associated with cracks or delaminations propagat- widely used for computing energy release rates because frac-
ing between different materials 共the so-called bimaterial in- ture mode separation is determined explicitly. Recently new
terface兲 will be mentioned briefly, as well as a strategy to VCCT methods to compute mixed-mode energy release rates
minimize these problems. The selected engineering problems suitable for the application with the p version of the finite
shown as examples and given as references will focus on the element method have also been developed 关50兴. Some modi-
application of the technique related to composite materials as fied and newly developed formulations of the VCCT allow
mentioned above. applications that are not based on finite element analysis and
are suitable for boundary element analysis 关25,51兴.
2 BACKGROUND
2.1 Crack closure method using two analysis steps
A variety of methods are used to compute the strain energy
Even though the virtual crack closure technique is the focus
release rate based on results obtained from finite element
of this paper and is generally mentioned in the literature, it
analysis. The finite crack extension method 关26,27兴 requires
appears appropriate to include a related method: the crack
two complete analyses. In the model the crack gets extended
closure method or two-step crack closure technique. The ter-
for a finite length prior to the second analysis. The method
minology in the literature is often inexact and this two-step
provides one global total energy release rate as global forces
method is sometimes referred to as VCCT. It may be more
on a structural level are multiplied with global deformations
appropriate to call the method the crack closure method be-
to calculate the energy available to advance the crack. The
cause the crack is physically extended, or closed, during two
virtual crack extension method 关28 –37兴 requires only one
complete finite element analyses as shown in Fig. 3. The
complete analysis of the structure to obtain the deformations.
crack closure method is based on Irwin’s crack closure inte-
The total energy release rate or J integral is computed locally
gral 关52,53兴. The method is based on the assumption that the
at the crack front and the calculation only involves an addi-
energy ⌬E released when the crack is extended by ⌬a from
tional computation of the stiffness matrix of the elements
a 关Fig. 3共a兲兴 to a⫹⌬a 关Fig. 3共b兲兴 is identical to the energy
affected by the virtual crack extension. The method yields
the total energy release rate as a function of the direction in

1
http://www.mil17.org/

Fig. 1 Fracture modes Fig. 2 Mixed-mode delamination criterion for IM7/8552

Appl Mech Rev vol 57, no 2, March 2004 Krueger : Virtual crack closure technique: History, approach, and applications 111

required to close the crack between location ᐉ and i 关Fig. surface. Forces at constraints may also be used if this option
3共a兲兴. Index 1 denotes the first step depicted in Fig. 3共a兲 and is available in the finite element software used. The options
index 2 the second step as shown in Fig. 3共b兲. For a crack are discussed in detail in the Appendix. The displacements
modeled with two-dimensional four-noded elements as
shown in Fig. 3 the work ⌬E required to close the crack ⌬u 2ᐉ and ⌬w 2ᐉ are obtained from a second finite element
along one element side can be calculated as analysis where the crack has been extended to its full length
a⫹⌬a as shown in Fig. 3共b兲.

⌬E⫽ 21 关 X 1ᐉ ⌬u 2ᐉ ⫹Z 1ᐉ ⌬w 2ᐉ 兴 , (1)
2.2 The modified crack closure method
The modified, or virtual, crack closure method 共VCCT兲 is
where X 1ᐉ and Z 1ᐉ are the shear and opening forces at nodal based on the same assumptions as the crack closure method
point ᐉ to be closed 关Fig. 3共a兲兴 and ⌬u 2ᐉ and ⌬w 2ᐉ are the described above. Additionally, however, it is assumed that a
differences in shear and opening nodal displacements at node crack extension of ⌬a from a⫹⌬a 共node i兲 to a⫹2⌬a
ᐉ as shown in Fig. 3共b兲. The crack closure method estab- 共node k兲 does not significantly alter the state at the crack tip
lishes the original condition before the crack was extended. 共Fig. 4兲. Therefore, when the crack tip is located at node k,
Therefore the forces required to close the crack are identical the displacements behind the crack tip at node i are approxi-
to the forces acting on the upper and lower surfaces of the mately equal to the displacements behind the crack tip at
closed crack. The forces X 1ᐉ and Z 1ᐉ may be obtained from node ᐉ when the crack tip is located at node i. Further, the
a first finite element analysis where the crack is closed as energy ⌬E released when the crack is extended by ⌬a from
shown in Fig. 3共a兲 by summing the forces at common nodes a⫹⌬a to a⫹2⌬a is identical to the energy required to close
from elements belonging either to the upper or the lower the crack between location i and k. For a crack modeled with

Fig. 3 Crack closure method 共two-step method兲.
a兲 First step—crack closed and b兲 second step—
crack extended.

The mode I and mode II components of the strain energy mode strain energy release rates using two-dimensional finite release rate. 4. modeled using single nodes. and applications Appl Mech Rev vol 57. ⌬E⫽ 21 关 X i ⌬u ᐉ ⫹Z i ⌬w ᐉ 兴 . 5 Crack modeled as one-dimensional discontinuity. ment as shown in Fig. 4. . four-noded elements. 6共b兲. the crack of length a is represented as a one- dimensional discontinuity by a line of nodes as shown in Fig. and are not connected with each other as shown in Fig. or the section where the crack is closed and the structure is still intact. 1 G II⫽⫺ X 共 u ⫺u ᐉ * 兲 . In the following. crack may be obtained from one single finite element analy. 6共a兲.关54兴. or plane strain model. it is important to ad- vance the crack in a kinematically compatible way. results in kinematically incompatible interpenetration for the sis. as shown in Fig. 5共b兲. a兲 Ini- tially modeled. The crack tip and the undamaged section. are calculated for four-noded ele- element models such as plane stress or plane strain.112 Krueger : Virtual crack closure technique: History. 5. is Fig. G I and G II . where ⌬A is the crack surface created. 7共a兲: approaches are also discussed for the cases where the crack 1 or delamination is modeled with plate/shell elements or with G I⫽⫺ Z 共 w ⫺w ᐉ * 兲 . The details of calculating the energy release rate G eight-noded elements with quadratic shape functions as ⫽⌬E/⌬A. March 2004 two-dimensional. no 2. Thus. 4 Modified crack closure method 共one-step VCCT兲 finite element mesh. forces and dis. This is discussed in detail in the Appendix. which caused initial problems when separation into the individual mode components will be dis.关55兴 and later generalized CRACK CLOSURE TECHNIQUE expressions to achieve this were derived by Raju 关21兴. however. Different ments as shown in Fig. approach. 关5兴. which explains specific modeling issues. tual crack closure technique.1 Formulas for two-dimensional analysis In a two-dimensional finite element plane stress. undeformed finite element mesh and b兲 deformed Fig. equations are presented to calculate mixed. where node after node is sequentially where X i and Z i are the shear and opening forces at nodal released along the crack. eight-noded elements were used in connection with the vir- cussed in the following section. Elementwise opening—where edge and midside nodes are released—provides a kinemati- cally compatible condition and yields reliable results. and the shown in Fig. Node- wise opening/closing. (4) 2⌬a i ᐉ 3. Nodewise opening/closing. It is identical to elementwise ments at node ᐉ as shown in Fig. is possible for the four-noded ele- point i and ⌬u ᐉ and ⌬w ᐉ are the shear and opening displace. 5共a兲. opening in this case as the crack is opened over the entire placements required to calculate the energy ⌬E to close the length of the element. Nodes at the top surface and the bottom surface of the discontinuity have identical coordinates. or two nodes with identical co- the work ⌬E required to close the crack along one element ordinates coupled through multipoint constraints if a crack side therefore can be calculated as propagation analysis is desired. however. (2) For a crack propagation analysis. This lets the elements connected to the top surface of the crack deform independently from those connected to the bottom surface and allows the crack to open as shown in Fig. which 3 EQUATIONS FOR USING THE VIRTUAL was demonstrated in Refs. (3) 2⌬a i ᐉ three-dimensional solids.

where ⌬a is the length of the elements at the crack front as where it is assumed that the two-dimensional model is of above. a兲 Node- wise crack opening for four-noded element and b兲 crack opening for eight-noded element. are calculated for eight-noded el.Appl Mech Rev vol 57. The G I⫽⫺ 关 Z 共 w ⫺w ᐉ * 兲 ⫹Z j 共 w m ⫺w m * 兲兴 . He also showed that the sliding and opening behind the crack tip are calculated at equations are applicable if triangular elements. The crack surface ⌬A created is calculated as ⌬A⫽⌬a⫻1. In addition to the relative displacements at release rate. 6 Kinematic compatible crack opening/closure. face u ᐉ and w ᐉ and the displacements u ᐉ * and w ᐉ * at the The mode I and mode II components of the strain energy lower crack face. While the original paper by Rybicki and 共nodal point i兲 the forces X j and Z j at the midside node in Kanninen is based on heurisitic arguments 关19兴. G I . (5) 2⌬a i ᐉ relative displacements behind the crack tip are calculated from the nodal displacements at the upper crack face u ᐉ and 1 w ᐉ 共nodal point ᐉ兲 and the nodal displacements u ᐉ * and w ᐉ * G II⫽⫺ 关 X 共 u ⫺u ᐉ * 兲 ⫹X j 共 u m ⫺u m * 兲兴 . (6) 2⌬a i ᐉ at the lower crack face 共nodal point ᐉ*兲. . In addition to the forces X i and Z i at the crack tip unit thickness 1. 7共b兲: points m and m * are required. March 2004 Krueger : Virtual crack closure technique: History. no 2. nodal points ᐉ and ᐉ* the relative displacements at nodal ements as shown in Fig. approach. and G II . obtained by nodal points ᐉ and ᐉ* from displacements at the upper crack collapsing the rectangular elements. are used at the crack tip. and applications 113 where ⌬a is the length of the elements at the crack front and 1 X i and Z i are the forces at the crack tip 共nodal point i兲. Raju proved front of the crack 共nodal point j兲 are required. respectively. which are calculated from Fig. The relative the validity of the equation 关21兴.

The mode I and mode mode components as II components of the strain energy release rate. . G I .114 Krueger : Virtual crack closure technique: History. ity of the stress field at the crack tip. 关21兴: G T ⫽G I⫹G II⫹G III . no 2. 8 have crack surface ⌬A created is calculated as ⌬A⫽⌬a⫻1. and applications Appl Mech Rev vol 57. singularity elements are not required ⫹Z j 兵 t 21共 w ᐉ ⫺w ᐉ * 兲 ⫹t 22共 w m ⫺w m * 兲 其 兴 . these unit thickness 1. The elements with quarter-point nodes as shown in Fig. and G II are calculated for eight-noded singularity elements using the simplified equations given in Ref.0.25. special two-dimensional crack tip displacements u m * and w m * at the lower crack face 关21兴.0. The elements are obtained by collapsing one side of the rectan- total energy release rate G T is calculated from the individual gular elements. 0. The equations are also applicable if trian. March 2004 displacements at the upper crack face u m and w m and the at the crack tip. However. quarter-point elements accurately simulate the 1/冑r singular- gular parabolic elements. obtained by collapsing the para. are used at the crack tip 关21兴. been proposed in the literature 关21. as shown in Fig. and 1. Triangular quarter-point bolic rectangular elements. approach. 7 Virtual crack closure technique for 2D solid elements. a兲 Virtual crack closure technique for four-noded element 共lower surface forces are omitted for clarity兲 and b兲 virtual crack closure technique for eight-noded element 共lower surface forces are omitted for clarity兲. Based on the where it is assumed that the two-dimensional model is of location of the nodal points at ␰⫽0. (7) where G III⫽0 for the two-dimensional case discussed. 8共b兲. Therefore. 1 The VCCT proposed by Rybicki and Kanninen did not G I⫽⫺ 关 Z 兵 t 共 w ⫺w ᐉ * 兲 ⫹t 12共 w m ⫺w m * 兲 其 2⌬a i 11 ᐉ make any assumptions of the form of the stresses and dis- placements.56 –58兴. (8) Fig.

8共b兲. Eqs. (9) obtained by collapsing one side of the rectangular elements. C. . a兲 Quadtrilateral elements with quarter-point nodes and b兲 collapsed quarter- point element.Appl Mech Rev vol 57. ⫹X j 兵 t 21共 u ᐉ ⫺u ᐉ * 兲 ⫹t 22共 u m ⫺u m * 兲 其 兴 . 共8兲 and 共9兲 for described in Ref. B. March 2004 Krueger : Virtual crack closure technique: History. no 2. Due to the fact that these elements are the quarter-point elements have cross terms involving the not readily available in most of the commonly used finite Fig. and applications 115 1 corner and quarter-point forces and the relative displace- G II⫽⫺ 关 X 兵 t 共 u ⫺u ᐉ * 兲 ⫹t 12共 u m ⫺u m * 兲 其 ments at the corner and quarter-point nodes. are used at the crack tip as shown in Fig. t 12⫽6 ␲ ⫺20. Note that for triangular elements the nodal forces have to be calculated where from elements A. (10) shape functions are also discussed in Ref. approach. Special 3␲ 1 rectangular and collapsed singularity elements with cubic t 11⫽6⫺ . Equations 共8兲 2⌬a i 11 ᐉ and 共9兲 are also valid if triangular quarter-point elements. t 21⫽ . 8 Singularity elements with quarter-point nodes at crack tip. 关21兴. t 22⫽1. 关59兴. and D around the crack tip. A special 2 2 six-noded rectangular element with quarter-point nodes is In contrast to regular parabolic elements.

no 2. . a兲 Delamination modeled with bi- linear 3D solid elements and b兲 delamination modeled with bilin- ear plate/shell type elements.2 Formulas for three-dimensional solids and plateÕ where the delamination is closed and the structure is intact is shell elements modeled using single nodes or two nodes with identical co- In a finite element model made of three-dimensional solid ordinates coupled through multipoint constraints if a delami- elements 关Fig.116 Krueger : Virtual crack closure technique: History. Nodes at the top surface and the bottom is modeled with eight-noded three-dimensional solid ele- Fig.2. coupled through multipoint constraints. For additional surface have identical coordinates and are not connected with information about singularity elements the interested reader each other as explained in the preceding section. 关58兴. only a section of the delaminated area that dimensional models. approach. equations are not provided. This is discussed in 9共b兲兴 the delamination of length a is represented as a two. The additional sues. nation front is represented by either a row of single nodes or two rows of nodes with identical coordinates. March 2004 element codes. For convenience.关60– 64兴. nation propagation analysis is desired. The delami- is referred to Refs. The undamaged section 3. which explains specific modeling is- dimensional discontinuity by two surfaces. 9 Delaminations modeled as two-dimensional discontinuity.1 Formulas for three-dimensional solids possible to obtain G III . detail in the Appendix. dimension allows us to calculate the distribution of the en- ergy release rates along the delamination front and makes it 3. 9共a兲兴 or plate or shell type elements 关Fig. and applications Appl Mech Rev vol 57. which is identical to zero for two.

(11) displacements behind the delamination at the top face node 2⌬A Li Lᐉ row ᐉ are denoted u Lᐉ . and w Lᐉ and at the lower face node row ᐉ* are denoted u Lᐉ * . Here ⌬A is the For twenty-noded solid elements. Hence. mode II. are calculated as ters as illustrated in the top view of the upper surface shown in Fig. and better identification in this and the following figures. and G III . Y Li . v Lᐉ * . row i. The mode I. 11 are Fig. ⌬a is the length of the elements at the late the strain energy release rate components at the element delamination front. and applications 117 ments is illustrated in Fig. X Li . All forces and displacements are obtained from the G II⫽⫺ X 共 u ⫺u Lᐉ * 兲 . (12) 2⌬A Li Lᐉ finite element analysis with respect to the global system. a兲 3D view 共lower surface forces are omitted for clarity兲 and b兲 top view of upper surface 共lower surface terms are omitted for clarity兲. (13) 2⌬A Li Lᐉ the illustration.Appl Mech Rev vol 57. and w Lᐉ * as shown in 1 Fig. col- mode III components of the strain energy release rate. . with ⌬A⫽⌬ab as shown in Fig. 10. The corresponding G I⫽⫺ Z 共 w ⫺w Lᐉ * 兲 . approach. 10 关65兴. and b is the width of the elements. 10 Virtual crack closure technique for four- noded plate/shell and eight-noded solid elements. v Lᐉ . 10共b兲. and Z Li denote the forces at 1 the delamination front in column L. For corner nodes 共location Li) as shown in Fig. 10. the equations to calcu- area virtually closed. Its use with respect to geometrically nonlin- ear analyses will be discussed later.z ⬘ ) that defines the normal and tangential coordinate directions at the delamina- 1 tion front in the deformed configuration has been added to G III⫽⫺ Y 共 v ⫺ v Lᐉ * 兲 .y ⬘ . March 2004 Krueger : Virtual crack closure technique: History. no 2. A local crack tip coordinate system (x ⬘ . umns are identified by capital letters and rows by small let- G II . G I .

Y Ki . Only one ⫹X L j 共 u Lm ⫺u Lm * 兲 ⫹ 21 X M i 共 u M ᐉ ⫺u M ᐉ * 兲兴 . approach. March 2004 1 and Z Ki denote the forces at the delamination front in column G I ⫽⫺ 关 1 Z 共 w ⫺w Kᐉ * 兲 ⫹Z Li 共 w Lᐉ ⫺w Lᐉ * 兲 K. and w Kᐉ * and at the top face node row ᐉ. 11共b兲兴. ⫹Y L j 共 v Lm ⫺ v Lm * 兲 ⫹ 21 Y M i 共 v M ᐉ ⫺ v M ᐉ * 兲兴 . . v Kᐉ * .67兴: Fig. (16) The equations to calculate the strain energy release rate components at the midside node 共location M i) as shown in where ⌬A L ⫽⌬ab as shown in Fig. respectively. The relative displacements at the corresponding 2⌬A L 2 Ki Kᐉ column K are calculated from the displacements behind the ⫹Z L j 共 w Lm ⫺w Lm * 兲 ⫹ 21 Z M i 共 w M ᐉ ⫺w M ᐉ * 兲兴 . Here X Ki . 11 关66兴. row i. 12 are as follows 关66. v Kᐉ and w Kᐉ 关Fig. as u Kᐉ . Similar definitions are applicable in column 1 M for the forces at node row i and displacements at node row G II ⫽⫺ 关 1 X 共 u ⫺u Kᐉ * 兲 ⫹X Li 共 u Lᐉ ⫺u Lᐉ * 兲 ᐉ and in column L for the forces at node row i and j and 2⌬A L 2 Ki Kᐉ displacements at node row ᐉ and m. no 2. Fig. 11 Virtual crack closure technique for cor- ner nodes in eight-noded plate/shell and twenty- noded solid-elements. (15) half of the forces at locations Ki and M i contribute to the energy required to virtually close the area ⌬A L . and applications Appl Mech Rev vol 57. Half of the 1 forces at location Ki contribute to the closure of the adjacent G III ⫽⫺ 关 1 Y 共 v ⫺ v Kᐉ * 兲 ⫹Y Li 共 v Lᐉ ⫺ v Lᐉ * 兲 area ⌬A J and half of the forces at location M i contribute to 2⌬A L 2 Ki Kᐉ the closure of the adjacent area ⌬A N .118 Krueger : Virtual crack closure technique: History. a兲 3D view 共lower surface forces are omitted for clarity兲 and b兲 top view of upper surface 共lower surface terms are omitted for clarity兲. (14) delamination at the lower face node row ᐉ* as u Kᐉ * .

March 2004 Krueger : Virtual crack closure technique: History.Appl Mech Rev vol 57. a兲 3D view 共lower surface forces are omitted for clarity兲 and b兲 top view of upper surface 共lower surface terms are omitted for clarity兲. Fig. . L j and Ni. 12 Virtual crack closure technique for mid- side nodes in eight-noded plate/shell and twenty- noded solid elements. 2 册 (19) 冋 where only one half of the forces at locations Li. 册 Instead of computing the strain energy release rate com- 1 ⫹ X N j 共 u Nm ⫺u Nm * 兲 . and applications 119 G I⫽⫺ 1 1 冋 1 Z 共 w ⫺w Lᐉ * 兲 ⫹ Z L j 共 w Lm ⫺w Lm * 兲 2⌬A M 2 Li Lᐉ 2 G III⫽⫺ 1 1 冋 1 Y 共 v ⫺ v Lᐉ * 兲 ⫹ Y L j 共 v Lm ⫺ v Lm * 兲 2⌬A M 2 Li Lᐉ 2 1 1 ⫹Z M i 共 w M ᐉ ⫺w M ᐉ * 兲 ⫹ Z Ni 共 w Nᐉ ⫺w Nᐉ * 兲 ⫹Y M i 共 v M ᐉ ⫺ v M ᐉ * 兲 ⫹ Y Ni 共 v Nᐉ ⫺ v Nᐉ * 兲 2 2 1 ⫹ Z N j 共 w Nm ⫺w Nm * 兲 . approach. 2 册 (17) 1 ⫹ Y N j 共 v Nm ⫺ v Nm * 兲 . 1 1 1 N j contribute to the energy required to virtually close the G II⫽⫺ X Li 共 u Lᐉ ⫺u Lᐉ * 兲 ⫹ X L j 共 u Lm ⫺u Lm * 兲 2⌬A M 2 2 area ⌬A M . 2 G I . G II . and G III may be calculated for an entire element. no 2. (18) ponents at the corner or midside nodes as described above. Half of the forces at locations Li and L j contrib- ute to the closure of the adjacent area ⌬A K and half of the 1 forces at locations Ni and N j contribute to the closure of the ⫹X M i 共 u M ᐉ ⫺u M ᐉ * 兲 ⫹ X Ni 共 u Nᐉ ⫺u Nᐉ * 兲 2 adjacent area ⌬A 0 .

14. . (22) 1 G I⫽⫺ 关 Z 共 w ⫺w Lᐉ * 兲 ⫹Z L j 共 w Lm ⫺w Lm * 兲 2⌬A M Li Lᐉ where the forces at locations Li. which are shaded in Fig. L j and Ni. The force at location M i is also calculated from ele- 1 ments A and B. which is identical to the procedure above.120 Krueger : Virtual crack closure technique: History. 13 Virtual crack closure technique 共ele- ment method兲 for eight-noded plate/shell and twenty-noded solid elements. March 2004 which may be advantageous in cases where the elements are 1 not of square or rectangular shape. For example. G II⫽⫺ 关 X 共 u ⫺u Lᐉ * 兲 ⫹X L j 共 u Lm ⫺u Lm * 兲 A three-dimensional twenty-noded singular brick element 2⌬A M Li Lᐉ with quarter points is shown in Fig. and applications Appl Mech Rev vol 57. G III⫽⫺ 关 Y 共 v ⫺ v Lᐉ * 兲 ⫹Y L j 共 v Lm ⫺ v Lm * 兲 2⌬A M Li Lᐉ putation of the strain energy release rate components along a circular or elliptical front where elements are trapezoidal the user may find this approach more suitable. approach. This is unlike the previous equations where four ele- ments contributed to the forces at locations Li. 13 are as follows 关66 – 68兴: ⫹Y N j 共 v Nm ⫺ v Nm * 兲兴 . A prism-shaped singular element is obtained by collaps- Fig. As mentioned above ⫹X M i 共 u M ᐉ ⫺u M ᐉ * 兲 ⫹X Ni 共 u Nᐉ ⫺u Nᐉ * 兲 the desired 1/冑r singularity of the stress field at the crack tip is achieved by moving the midside node to the quarter posi- ⫹X N j 共 u Nm ⫺u Nm * 兲兴 . (20) N j. for the com. The equations to ⫹Y M i 共 v M ᐉ ⫺ v M ᐉ * 兲 ⫹Y Ni 共 v Nᐉ ⫺ v Nᐉ * 兲 calculate the strain energy release rate components for one element as shown in Fig. ⫹Z N j 共 w Nm ⫺w Nm * 兲兴 . no 2. (21) tion. a兲 3D view 共lower surface forces are omitted for clarity兲 and b兲 top view of upper surface 共lower surface terms are omitted for clarity兲. N j are calcu- lated only from elements A and B. ⫹Z M i 共 w M ᐉ ⫺w M ᐉ * 兲 ⫹Z Ni 共 w Nᐉ ⫺w Nᐉ * 兲 13共b兲. L j and Ni.

G II . approach. G I . mode II. (23) Fig. 14 Virtual crack closure technique 共ele- ment method兲 for twenty-noded quarter point el- ements. Each set of node. The mode I. 15. are calculated single node or the degrees of freedom must be connected for a brick or prism singularity element using the simplified through multipoint constraints as if each set was a single equations given in Ref. March 2004 Krueger : Virtual crack closure technique: History. a兲 3D view 共lower surface forces are omitted for clarity兲 and b兲 top view of upper sur- face 共lower surface terms are omitted for clarity兲.Appl Mech Rev vol 57. and applications 121 ing a face of the element as shown in Fig. 关68兴: G I⫽⫺ 1 2⌬A M 冋 再 t 12 Z Li t 11共 w Lm ⫺w Lm * 兲 ⫹t 12共 w M ᐉ ⫺w M ᐉ * 兲 ⫺2t 12共 w Lᐉ ⫺w Lᐉ * 兲 ⫺ 共 w Nᐉ ⫺w Nᐉ * 兲 2 冎 再 ⫹Z Ni t 12共 w M ᐉ ⫺w M ᐉ * 兲 ⫺ t 12 2 共 w Lᐉ ⫺w Lᐉ * 兲 ⫺2t 12共 w Nᐉ ⫺w Nᐉ * 兲 ⫹t 11共 w Nm ⫺w Nm * 兲 冎 ⫹Z N j 再 1 2 冎 再 共 w Nᐉ ⫺w Nᐉ * 兲 ⫹ 共 w Nm ⫺w Nm * 兲 ⫹Z M i t 11 2 共 w Lm ⫺w Lm * 兲 ⫹t 21共 w M ᐉ ⫺w M ᐉ * 兲 ⫹t 22共 w Nᐉ ⫺w Nᐉ * 兲 ⫹t 22共 w Lᐉ ⫺w Lᐉ * 兲 ⫹ t 11 2 冎 再 1 共 w Nm ⫺w Nm * 兲 ⫹Z L j 共 w Lm ⫺w Lm * 兲 ⫹ 共 w Lᐉ ⫺w Lᐉ * 兲 2 冎册 . no 2. . and G III . and mode III components of the collapsed nodes at the front must either be defined as a strain energy release rate.

and applications Appl Mech Rev vol 57. no 2.122 Krueger : Virtual crack closure technique: History. March 2004 G II⫽⫺ 1 冋 再 X t 共 u ⫺u Lm * 兲 ⫹t 12共 u M ᐉ ⫺u M ᐉ * 兲 2⌬A M Li 11 Lm ⫺2t 12共 u Lᐉ ⫺u Lᐉ * 兲 ⫺ t 12 2 Nᐉ 共 u ⫺u Nᐉ * 兲 冎 再 ⫹X Ni t 12共 u M ᐉ ⫺u M ᐉ * 兲 ⫺ t 12 2 Lᐉ 共 u ⫺u Lᐉ * 兲 ⫺2t 12共 u Nᐉ ⫺u Nᐉ * 兲 ⫹t 11共 u Nm ⫺u Nm * 兲 冎 ⫹X N j 再 1 2 冎 再 共 u Nᐉ ⫺u Nᐉ * 兲 ⫹ 共 u Nm ⫺u Nm * 兲 ⫹X M i t 11 2 Lm 共 u ⫺u Lm * 兲 ⫹t 21共 u M ᐉ ⫺u M ᐉ * 兲 ⫹t 22共 u Nᐉ ⫺u Nᐉ * 兲 ⫹t 22共 u Lᐉ ⫺u Lᐉ * 兲 ⫹ t 11 2 冎 再 1 共 u Nm ⫺u Nm * 兲 ⫹X L j 共 u Lm ⫺u Lm * 兲 ⫹ 共 u Lᐉ ⫺u Lᐉ * 兲 2 冎册 . approach. 15 Collapsed twenty-noded solid singular- ity elements with quarter point nodes at crack tip . (24) Fig.

t 21⫽ ␲ ⫺2. a兲 3D view 共lower surface forces are omitted for clarity兲 and b兲 top view of upper surface 共lower surface terms are omitted for clarity兲. Eqs. 16 Virtual crack closure technique for four- noded plate/shell elements. L j and Ni. approach. 共23兲–共25兲 for the quarter-point elements 5␲ have cross terms involving the corner and quarter-point t 22⫽⫺ ⫹4. (26) forces and the relative displacements at the corner and 4 Fig. the forces at locations Li. In contrast to Eqs. no 2. 共20兲–共22兲 for regular brick elements. N j as well as M i are calculated only from elements A and B. t 12⫽ ␲ ⫺4.Appl Mech Rev vol 57. (25) where As mentioned in the section for regular twenty-noded brick elements. shaded in Fig. March 2004 Krueger : Virtual crack closure technique: History. 14共b兲. . which are t 11⫽6 ␲ ⫺20. and applications 123 G III⫽⫺ 1 冋 再 t 12 Y t 共 v ⫺ v Lm * 兲 ⫹t 12共 v M ᐉ ⫺ v M ᐉ * 兲 ⫺2t 12共 v Lᐉ ⫺ v Lᐉ * 兲 ⫺ 共 v Nᐉ ⫺ v Nᐉ * 兲 2⌬A M Li 11 Lm 2 冎 再 ⫹Y Ni t 12共 v M ᐉ ⫺ v M ᐉ * 兲 ⫺ t 12 2 共 v Lᐉ ⫺ v Lᐉ * 兲 ⫺2t 12共 v Nᐉ ⫺ v Nᐉ * 兲 ⫹t 11共 v Nm ⫺ v Nm * 兲 冎 ⫹Y N j 再 1 2 冎 再 共 v Nᐉ ⫺ v Nᐉ * 兲 ⫹ 共 v Nm ⫺ v Nm * 兲 ⫹Y M i t 11 2 共 v Lm ⫺ v Lm * 兲 ⫹t 21共 v M ᐉ ⫺ v M ᐉ * 兲 ⫹t 22共 v Nᐉ ⫺ v Nᐉ * 兲 ⫹t 22共 v Lᐉ ⫺ v Lᐉ * 兲 ⫹ t 11 2 冎 再 1 共 v Nm ⫺ v Nm * 兲 ⫹Y L j 共 v Lm ⫺ v Lm * 兲 ⫹ 共 v Lᐉ ⫺ v Lᐉ * 兲 2 冎册 .

the components for a nine-noded plate element as shown in Fig. The second technique allowed the nodes to have independent The equations to calculate the strain energy release rate rotations. For nodes of the upper and lower surfaces. and applications Appl Mech Rev vol 57.关68 –73兴. for the mode I components. are 3. 15. 2⌬A Li Lᐉ tional and rotational degrees of freedom for nodes with iden- tical coordinates. For the configurations and loads considered. approach.124 Krueger : Virtual crack closure technique: History. 16. a兲 3D view 共lower surface forces are omitted for clarity兲 and b兲 top view of upper surface 共lower surface terms are omitted for clarity兲. dimensional quarter-point brick element with 27 nodes is dis. (27) 2⌬A Li Lᐉ mode strain energy release was demonstrated in Ref. There- nodal forces have to be calculated from elements A. C. paper. the equations to calculate the strain energy release different equations for solid elements are given in Refs. shown in Fig. Fig. rate components at the element corner nodes 共location Li) 关22兴. A three. 关23兴. second technique appeared to provide preferable constraints. Two techniques to tie the nodes at the delamination 2⌬A Li Lᐉ front 共row i in Figs.2 Formulas for plate/shell elements 1 The use of four-noded plate elements to calculate the mixed. Additional information with respect to Fig. for the prism element. .and 1 nine-noded plate elements were derived by Wang and Raju G II⫽⫺ X 共 u ⫺u Lᐉ * 兲 . no 2. only the compat. For the four-noded rectangular plate element. March 2004 quarter-point nodes. 17 Virtual crack closure technique for nine-noded plate/shell elements. 关69兴. B. which are joined at the crack 1 G III⫽⫺ Y 共 v ⫺ v Lᐉ * 兲 . fore only the equations for this method are mentioned in this and D around the crack tip as shown in Fig. (29) tip. In the second technique. 16共b兲 关74兴. 17 are. as shown in cussed in Ref. Equations for the use of VCCT in conjunction with four. the yielding accurate strain energy release rates 关74 –76兴. the first method enforced the compatibility of transla.2. (28) 关74兴. Note that. G I⫽⫺ Z 共 w ⫺w Lᐉ * 兲 . 16 and 17兲 were discussed. where ⌬A⫽⌬a(b 1 ⫹b 2 )/2 is the crack surface closed as ibility of the translational degrees of freedom were enforced.

March 2004 Krueger : Virtual crack closure technique: History. The equations to calculate the mixed-mode energy re- G II兩 M ⫽⫺ 关 X 共 u ⫺u M ᐉ * 兲兴 . For example. kinematic constraints in the neighborhood of 2⌬A N Ni Nᐉ (32) the delamination front. 1 G I兩 M ⫽⫺ 关 Z 共 w ⫺w M ᐉ * 兲 Built-up structures are traditionally modeled and analyzed 2⌬A M M i M ᐉ using plate or shell finite elements to keep the modeling and ⫹Z M j 共 w M m ⫺w M m * 兲兴 .75兴. For geometric nonlinear analysis where large deformations 2⌬A N Ni Nᐉ may occur. with forces . (34) models showed differences in results near the free edges of the structure where the stress state is 3D 关77兴. Since many layers of brick elements through the thickness are often necessary to model the individual plies. delaminations or debondings are modeled with plate or shell 1 finite elements 关74. 2⌬A N Ni Nᐉ analyses are required when matrix cracks and multiple (35) delaminations need to be modeled at different ply interfaces. however. both forces and displacements obtained in the (38) global coordinate system need to be transformed into a local as given in Refs. (42) 2⌬A L Li Lᐉ 2⌬A M M i M ᐉ (30) with the equations for columns L and N unaltered. 关75兴 mesh re- G II兩 L ⫽⫺ 关 X 共 u ⫺u Lᐉ * 兲 ⫹X L j 共 u Lm ⫺u Lm * 兲兴 . The local crack tip system defines the tangential (x ⬘ . midside-crack front nodes.Appl Mech Rev vol 57. For future detailed modeling. A comparison G II兩 M ⫽⫺ 关 X 共 u ⫺u M ᐉ * 兲 2⌬A M M i M ᐉ of computed mixed-mode strain energy release rates ob- tained from plate models with values computed from 3D ⫹X M j 共 u M m ⫺u M m * 兲兴 . 2⌬A L Li Lᐉ come prohibitively large. which defines mode II. G III兩 M ⫽⫺ 关 Y 共 v ⫺ v M ᐉ * 兲兴 . Computed mixed-mode strain energy release rate components. and applications 125 1 1 G I兩 L ⫽⫺ 关 Z 共 w ⫺w Lᐉ * 兲 ⫹Z L j 共 w Lm ⫺w Lm * 兲兴 . 18共b兲. respectively. (31) computational effort affordable. assumptions. which defines mode I. The equivalent crack The local z ⬘ direction. approach. no 2.3 Formulas for geometrically nonlinear analysis 1 G III兩 N ⫽⫺ 关 Y 共 v ⫺ v Nᐉ * 兲 ⫹Y N j 共 v Nm ⫺ v Nm * 兲兴 . The vector through nodes i and k in the deformed configu- are the equivalent crack surfaces apportioned to corner. finement studies showed that computed G I . ⌬A M ⫽⫺ 32 ⌬ab 2 ordinate directions at the crack tip in the deformed configu- and ⌬A N ⫽⫺ 61 ⌬a 共 b 1 ⫹b 2 兲 (39) ration as shown in Fig. M. Forces at lease rate components are constant across the width of an node row i and displacements at node row ᐉ need to be element 关74兴. as shown in Fig. or mode I兲 co- ⌬A L ⫽⫺ 61 ⌬a 共 b 2 ⫹b 3 兲 . 18共b兲 for the two-dimensional case. The accuracy of the computed mixed-mode energy G I兩 M ⫽⫺ 关 Z 共 w ⫺w M ᐉ * 兲兴 . The transformation is required to correctly com- to pute the mixed-mode energy release rates G I and G II . Furthermore. The total energy release rate remains unaffected by this transfor- 1 mation. 关74兴. Here indices L. (40) 2⌬A M M i M ᐉ release rates depends on the accuracy of the forces and dis- placements obtained form the nonlinear finite element analy- 1 sis. 18共a兲. (37) 3. is perpendicu- surfaces are obtained by assuming that the strain energy re. ⫹Y M j 共 v M m ⫺ v M m * 兲兴 .关75兴. and G III did 2⌬A L Li Lᐉ (33) not converge when the structure above and below the plane of delamination was modeled with plate elements with dif- 1 ferent section properties 共thickness or layup兲. the (36) shell/3D modeling technique offers great potential for saving modeling and computational effort because only a relatively 1 small section in the vicinity of the delamination front needs G III兩 M ⫽⫺ 关 Y 共 v ⫺ v M ᐉ*兲 2⌬A M M i M ᐉ to be modeled with solid elements 关78兴. lar to the local x ⬘ direction. or mode II兲 and normal (z ⬘ . For eight-noded plate elements the M m terms transformed to the local x ⬘ -z ⬘ system at the tip as shown in are equal to zero and the equations at column M are reduced Fig. 18. and continuity of material properties and section stiffness in the vicinity of the debond when mode II components.z ⬘ ) that originates at the crack tip as column location. the size of 1 finite element models required for accurate analyses may be- G III兩 L ⫽⫺ 关 Y 共 v ⫺ v Lᐉ * 兲 ⫹Y L j 共 v Lm ⫺ v Lm * 兲兴 . 3D G II兩 N ⫽⫺ 关 X 共 u ⫺u Nᐉ * 兲 ⫹X N j 共 u Nm ⫺u Nm * 兲兴 . 17共b兲 and shown in Fig. and mode III components. G II .and ration defines the local x ⬘ direction as shown in Fig. in Ref. depend on 1 many variables such as element order and shear deformation G I兩 N ⫽⫺ 关 Z 共 w ⫺w Nᐉ * 兲 ⫹Z N j 共 w Nm ⫺w Nm * 兲兴 . N denote the coordinate system (x ⬘ . These prob- 1 lems may be avoided by using 3D models. (41) 2⌬A M M i M ᐉ lease rate components remain the same as before.

or mode III兲 coordinate direction. 18 Virtual crack closure technique for geo- metrically nonlinear analysis.4 Corrections for elements with different lengths or widths at the crack tip 1 3. The forces ments at the upper crack face u m ⬘ and w m⬘ and the displace. Additional to the relative displacements at nodal stress field at the crack tip to derive the corrected equations. hind the crack tip are calculated at nodal points ᐉ and ᐉ* the ideal case of identical element length can no longer be from the transformed displacements at the upper crack face assumed and corrections are required. Three. The relative sliding and opening be. Z i at the crack tip 共nodal point i兲 calculated for an ele- ments u m ⬘ * and w m⬘ * at the lower crack face. .1 Correcting for elements with different lengths at the G II⫽⫺ 关 X ⬘ 共 u ⬘ ⫺u ᐉ⬘ * 兲 ⫹X ⬘j 共 u m ⬘ ⫺u m⬘ * 兲兴 . 19.4. 3. March 2004 and displacements now expressed in the local system. 1 G I⫽⫺ 关 Z ⬘ 共 w ⬘ ⫺w ⬘ᐉ * 兲 ⫹Z ⬘j 共 w m ⬘ ⫺w m⬘ * 兲兴 . approach. In their original paper u ᐉ⬘ and w ᐉ⬘ and the displacements u ⬘ᐉ * and w ⬘ᐉ * at the lower Rybicki and Kanninen 关19兴 use the 1/冑r singularity of the crack face. Once auto- local crack tip system. which are calculated from displace. Z ⬘j are the forces at the crack tip the assumption that the element lengths ⌬a for the element 共nodal point i兲 and in front of the crack 共nodal point j兲 in the in front of the crack tip and behind are identical. matic mesh generators are used to create complex models. Z i⬘ and X ⬘j . ment length ⌬a 2 are known from the finite element analysis. For dimensional analysis additionally requires the definition of the two-dimensional eight-noded quadrilateral element with the tangential (y ⬘ . no 2. (44) 2⌬a i ᐉ crack tip All equations in previous sections have been derived under where X ⬘i .5. which quadratic shape functions this yields will be discussed in Sec. the relative displacements at nodal points m A sketch of a crack tip modeled with two-dimensional finite and m * are required. X i . and applications Appl Mech Rev vol 57.126 Krueger : Virtual crack closure technique: History. elements of unequal length is shown in Fig. Fig. points ᐉ and ᐉ*. a兲 Definition of lo- cal crack tip coordinate system and b兲 forces and displacements in local coordinate system 共lower surface forces are omitted for clarity兲. (43) 2⌬a i ᐉ 3.

which is a function of the distance r from the crack The displacement at locations ᐉ̃ and ᐉ̃ * are calculated using tip. the re- quired forces X̃ i . March 2004 Krueger : Virtual crack closure technique: History. The displacements are adjusted ␴ 共 r 兲 ⫽b ␴ ⬁ ⫽ ⫽ . the second method is less restrictive be- cause the results only depend on the finite element discreti- zation at the crack tip. both methods are identical. (46) Fig. however. approach. With the relationship of the forces established. 20共b兲 yielding ⌬a 2 ⌬a 2 X i ⫽2b ␴ ⬁ ⌬a 1/2 2 . 关53兴 which were computed for a length ⌬a 1 . With the relationship of the displacements ⌬a 1 1/2 established. (51) ⌬a 1 ⌬a 1 A relationship between the forces can be derived. however. crack. A variation of only the element length. 2D case. The different element lengths are accounted for by The stress tip field at the crack tip can be expressed as correcting the displacements behind the crack tip 共at node ᐉ兲. Z i at the crack tip 共nodal point i兲. the required displacements at locations ᐉ̃ and ᐉ̃ * X̃ i ⫽ Xi . yielding the corrected The calculation of the crack opening force Z̃ i is done accord. Meshing of arbi- trarily shaped delamination front contours will. yields equations equivalent to the 2D case discussed behind the crack tip earlier: . and applications 127 Required for the virtual crack closure technique. 21. which are the forces X̃ i . Fig. (53) G II⫽⫺ 1 2⌬a 1 X i 共 u ᐉ ⫺u ᐉ * 兲 ⌬a 1 ⌬a 2 冉 冊 1/2 . (47) u ᐉ̃ ⫽u ᐉ . The forces at the crack tip for element length ⌬a 1 and a linear interpolation for ⌬a 1 ⬎⌬a 2 as shown in Fig. G I⫽⫺ Z i 共 w ᐉ ⫺w ᐉ * 兲 .Appl Mech Rev vol 57.4. where the displacement varies turbed far field stress. a two-dimensional representation of a crack or delamination in a plate/shell or a 3D solid finite element models also requires a correction accounting for differences in element widths b along the front. (49) G II⫽⫺ 1 2⌬a 2 X i 共 u ᐉ ⫺u ᐉ * 兲 ⌬a 2 ⌬a 1 . A different approach for correcting the equations. (45) 冑r dA bdr by taking into account the shape functions of the elements or approximated by simple linear interpolation. Therefore. no 2. equations for the energy release rate components: ingly. For elements where b is the element width or thickness. which were computed 1 dX dX for an element length ⌬a 2 . 3. u ᐉ̃ * ⫽u ᐉ * . which have been calculated for the stress field at the crack tip. 19 Correction for elements with different lengths in front and ⌬a. cause element length ⌬a and width b to vary along the front. Z̃ i may be substituted with the forces X i . ␴ ⬁ is the undis.2 Correcting for elements with different widths along the delamination front For the derivation of the equations in the previous sections it had been assumed that the front is straight and that element widths b remain constant along the front. to match the forces X i . However. 20 for the an element length ⌬a 1 . 1 ⌬a 2 Z i obtained from finite element analysis. Z̃ i matching the relative displacements at does not depend on the assumption of the 1/冑r singularity of node ᐉ behind the crack tip. is depicted in Fig. For components mod- eled with four-noded plate/shell type elements or eight- noded solid brick elements the correction is straightforward as shown in Fig. (52) rected equations for the energy release rate components 2⌬a 2 ⌬a 1 G I⫽⫺ 1 Z 共 w ⫺w ᐉ * 兲 2⌬a 1 i ᐉ ⌬a 1 ⌬a 2 冉 冊 1/2 . and ␴ (r) is the stress in front of the linearly along the element edge. yielding the cor. with linear shape functions. (48) ⌬a 2 may be substituted with the displacements ᐉ and ᐉ* obtained directly from finite element analysis. which only depends on the length of the elements in front and be- hind the crack tip The calculation of the crack opening displacement w is 冉 冊 done accordingly. 20共a兲 ⌬a 2 are obtained through integration: and linear extrapolation is used for ⌬a 1 ⬍⌬a 2 as shown in X̃ i ⫽ 冕 0 ⌬a 1 dr r 1/2 ⫽2b ␴ ⬁ ⌬a 1/2 1 . (50) The method first described imposes an analytical relation- ship based on the 1/冑r singularity of the stress field at the crack tip.

yielding the corrected the higher-order plate elements 关74兴 may be used since the equations for the energy release rate components change of element width is already accounted for in the ex- pressions for the mixed-mode strain energy release rates. and applications Appl Mech Rev vol 57. 20 Correction for elements with different lengths in front and behind the crack tip.128 Krueger : Virtual crack closure technique: History.2. 3. approach. ⌬a.2 for ⌬A 1 ⫽ 21 ⌬a 2 b 1 and ⌬A 2 ⫽ 21 ⌬a 2 b 2 . may be compensated as G I⫽⫺ Z Li 共 w Lᐉ ⫺w Lᐉ * 兲 . no 2. (58) 2⌬a 2 b Li Lᐉ ⌬a 1 2 ⌬A 1 ⫹⌬A 2 Li Lᐉ ⌬a 1 1 ⌬a 2 G II⫽⫺ X Li 共 u Lᐉ ⫺u Lᐉ * 兲 . (55) 2⌬a 2 b ⌬a 1 1 1 ⌬a 2 G III⫽⫺ Y Li 共 v Lᐉ ⫺ v Lᐉ * 兲 . (59) 2 ⌬A 1 ⫹⌬A 2 ⌬a 1 1 ⌬a 2 G III⫽⫺ Y 共 v ⫺ v Lᐉ * 兲 . The set of equations given in Sec. (57) 2 ⌬A 1 ⫹⌬A 2 ⌬a 1 described in the preceding section. Fig. a兲 Shorter elements in front of the crack tip and b兲 longer elements in front of the crack tip. . (56) 2⌬a 2 b Li Lᐉ ⌬a 1 A simple equivalent expression for twenty-noded solid The additional variation of the element width b requires brick elements based on the equations presented in Sec. 3. March 2004 1 ⌬a 2 1 1 ⌬a 2 G I⫽⫺ Z 共 w ⫺w Lᐉ * 兲 .2.1 the separate calculation of the contributing element surfaces is not available. (54) G II⫽⫺ X 共 u ⫺u Lᐉ * 兲 . The 1 1 ⌬a 2 variation in element length.

Appl Mech Rev vol 57. 22 Virtual crack closure technique for arbi- trarily shaped front . approach. and applications 129 Fig. 21 Virtual crack closure technique for eight-noded solid elements with different widths Fig. no 2. March 2004 Krueger : Virtual crack closure technique: History.

The formulation of an extraction method for sample node in Fig. The pose a problem when computing mixed-mode energy release rates as the separation into in-plane shear 共mode II兲 and tear- ing 共mode III兲 is not defined. For an arbitrarily shaped front crack front on the computed energy release rates was studied the mode definition constantly changes along the contour as in Ref. rived with the assumption that the delamination front is For the procedure described above it is assumed that the straight. Modern pre- the definition of the modes is intuitive and constant for the processing software allows the modeling of almost any com- entire front: Mode I is caused by the out of plane crack plex configuration. release rates. The secant through adjacent nodes VCCT that yields accurate energy release rates for models defines the y ⬘ direction and mode III. appears to be an acceptable alternative. 关80兴. which defines 3. Ideally sharp corners generally do not exist so that modeling a rounded corner. 22. Forces at corners node row i and displacements at node row ᐉ need to be Delaminations with sharp corners—as shown in Fig. is perpendicular to the plane of delamination. 23共b兲. 22. A recently suggested modified approach uses stair stepped instead of smoothed fronts and thus avoids an increase in model size 关82兴. were not devel- opening. 24 Detail of a finite element mesh with modeled square insert . Finally. 关80兴. For a straight front as shown in Figs. as shown in Fig. A local crack tip coordinate system there- that did not have normality at the crack front when compared fore needs to be defined at each nodal point along the front to reference solutions. The method suggested has been applied successfully to a specimen with an embedded. The local z ⬘ direction.5 Procedures for arbitrarily shaped delamination transformed forces and displacements are used in the previ- front contours ously derived equations to calculate the mixed-mode energy The equations presented in the previous sections were de. however.130 Krueger : Virtual crack closure technique: History. delamination front. the local x ⬘ direction. no 2. The vector through nodes i and k in the deformed con- variation from the normal condition yielded a greater dis- figuration defines the local x * direction as shown for one crepancy. The method results in an increase of nodes and elements and the model may become large. Fig. 10–17 and 21 mesh remains normal to the delamination front. It was also found that an increased 关79兴. March 2004 3. The effect of a lack of normality at the nent tangential to the front. a兲 Unde- fined mode separation at corner nodal point and b兲 well-defined mode separation at rounded corner. The local plane of that lack normality is given in Ref. approach. 24 关81兴. these programs. The mode separation is well defined at the rounded corner and the pro- cedure described in the preceding section may be applied to define the appropriate local crack tip coordinate system. and mode III by the shear compo. delamination is defined by x * and y ⬘ . mode II by the shear perpendicular to the straight oped to guarantee the normality of the mesh and the delamination/crack front. Fig. is perpendicular to the y ⬘ and z ⬘ plane. 23共a兲— transformed to the local x ⬘ -y ⬘ -z ⬘ system at the tip. which defines mode I. and applications Appl Mech Rev vol 57. 23 Definition of mode separation at sharp corners. It was shown that the standard formulations for VCCT were not able to extract accurate values from models shown in Fig.6 Suggested solutions for delaminations with sharp mode II. square delamination as shown in Fig.

tual crack closure length and a the crack length兲. In addition it was veri- Fig. and G III obtained exist between the plies 关83.87兴. mentioned problem does not exist. 25共a兲. no 2. duction of an artificial thin resin rich layer that is assumed to cised in interpreting the values for G I . For crack virtual crack closure length. Although this technique Mathematical solutions of the near crack tip field indicate circumvents the issue. crack closure length ⌬a goes to zero has been associated It was shown in the literature 关84兴 that finite values for the with these stress oscillations near the crack tip. ⌬a/a⬎0. 25 Bimaterial interface. G II . result in this phenomenon has to be considered as the delamination is nearly constant mixed-mode ratios. b兲 depen- dence of computed energy release rate on element size at crack tip and c兲 upper and lower bounds for element size at crack tip.05 (⌬a is the vir- growth and delamination propagation in composite materials. Several other papers tip when crack growth occurs at interfaces between materials have addressed the problems of a bimaterial interface and the with dissimilar properties as shown in Fig. eg. The fact associated stress oscillations near the crack tip in the past that the mixed-mode ratio is undefined when the virtual 关88 –98兴. and applications 131 4 DEALING WITH THE PROBLEMS AT A rarely located at an interface between two plies of identical BIMATERIAL INTERFACE orientation. One way to circumvent this problem is the intro- Previous investigations have shown that care must be exer. Delamination propagation in using the virtual crack closure technique for interfacial this case occurs in a homogenious material and the above- delaminations between two orthotropic solids 关83– 86兴. a兲 Bimaterial interface. March 2004 Krueger : Virtual crack closure technique: History. it requires larger models with signifi- that stresses start to oscillate in the immediate vicinity of the cant refinement in the thin resin layer. approach. .Appl Mech Rev vol 57.

The approach used must be consistent strain energy release rate distribution across the width is when the definition of the energy release rates used for frac. larger values of ⌬a the energy release rate components were In a previous investigation. It was also shown that evaluating energy in Table 1. delaminated interface was located between a ⫹30° and ⫺30° the energy release rates evaluated with a larger ⌬a are in. prerequisite for an energy release rate that is to be used as a refined mesh of length c was used in the vicinity of the fracture criterion in a real situation.103兴.079 792 0. release rates by using larger values of ⌬a/a. mode mixety between these upper and lower bounds are cedure yields results similar to the analytical evaluation of typically very small and should prove acceptable for practi- the crack closure integral based on the near tip fields.015 625 0.246 0.132 Krueger : Virtual crack closure technique: History. approach. converges between a ⫹30° and ⫺30° ply 关100. ⫽1 mm) need to be avoided as shown in Fig. ⌬a⫽c/n ture predictions as well as that employed for material char.920 20 0. The distribution of the mixed- manner that the computed results are insensitive to the varia. it had been of the specimen and peaks near the edges accompanied by suggested to use element lengths at the crack tip in such a local mode III contribution. March 2004 Table 1. whereas for very small ⌬a they rates were computed for an ENF specimen with multidirec- are functions of element length as sketched in Fig. h. in this zone as shown in the detail sponds to the lengths of the elements adjacent to the crack of Fig.492 0. In another study the influence of the mesh size at the ergy release rate components become stationary. used to describe the stress state at the crack tip. where the en.078 125 0. 28 and 29 where the mode II strain energy the data reduction scheme is in agreement with the definition release rate is fairly constant across almost the entire width of a finite crack closure length. Variations in via nodal forces and displacements in the finite element pro. Along the length of the model. The ply 5 CONVERSION OF ENERGY RELEASE RATES thickness was suggested as a practical upper limit for ele. The influence of mesh refinement on the mode I oscillating results. Consequently. 26. This is a specimen is shown in Fig. 共SLB兲 specimen with multidirectional layup. 27. 关75兴.920 48 0. n.003 906 25 0.915 62 0. The element size 共length specimens confirm the suggestion made earlier for lower and height兲 should not be less than 101 of a ply thickness. the energy release mixed-mode strain energy release rates was studied by keep- rates are defined as the virtual crack closure integral over a ing the length of the refined zone c constant.101兴. 25共c兲.0) of ele- which corresponds to the diameter of two carbon tows in the ment lengths to be used. bounds (⌬a/h⬎0. A study indicated to a constant value as shown in Ref. however. 25共b兲. obtained for a wide range of materi- . ⌬a. 30. front. moderate and only very long elements (n⫽3. puted mixed-mode ratio on element size ⌬a.关86兴 that evaluating the energy release rates where the energy release rates are calculated. mixed-mode energy release found to be nearly constant.084 262 0.079 548 0. the number of elements. This element length. However.127 mm. Dependence of mixed-mode ratio in bimaterial interface on element size at the crack tipa Element length ⌬a Relative element Relative crack „mm… size ⌬aÕh closure length ⌬aÕa Mode ratio GII ÕGT Mode ratio GIII ÕGT 0.000 492 0. there is only a small dependence of com- Fig.030 7 0. K C . A consider- ement as shown in Fig. are given in to ensure a converged FE solution but large enough to avoid Table 2. The three-dimensional model of the SLB sensitive to material inhomogeneities that exist.000 984 0. For cal applications. where the delaminated interface was located total energy release G T ⫽G I⫹G II⫹G III . C12K/R6376 tape.123 0.079 509 0. but it should be established that shown in Figs. This crack closure length corre. fied in Refs. carbon/epoxy material modeled as shown in Fig.102. The ⌬a/a values where the oscillation begins is connected with the value of ␥ that computed energy release rates G II and G III . ⌬a/a.084 377 a Layup 关 ⫾30/⫺30/30/⫺30/30/30/⫺30/30/⫺30/⫺30/30/↑ ⫺30/30/30/⫺30/30/⫺30/⫺30/30/⫺30/30/⫾30兴 . This does not imply that the material tests must confirmed by the mode II and mode III distributions as be evaluated by FE models. must be chosen small enough ⌬a/h.000 246 0. Smeared or homogenized able database exists of experimentally determined fracture- properties would result in altered properties at the interface toughness values. For the range studied tion of the element length ⌬a at the crack tip as sketched in (n⫽3 up to 48兲. ply 关100. delamination length a⫽31. The corresponding values of relative element size. no 2. ply thickness h⫽0. delamination front.915 74 0. For smaller element sizes the assumption of modeling each ply as an orthotropic continuum is no longer valid.75 mm. The results Upper and lower bounds may be assumed for practical from the above studies of the multidirectional ENF and SLB applications as shown in Fig. This is acterization. mode ratio G I /G II is shown in Fig.001 969 0. 关85兴. The influence of mesh size on computed For the virtual crack closure technique.000 123 0. In metal fracture the stress intensity factor K is commonly quire smearing of the different layer properties over one el.920 45 0. 25共c兲 关99兴.031 25 0. 25共c兲. and applications Appl Mech Rev vol 57. which is hibit a significant variation with mesh refinement as shown given by r ⫺1/2⫹i ␥ .061 5 0. yields simi. and relative crack closure length. where the suming a resin-rich region mentioned above 关83兴. 26.062 5 0. 25共b兲. TO STRESS INTENSITY FACTORS ment length and height because larger elements would re. The tional layup. and increasing finite crack closure length. delamination tip was investigated for a single leg bending lar results to values obtained by an analysis performed as.1) and upper bounds (⌬a/h⬍1. do not ex- found in the expression for the crack tip singularity.

the calculated total energy while the assumption of a plane strain condition yields release rate G T exceeds the interlaminar fracture toughness G c as shown in Fig. Due to the fact that K I⫽ 冑 G IE 共 1⫺ ␯ 2 兲 . (62) the calculation of stress intensity factors is not necessary for the prediction of delamination onset or growth in composites the lengthy equations required to convert energy release rates 冑 to stress intensity factors for orthotropic and anisotropic ma- G IIE terials are not given in this paper and the interested reader is K II⫽ . K III⫽ 冑 G IIIE 共 1⫹ ␯ 兲 .Appl Mech Rev vol 57. 2 for IM7/8552 关14兴. for a given mixed mode ratio G II /G T . The as. Failure in composite materials is expected when. (60) where E is the Young’s modulus and ␯ is the Poisson’s ratio K II⫽ 冑G IIE. Fig. 关104兴. and applications 133 als. 26 Three-dimensional finite element model of SLB specimen . (63) 共 1⫺ ␯ 2 兲 referred to the Ref. no 2. It may therefore be desirable to convert computed strain energy release rates G to stress intensity factors K. approach. (61) 关53兴. (64) sumption of a plane stress condition yields K I⫽ 冑G IE. March 2004 Krueger : Virtual crack closure technique: History.

putational time. studies showing the application of VCCT to 2D and 3D finite and other properties across the entire width are inherently element analyses are discussed in more detail.25 1. Mesh size at the crack tip for SLB specimen with multidirectional layupa Length c of refined Number of elements Element length ⌬a Relative element Relative crack section „mm… n „mm… size ⌬aÕh closure length ⌬aÕa 3. approach.125 0. sively in the past. out of plane strains to be zero (␧ zz ⫽ ␥ xz ⫽ ␥ yz ⫽0). and applications Appl Mech Rev vol 57.5 3. the geometry.874 0.984 0. ply thickness h⫽0.134 Krueger : Virtual crack closure technique: History.0 48 0.0 24 0. However. March 2004 Table 2. 28 Influence of the number of elements in refined section on computed mode II strain energy release rate distribution across the Fig. a two- dimensional plane-stress model in the x-y plane imposes the 6. especially if many dif- sure technique is applied to fracture of composite structures ferent configurations have to be analyzed during the initial are referenced in this section.007 3 3. 30 Influence of the number of elements in refined section on width of a SLB specimen mode ratio distribution across the width of a SLB specimen . In general 2D models are preferred by SURE TECHNIQUE TO ENGINEERING PROBLEMS industry due to the fact that modeling time.0 7. Two examples from recent design phase. boundary conditions.3 mm.492 0.1 Applying the virtual crack closure technique with out of plane stresses to be zero ( ␴ zz ⫽ ␶ xz ⫽ ␶ yz ⫽0) and al- two-dimensional finite element analysis lows the displacement to be the free parameter. remains affordable. A plane- Two-dimensional finite element models where the crack or strain model in the x-y plane.0 12 0. as well as com- Selected engineering problems where the virtual crack clo. The effect of 2D modeling Fig.0 6 0.0 3 1. 27 Influence of the number of elements in refined section on computed mode I strain energy release rate distribution across the Fig.127 mm. 29 Influence of the number of elements in refined section on width of a SLB specimen computed mode III strain energy release rate distribution across the width of a SLB specimen Fig. on the other hand. 6 APPLICATION OF THE VIRTUAL CRACK CLO.003 65 3.937 0.014 5 3.001 83 a Layup 关 ⫾30/0/⫺30/0/30/04 /30/0/⫺30/0/⫺30/30/↑ ⫺30/30/30/0/30/0/⫺30/04 /⫺30/0/30/0/⫾30兴 . constant in a 2D finite element model. C12K/ R6376 tape delamination length a⫽34. which excessively constrains the plies.062 5 0. Additionally. no 2.029 3. imposes the delamination is simulated as a line have been used exten.968 5 0.

. The behavior of edge delaminations The skin was made of prepreg tape and had a nominal ply was investigated in Refs. 关122兴 and the durability of bonded joints in computed energy release rates was studied in detail in Ref. 关121兴. 关118兴 and the failure of thickness of 0. 31 关133兴. Refs. The failure of lap joints was investigated The influence of 2D finite element modeling assumptions on in Refs. 关116兴. 关131兴. 31 Skin/flange debonding specimen. ture toughness testing in Refs. 关123兴. The virtual crack closure technique was also used 关105兴. tion length. a兲 Specimen configura- tion and b兲 finite element model of skin/flange specimen. in Refs. Based on the results of this investigation it is recom. For more accurate predictions. an example is taken from the study of skin/stringer Two-dimensional finite element models have been used to debond failure as shown in Fig. representing the stringer. 关120兴. 关132兴. 关117兴. 关106 –114兴 and three-point bonded onto a skin as shown in Fig.129兴. to investigate the facesheet delaminating from the sandwich mended to use results from plane stress and plane strain core 关124 –127兴 as well as delamination buckling 关128. 关130兴 while the initiation specific to rotorcraft energy release rates and mixed-mode ratios with delamina. The specimens study the behavior of specimens used and suggested for frac. consisted of a tapered laminate.142 mm and a 关45/⫺45/0/⫺45/45/90/90/⫺45/ Fig. models as upper and lower bounds. An IM7/8552 bending specimen 关115兴. Two-dimensional models Delamination initiation from ply drops in general was stud- may also be used to qualitatively evaluate the variation of ied in Ref. Iosipescu specimens were studied graphite/epoxy system was used for both the skin and flange. and applications 135 assumptions is most marked for 45° plies because of their composite hat stringer pull-of specimens was examined in high in-plane Poisson’s ratio.Appl Mech Rev vol 57. Ref. a 3D To illustrate the application of VCCT to structural delami- analysis is required. nation. approach. however. 关119兴. 31共a兲. March 2004 Krueger : Virtual crack closure technique: History. flexbeams was investigated in Refs. while it is small for 0° and 90°. 关5兴. no 2. 关20兴.

no 2. 31共b兲. For this investigation. and ‘‘45’’ represents generally identified based on the shape of the total energy a 0°–90° fabric ply rotated by 45°. The initial crack was modeled at the loca- tion suggested by the microscopic investigation as shown in the detail in Fig.8 kN with a coefficient of variation of ascertain that the peak value had been captured. The tests were terminated when the flange debonded from the skin. approach. March 2004 45/0/45/⫺45兴 layup. a delamination formed at the top 45°/⫺45° skin ply interface. The flange layup ments. The value of the virtual delamination length and then decreased.9%. 35 Computed mixed-mode energy release rate along detected 45°/⫺45° skin plies delamination front after 200 000 load cycles . 32兲 at some length of 101. The total 8. the delami- nation growing between the skin top 45° and ⫺45° plies was extended by releasing multipoint constraints at the crack tip and in front of the crack tip.6 mm as shown in Fig.136 Krueger : Virtual crack closure technique: History. and applications Appl Mech Rev vol 57. The delami- damage onset load. This critical G is fabric. During a series of nonlinear finite element analyses. which is deter- were performed in a servohydraulic load frame where the mined through analysis as shown in Fig. a兲 Delamination buck- ling specimen and section modeled with three-dimensional solid elements and b兲 x ray showing the initial circular delamination and Fig. 32. The G T versus x specimens were mounted in hydraulic grips with a gage curve reached a maximum 共as marked in Fig. 0 represents a 0°–90° fabric ply. between top 45°/⫺45° skin plies Fig.2 mm to termined to be 17.208 mm nominal ply thickness. P. where the subscript f denotes mined to predict delamination onset. Micrographic investigations showed that at corners 2 and 3. 31共b兲. 32 Computed total strain energy release rate for delamination detected growth after 200 000 load cycles. 33 Computed mixed mode ratio for delamination between top Fig. Quasistatic tension tests release rate versus delamination length curve. Finite element solutions were ob- tained using the commercial ABAQUS®/Standard finite ele- ment software. was averaged from four tests and de. A critical energy release rate G c needs to be deter- was 关 45/0/45/0/45/0/45/0/45兴 f . The flange was made of a plain-weave puted at each tip location for the loads applied in the experi- fabric of 0. strain energy release rates were com- Fig. Eight-noded quadrilateral plane-stress 共CPS8R兲 elements with quadratic shape functions and a re- duced 共2⫻2兲 integration scheme were utilized for the geo- metric nonlinear analyses. nation was extended to a total simulated length of 2. 34 Delamination buckling specimen.

A comparison of 2D models in Refs.27.162–172兴.3 Applying the virtual crack closure technique to maximum mode I percentage. The problem of a composite stringer separating Delamination buckling has been investigated extensively us- from the skin has been investigated using plate and shell ing 3D models 关69. .Appl Mech Rev vol 57.78. Fig. nation 关138兴. Three-dimensional finite element models have been used to study the behavior of specimens traditionally used in fracture toughness testing 关18. chosen as the critical value to be used for the fatigue life prediction 关133兴. Ireman et al studied the damage propagation in composite ordinates. Similar investigations of skin/flange deb- onding are reported in Refs. the minimum value of analysis with solid finite elements G II /G T in Fig. 关158兴 and discontinuity by two surfaces with identical nodal point co. was analyzed in Refs. a兲 In- tact region in front of the crack modeled with single nodes and b兲 intact region in front of the crack modeled with double nodes tied by multipoint constraints. was els 关69.23. and applications 137 energy release rates and the corresponding mixed mode ra.77. of stitched composite structures 关141–143兴. no 2.144 –148兴. 36 Crack modeled as one- dimensional discontinuity. This peak in G T also corresponded to the 6. March 2004 Krueger : Virtual crack closure technique: History. 关139兴. Three-dimensional models were analysis and demonstrated its application to simple fracture also used to study the growth of near-surface delaminations toughness specimens. and 3D analysis for skin/stringer debond failure was per- ses performed by Glaessgen et al focused on the debonding formed in Refs. Damage in titanium- ments the delamination is represented as a two-dimensional grahite hybrid lamiantes was investigated in Ref. 关134 –137兴. 33. 关74兴.2 Applying the virtual crack closure technique to three-point bending specimen 关154兴 as well as the behavior analysis with plate and shell elements of edge delaminations 关155–157兴. 关173–175兴. Skin-stiffener debonding In a finite element model made of plate or shell type ele. Flanagan developed a code based on sublaminate structural elements 关159兴. 关66兴. 关105兴.160.130. ie.79.161兴. the curved-beam test and edge delami. in composite laminates under fatigue loading 关81. 6. Delamination tios G II /G T are plotted in Figs.149–153兴. 32 and 33. approach.67. 关140兴.98. Fracture mechanics analy.100–102. The maximum buckling has been investigated as well using plate/shell mod- total energy release rate G T as marked in the graph.

25 and s tigue loading. Stress maxima in the range of 220–240 ⬇0. Due to the extensive computation times seen for 7 CONCLUDING REMARKS 3D models. March 2004 The example chosen is taken from an investigation of using a contact processor that utilizes a contractor target con- delamination buckling and growth as shown in Fig. The distributions of the mixed-mode ⫿5兴 layup. approach to assess the damage tolerance of composite struc- was used for the nonlinear simulations 关177兴. and applications Appl Mech Rev vol 57. their maximum perpendicular to the load direction (s⬇0. shown in Fig. global maximum of the total energy release rate G T is now mens were subjected to tension-compression (R⫽⫺1) fa.138 Krueger : Virtual crack closure technique: History.22 tion was monitored by the Moiré technique. Ef- Fig. 34共b兲. Figure 34共a兲 shows 关81. Interpenetra. Additional x-ray photographs 关as the entire investigation are discussed in Ref. The damage near the surface as shown in Fig. 34共b兲兴 were used to verify this method. 34共b兲. Dur.160. The increased interest in using a fracture mechanics–based mulated according to a continuum-based 3D shell theory. The values for interlaminar shear failure also reach ing the experiment. The specimens were made of T300/914C tape the specimen outline. the out-of-plane 共ie. mum is reached perpendicular to the loading direction. The speci. tures in the design phase and during certification has re- tion of the layers in the delaminated region was prevented by newed the interest in the virtual crack closure technique. for. post-processing procedures.f to calcu- late strain energy release rates . approach. 35. however. 37 Flow chart of routine extract.5.176兴. after 200 000 load diameter at interface 2/3 was assumed to simulate an impact cycles兲 as shown in Fig. buckling兲 deforma. 关179兴. Finite element solutions were obtained using NOVA finite element software.75) with a second maximum occurring in the loading N/mm2 had shown to yield stable delamination growth.72) and a second maximum for s⫽0 and s⫽0. 34共a兲 cept applying the penalty method 关178兴. the size and shape of the delami. a special layered element with eight nodes. direction. This also holds for the G I distribution where again the maxi- nated sublaminate were determined from this information. Using numerical and s⬇0. are plotted in Fig. details of the section modeled and the material with a 关⫾5/⫹45/⫾5/⫺45/0/⫾85/0/⫺45/⫿5/⫹45/ deformed FE model. found perpendicular to the loading direction (s⬇0. An embedded circular delamination of 10 mm energy release rates along the front s 2 共ie. Details of to the numerical model. no 2. which were used as input loading direction. in yielding the delamination contours. it decreases to zero.

4兲 Read element forces. 37 as an example. 36共b兲兴. 3. however. The same logic applies to a finite element model made of plate/shell type elements or 3D solid ele- was discussed. Manfred König of the University of Stuttgart and 2兲 Provide input interface to read variable problem specific Jeffery Schaff of United Technologies Research Center for external data such as crack length. to the element forces at nodes. 36共a兲. using nodal point attention has to be given to the modeling of the discontinuity coordinates. 3兲 Create output file to store echo print of control param- APPENDIX: IMPLEMENTATION OF THE VIRTUAL eters. The flow chart.3. or internal common blocks and arrays for further particular software. processing step. is not a comprehensive literature survey. etc. no 2. and 3. however. Fritz Buchholz of during execution: the University of Paderborn for his generous help in provid- ing old manuscripts and insight into the history of the Virtual 1兲 Establish interface with finite element sortware if re- Crack Closure Technique. y ⬘ . approach. the equations given for the individual element type and gation analysis is to be performed. and G III need to be computed by ment software if a user interface is provided. 5兲 Store the data retrieved in step 共4兲 in external files. 36 and the calculation of the mixed-mode 7兲 Calculate local coordinate system x ⬘ . input data. In both cases the strategy is similar and 6兲 Calculate area virtually closed.1. or output file in ASCII format or data bank if data are lease rates using the virtual crack closure technique 共VCCT兲 stored externally. etc. ments. Fig. This Composites Material MIL-17 Handbook. March 2004 Krueger : Virtual crack closure technique: History. Young’s modulus. the energy release from specific common blocks or arrays of the finite ele- rate components G I . user-written subroutines. at nodes have to be retrieved and summed to obtain forces at The approach used in the virtual crack closure technique the crack tip. The use of multipoint correct for different element lengths and widths if neces- constraints may also be preferable if the finite element code sary. provided this option has been made available for this base. Nodes at the top surface and the bottom crack tip from the forces at element nodes by summing surface have identical coordinates. 2D discontinuity by two surfaces. is modeled using single nodes 关Fig. Alternatively retrieve required data as described in the main text. the nodal point displacements but more a summary and review of issues relevant to the and the nodal point coordinates. dimensional finite element model is represented as a one. 3. as shown in Fig. However. Thanks are due to and control parameters such as element type. This step may is represented by either a single node 关Fig. The author would also like to quired. 36共a兲兴 or two nodes with c兲 Transform forces and displacements to local crack tip identical coordinates coupled through multipoint constraints coordinate system. shows an independent post- Nevertheless. which has been the feature would render obsolete the step where element forces motivation for the overview presented. These subroutines may either inter- face directly with the finite element code during its execu. . depicted in successful application of the virtual crack closure technique. or forces at constraints. Ed Glaessgen of NASA Langley Research Center and Dr. 3. three-element sol. nodal point Currently. and retrieved data as well as intermedi- CRACK CLOSURE TECHNIQUE ate and final results. The crack tip either to the upper or the lower surface. strategy to use VCCT.5. Vasyl Harik of ICASE for the thorough review of this report. and nodal displacements from binary result offer the choice for calculating the mixed-mode energy re- file. an effort was made to pay proper tribute to the processing procedure where input data for the VCCT was relevant contributions made during a quarter century since extracted directly from an ABAQUS® binary result file. many discussions and their valuable input. Equations for the use of the technique in con. 5 and 36. based on results from finite element analysis requires access The paper. shaped delamination front as discussed in Secs. and z ⬘ at crack energy release rate as outlined in the flowchart of Fig. 36共a兲兴 or two be skipped if forces at constraints are available directly nodes with identical coordinates coupled through multipoint from the finite element software. It is generally up to the user to decide how to model the crack tip and the intact region. the large commercial finite element codes do not coordinates.Appl Mech Rev vol 57. dimensional discontinuity by a line of nodes as shown in 8兲 a兲 Obtain the forces at the crack tip and in front of the Figs. constraints 关Fig. is the same if data are re- trieved from a data file in ASCII format or if a user-written ACKNOWLEDGMENTS subroutine interfaces directly with the finite element software The author gratefully acknowledges Dr. generally not hard coded into the subroutine. the use 9兲 Calculate mixed-mode strain energy release rates using of multipoint constraints may be preferred if a crack propa. The undamaged section or the sec- b兲 Obtain relative displacements between the surfaces tion where the crack is closed and the structure is still intact from the nodal point displacements. however. are not con. ids as well as plate/shell elements were given and insight The application of the virtual crack closure technique into the application to engineering problems was provided. 36共b兲兴. tip for geometrically nonlinear analysis and arbitrarily As mentioned in Sec. where the delamination of length a is represented as a junction with two-dimensional elements. and applications 139 forts are underway to incorporate these approaches in the used provides forces at constraints as a standard output. 37. or operate as an entirely separate post. the forces at common nodes from elements belonging nected with each other as shown in Fig. Therefore. data- tion. the crack of length a in a two. G II . thank Dr.6. 关Fig. which is Dr. The the original publication. ⌬A. access during the calculation.

Fract. 1801–1821. Int. Failure mechanics of edge delamination of compos. An equivalent domain integral stress intensity factors by a modified crack closure integral. 关22兴 Buchholz FG. Am Inst of matrix composites: Problems and assessment. J. Phila. nation. Stress intensity factors as the frac- 关21兴 Raju IS 共1987兲. Eng. Eng. 57– 62. Rev. Int. intensity factors and energy release rate calculations for mixed frac- ior of a first generation graphite/epoxy composite. Part B 27B共2兲. pp. Mech. 关19兴 Rybicki EF and Kanninen MF 共1977兲. PA. 24. 129–143. 89–105. 140–167. Compos. 174 –183. Composite Materials: Testing and Design. Crack tip and associated domain inte- terials Engineering Research Laboratory Ltd. Delamination—A damage mode in composite struc. energy release rates along straight and curved delamination fronts of buckling. 25. Swansea. R33–R42. 433–528. Philadelphia. G IIc : Shear measurement or sheer Myth?. 28. 59共4兲. rials: Testing and Design. 707–725. J. West Conshohocken. Eng. H. 405– 关14兴 Hansen P and Martin R 共1999兲. Struct. 52共2兲. 353–364. Dreyer. PA. Technol. J. J. winding road to standardization. Publication No 关1兴 Garg AC 共1988兲. Fract. Materials. An interaction energy integral method and 3D-applications of the improved and generalized modified crack for computation of mixed-mode stress intensity factors along non- closure integral method. SJ Hooper 关40兴 Li FZ. Rev. Eng. Fract. Maximum energy release rate distribution from delamination testing. Eng. Int. unidirectionally laminated DCB. Uni- versal crack closure integral for SIF estimation. v. Lee SM. for combined tensile and shear loading by only a virtual crack exten- 303–322. Fract. Mixed mode bending method for 关36兴 Claydon PW 共1992兲. J. Elsevier Science. Int. Methods Eng. Application of the finite element 关49兴 Park O and Sankar BV 共2002兲. 9. N68171-98-M-5177. Numer. London. 615– 642. Virtual crack extension: A general finite element delphia. Technol. 187–207. An energy release rate 关46兴 Banks-Sills L 共1991兲. ASCE Press. Fract. Part B 29共1兲. method for three-dimensional mixed-mode fracture problems. 11. M. 487–502. Fract. 关24兴 Ingraffea A and Wawrzynek. Carter BJ. A finite element analysis of stress intensity factors ASTM STP 1206. 60– 81. 27共6兲. 关20兴 Rybicki EF. Compos. Philadelphia. 55共4兲. R. On the virtual crack extension technique for stress 共1997兲. nates. 233–344. 关15兴 Lee SM 共1990兲. H 共1988兲. UK. Mech. 42共6兲. Part B 28B共4兲. 4ENF and MMB delamination 421. Mech. PA. U. Mech.140 Krueger : Virtual crack closure technique: History. Fract. Mech. Eleventh Volume. 19. and G Yagawa 共eds兲. 关34兴 Delorenzi HG 共1985兲. Compos. Damage in Composite material behavior. 21. 共MERL兲. Composite Materials: Fatigue equivalent domain integral method. Mech. Am Soc for Testing and Materials. Int Conf on Designing Cost-Effective Composites. pp. energy release rate in delaminated plates. Fract. ness of laminated composites. Kussmaul. by the III delamination testing. 464 – 479. Interlaminar fracture toughness: The long and Mechanics. 1332–1342. method in the two-dimensional analysis mixed-mode crack problems. 217–235. Grebner. Compos. 38. 14共1兲. An equivalent domain integral edge crack torsion ECT test for Mode III interlaminar fracture tough. Research Transformed into Practice: ergy release rates during delamination growth. Proc of the 关3兴 Pagano NJ and Schoeppner. 470– 487. Eng. 12–19. 299–319. Round robin testing for mode I of fracture parameters in side-grooved compact specimen. Aeronaut and Astronaut. 关38兴 Ishikawa H 共1980兲. 129–145. 429– 434. PA. Eng. A bilinear failure criterion for mixed-mode delami. March 2004 10兲 Repeat steps 共6兲 through 共9兲 for all nodes along the Implementations of NSF Research. Krüger. 24共9兲. 251–274. Appl. M 关39兴 Sha GT 共1984兲. Fract. 55共2兲. lem. 557–584. and O’Brien TK 共1997兲. Streiner P. Rev C INTES GmbH. Schmueser DW. ASTM STP 1330. Technol. FRANC2D: A case study in 关50兴 Schön J and Andersson B 共1998兲. linear elastic fracture mechanics. grals from momentum and energy-balance. Proc of Am Soc for . 269–281. 16. tures. Rev. interlaminar fracture toughness of composite materials. Composite Materials: Fa. 44共10兲. Mech. Mater. KH. Mech. the virtual crack extension method for calculation of the rates of 关12兴 Reeder JR 共1993兲. Tayebi AK. Fract. A finite element calculation of 关45兴 Shivakumar KN and Raju IS 共1992兲. Am Soc for integration form of the virtual crack extension method. Testing and Materials. Application of the finite element method to approach for stable crack growth in the free-edge delamination prob. Fract. Composite Mate. Int. 关37兴 Hwang CG. Fract. 21. approach. determination of crack tip stress intensity factors. Eng. 21共2兲. 关10兴 Reeder JR and Crews JH 共1990兲. Technol. method for computing crack-tip integral parameters in nonelastic. Mater. technique for J-integral evaluation. 关23兴 Miravete A and Jiménez MA 共2002兲. 183–193. Alberti. Calculation of fracture- 关16兴 Martin RH 共1991兲. Pineridge Press. 195–220. Res. and Needleman A 共1985兲. Mech. 19. Characterization and analysis of delamination fracture 关29兴 Parks DM 共1974兲. and Fracture. Characterization of delamination onset and 关30兴 Parks DM 共1977兲. 217–233. 关48兴 Gosz M and Moran B 共2002兲. energy release rate. 12. Wawrzynek PA. 375–384. Composites. Int. J. J. 3-D crack configurations. AR Luxmoore and DRJ Owen 共eds兲. and Krome. characterisation of S2/8552 and IM7/8552. 关35兴 Lin SC and Abel JF 共1988兲. Am Soc for Testing and Materials. Knapp H. Methods Eng. 37共4兲. Evaluation of the split cantilever beam for mode mechanics parameters for an arbitrary 3-dimensional crack. J. Redesign of the mixed-mode bend. 851– 867. An equivalent domain integral III delamination test for composites. Res. pp. 447– 461. P 共1995兲. Fract. 3–18. Springer Verlag. 28. Lee EW. 26共6兲. Mech. 10. Compos. Evaluation of the 关44兴 Raju IS and Shivakumar KN 共1990兲. 关18兴 Li J. 关25兴 Singh R. Composites. 关28兴 Hellen TK 共1975兲. Third Volume. ASTM STP 775. Methods Appl. J. Mech. K. 关27兴 Krüger R. Computation of local 关2兴 Bolotin VV 共1996兲. R243–R246. UK. The virtual crack extension method for nonlinear growth in a composite laminate. New York. and Fox J 共1977兲. Energy release rate calculations by the finite 关9兴 O’Brien TK 共1998兲. Composites. 2D. 133–146. A Kelly and C Zweben 共eds兲. and Ingraffea AR 共1998兲. ASTM STP 1110. Philadelphia. König M. User Manual. 931–938. J. and Wüstenberg H. Shih CF. 关42兴 Nikishkov GP and Atluri SN 共1987兲. J. GA 共2000兲. Washington. Hertford. Doltsinis J. posite Materials. 关33兴 Delorenzi HG and Shih CD 共1983兲. Compos. a generalized 3D virtual crack extension method.and ENF—specimens. J. Seventh Volume. IMechE Conference Transactions. On the energy release rate and the J-integral for design. 关7兴 Martin RH 共1998兲. Comput. Wawrzynek PA. Mech. 521–542. 关41兴 Moran B and Shih CF 共1987兲. 关31兴 Parks DM 共1978兲. Fract. Stuttgart. 关11兴 Reeder JR and Crews JH 共1992兲. 关4兴 Tay TE 共2003兲. ites. Appl. 共1985兲. A stiffness derivative finite element technique for in composites: An overview of developments from 1990 to 2001. A comparison of methods 共ed兲. 13th Volume. Eng. no 2. ing delamination test to reduce nonlinear effects. Fract. Characterizing static and fatigue interlaminar fracture behav. On the method of virtual crack extension. AIAA J. Fract. 243–266. SN Atluri planar crack fronts in three dimensions. Fracture mechanics. Compos. Development of an improved mode 关43兴 Nikishkov GP and Atluri SN 共1987兲. Mech. 关26兴 St. 29共5兲. Vol 2. growth and stability. 42共6兲. 1–32. Fract. Numer. 961–969. element method. Computational Mechanics ’88. and Ingraffea AR 共1998兲. Comprehensive Com. 69共3兲. J. Crack-tip force method for computing method to prediction of onset of delamination growth. J Colville and A Amde 共ed兲. 60共2兲. Fract. Calculation of strain-energy release rates with higher ture parameters for delamination crack growth in composite lami- order and singular finite elements. Fract. thermomechanical fracture. 9. 15. Int. Eng. Delamination of polymer 34th AIAA/ASME/ASCE/AHS/ASC SSDM Conf. Mech. and applications Appl Mech Rev vol 57. Mech. Numerical Methods in Fracture 关6兴 O’Brien TK 共1998兲. ture mode. ASTM STP 1242. 关13兴 König M. 关5兴 O’Brien TK 共1982兲. Appl. Sci. REFERENCES PERMAS-FM. 56共1兲. Delaminations in composite structures: its origin. On Res. 1270–1276. 1200–1212. Ma. Germany. 关17兴 Robinson P and Song DQ 共1994兲. Calculation of mode-separated en- transfer of software technology. Am Soc for Testing and Materials.K. and Gädke. and Schneider T 共1993兲. PA. Mech. Eng. La Jolla. Composite interlaminar shear fracture toughness. 关47兴 Chow WT and Atluri SN 共1997兲. Variational aproach for a new direct- tigue and Fracture. 83–92. 226. 3-D Elastic-plastic investigation 关8兴 O’Brien TK and Martin RH 共1993兲. New York. Incorporating interlaminar fracture mechanics into 关32兴 Delorenzi HG 共1982兲. Int. Mech. Eng. for calculating energy-release rates. delamination front and write results to output file. Eng. DCB. 935–959. Eng. sion.

关53兴 Broek D 共1991兲. Int. 2597–2613. J. Am Soc for Testing and Materials. Int. cial cracks in bimaterial media. release rates in fracture of interfacial cracks. Proc of Int Conf. J. Sixth Volume. Fract. Sistla R. 57共4兲. Eng. Ninth Volume. Johnston IW. 29th Volume. Fract. 679– 686. 2595–2609. Germany. putation by modified crack closure integral 共MCCI兲. VA. Strain-energy release rate 关91兴 Qian W and Sun CT 共1998兲. Mech. Mech. 13–20. 关86兴 Sun CT and Manoharan MG 共1989兲. 关77兴 König M. Fract. March 2004 Krueger : Virtual crack closure technique: History. Marques AT. Solids Struct. A virtual crack- modes in composite delamination problems. Krishnamurthy T. Dattaguru B. Res. Composite Materials: Theory and Practice. and Aminpour MA 共1988兲. 25共2兲. Philadelphia. 55共2兲. no 2. Fract. cracked bodies. and Vijaya. Convergence of strain 共1990兲. determining stress intensity factors in finite element analysis of Mech. Int. Accuracy Estimates and Adaptive Refinements in Finite El. personal communication. ASTM STP 1383. 28共1兲. 36. Robinson and Assoc. 975–983. nates. 558 –590. Solids Struct. Fract. ASTM STP 1321. Decomposed crack clo. Commun. Eng. On strain energy release rates for interfa- quarter point elements.Appl Mech Rev vol 57. George S. Ramamurthy TS. 977–988. 38共21兲. General- 关72兴 Singh R. Am Soc for Test- 关57兴 Barsoum RS 共1977兲. Triangular quarter-point elements as elastic and ing and Materials. 30共3兲. Int. Stress intensity factors and energy release unnecessary. Mech. Effect of shear de. Mech. Three. Mech. Dayton. A shell/3D modeling technique Revised Edition. Methods Eng. 25– 44. Flügge 共ed兲. 1493–1498. Fatigue and Fracture Mechan- elements. 165–178. Mater. directional composite DCB. Methods Eng. 9. Finite- kumar K 共1994兲. 关61兴 Ramamurthy TS. PA. 179–186. Fract. Am Soc for Testing and Materials. 10. Interline Publications. Eng. ASTM STP 1059. 62共5兲. Soc for Composites. AVK Murthy and FG Buchholz 关93兴 Sun CT and Qian W 共1997兲. Mech. 关51兴 Mukhopadhyay NK. Am Soc shear deformation and continuity effects related to the modeling of for Composities Dayton. 关69兴 Whitcomb JD and Shivakumar KN 共1989兲. and Dattaguru B 共1999兲. Eng. debonding of laminates. Fract. 关97兴 Zou Z. West Conshohocken. Modified crack closure integral using six-noded isoparametric energy release rate components for edge-delaminated composite lami- quadrilateral singular elements. Moura MFd. Fract. Fract. mulae for 3D finite element. 关94兴 Venkatesha KS. Composite Materials: Testing and Design. Fract. Kakodkar A. Solids Struct. 371– 324 –342. Am element. Mech. Structural Dy. 50. and Castro PTd 共2002兲. Calculation of stress intensity factors for 关68兴 Roeck GD and Wahab MMA 共1995兲. Dattaguru B. Mech. 496 –507. 关95兴 Venkatesha KS. 40共6兲. Eng. 241–252. 关54兴 Buchholz FG 共1984兲. Mech. 385. Modified crack closure integral 共MCCI兲 for 3-d element analysis of an interface crack with large crack-tip contact problems using 20-noded brick elements. Strain energy release rate for. J. Soden PD. Fract. 关99兴 O’Brien TK 共2001兲. Compos. Shivakumar KN. J. 关62兴 Lim IL. A study of 关73兴 He W-J. 409– 415. 关60兴 Henshell RD and Shaw KG 共1975兲. interlaminar cracks in composite laminates. Numer. zones. 13共4兲. 64共6兲. 35共25兲. West Conshohocken. 关55兴 Buchholz FG and Meiners B 共1984兲. 34共20兲. 237–258. 38共15兲. Mech. Fracture mechanics analyses for skin-stiffener debonding. and Newman JC 共1988兲. Elementary Engineering Fracture Mechanics. Dattaguru B. CMES 3共1兲. P 关52兴 Irwin GR 共1958兲. Fract. 关81兴 Krüger R and König M 共1997兲. Numer. and Vijayakumar K 关83兴 Raju IS. Composite Materials: Fatigue and Fracture. J. Am Soc for Testing Springer Verlag. Mech. Mech. Fract. Solids Struct. 关100兴 Krüger R 共1994兲. Mech. Sci. 关98兴 Morais ABd. Int. Lisbon. Further con. perfectly-plastic crack tip elements. Finite element based computa. Mixed-mode stress intensity 关67兴 Raju IS. I. Eng. specimen. Crews JH. Improved formulae for the FE-calculation of the 关79兴 Whitcomb JD 共1990兲. nents at the tip of an interface crack under mode I loading. and Ramamurthy TS 共1996兲. R43–R50. J. and Raju IS 共2002兲. Reid SR. Dordrecht. Fatigue Fract. Ramamurthy TS. 335–339. Separation of crack extension 关65兴 Shivakumar KN. 383–396. Prediction of delamination growth 关56兴 Barsoum RS 共1976兲. 59共3兲. Ramamurthy TS. Ramamurthy TS. 17共2兲. and Dattaguru B 共1998兲. Eng. tion of strain energy release rate by modified crack closure integral. 714 –734. OH. EA Armanios 共ed兲. and Maiti SK 共1998兲. 345–365. and Ramamurthy TS 共1996兲. Fract. and Choi SK 共1991兲. 关88兴 Beuth JL and Narayan SH 共1997兲. Compos. Fract. Technical University 共eds兲. A universal optimum quarter point Technical Conference on Composite Materials. Int. 677– 691. Fract. 665– 676. EA Armanios three dimensional bodies. West Conshohocken. Patel SK. Int. Mech. Jr 共1988兲. OH. Fract. Fract. 关90兴 Qian W and Sun CT 共1997兲. 60共4兲. SLB and ENF specimens ISD-Report . Eng. Kluwer Academic Publishers. J. Three dimensional finite element analysis of multi- 关76兴 Glaessgen EH. ments. Eng. Compos. 关89兴 Huang H and Kardomateas GA 共2001兲. Venkatesha KS. 103–116. 25–37. Finite element siderations in modified crack closure integral based computation of analysis of delamination growth in a multidirectional composite ENF stress intensity factor in BEM. Sixth Volume. J. Nodal constraint. factors for interfacial cracks between two orthotropic solids. Modified crack closure integral method with 关85兴 Sun CT and Jih CJ 共1987兲. Struct. Eng. 131–140. Certain aspects related to com. PA. On the accuracy of the modified 关80兴 Smith SA and Raju IS 共1998兲. Eng. 共1994兲. Eng. Lisbon. Fract. 7–12. specimen. 36共6兲. The use of transition. Mode-I interlaminar fracture of carbon/epoxy cross-ply composites. SP method. Int. Riddell WT. 54共3兲. Struct. Mech. Proceed. Comput. Blacksburg. A three-dimensional formula for the behavior of subinterface cracks in bimaterial plates. West Conshohocken. Mode separation of 关74兴 Wang JT and Raju IS 共1996兲. A method of applying VCCT to 85–98. and Rinderknecht S 共2000兲. Bangalore. ics. Dorset. 30共2兲. 269–279. 569–580. Proc of Am Soc for Composites. Eng. Fracture I. 211–228. PA. 关96兴 Venkatesha KS. 54共2兲. Eng. 460– 478. Narayana KB. On the use of isoparametric finite elements in under cyclic loading. TL Panontin and SD Sheppard ement Computations. 59共2兲. Riddell WT. 23. Mater. and Buchholz FG Mech. model of interface fracture. AIAA J. lands. Mater. Methods Eng. and Dattaguru B 共1986兲. Nether. Technol. Compos. 54共6兲. Lin Y. 57共6兲. 关75兴 Glaessgen EH. J. Evaluation of stress-intensity factors crack closure integral method in combination with higher order finite using general finite-element models. approach. Numer. Eng. Eng. 关63兴 Narayana KB and Dattaguru B 共1996兲. 共ed兲. Part A 32共1兲. ings of the Indo-German Workshop. and Crews JH. and Li S 共2001兲. Eng. Vijayakumar K. 49共3兲. and Krishnamurthy T 共1996兲. Eng. for the analysis of delaminated composite laminates. sublaminates. Crack tip finite elements are 关84兴 Hwu C and Hu J 共1992兲. Interlaken. 23. Proceedings of the 4th World Congress on Finite Elements. Strain energy release rate formulae for energy release rate for delamination in composite laminates using skin-stiffener debond modeled with plate elements. Sci. 42共6兲. Grant and CQ Rousseau 共eds兲. Technol. 关71兴 Narayana KB. 11. rates of delaminations in composite laminates. J. PA. 215–230. formation and continuity on delamination modeling with plate ele. 16th Annual 关58兴 Murti V and Valliappan S 共1986兲. Composites. Finite element analysis of a 3D mixed-mode 关92兴 Boniface V and Simha KRY 共1999兲. 关70兴 Buchholz FG 共1994兲. 关87兴 Dattaguru B. Methods for calculating stress intensity analysis of plates with postbuckled delaminations. interfacial crack between two orthotropic solids. Garbo 共ed兲. Indian Institute of Science. using plate elements. 847– 860. 关59兴 Narayana KB. 227–231. Strain energy release rates of an elements. corner crack nodes. 637– 650. Finite element estimates of strain energy release rate compo- 关64兴 Sethuraman R and Maiti SK 共1988兲. PA. and Raju IS 共1998兲. and Ding H-J 共1997兲. Mechanics of instability-related delamination strain energy release rate by the modified crack closure integral growth. linear fracture mechanics. ASTM STP 1285. of Lisbon. Mech. ized modified crack closure integral 共GMCCI兲 and its application to sure integrals for estimation of SIF variations. 4th 关78兴 Krueger R and O’Brien TK 共2001兲. 3317–3330. 63共2兲. Berlin. Eng. 451– 463. interface crack problems. Krüger R. 145–157. Re-examination of crack opening fracture problem by virtual crack closure integral method. The use of finite extension strain energy 共eds兲. Mech. namics and Materials Conference. ASTM STP 1285. 关66兴 Raju IS. Am Soc for Testing and Materials. 162–178. 650– 659. Dattaguru B. 3719–3734. Tan PW. 39rd AIAA/ASME/ASCE/AHS/ASC Structures. 26. 945–955. and Materials. factors for cracks located at or parallel to the interface in bimaterial dimensional elastic analysis of a composite double cantilever beam half planes. 关82兴 Ferrie CH and Rousseau CQ 共2001兲. Handbuch der Physik VI. and applications 141 Composites—13th Technical Conf on Composite Materials. Composite Materials: closure technique for calculating stress intensity factors for cracked Fatigue and Fracture.

ASME/ASCE/AHS/ASC Structures. 关111兴 Rikards R. and Rao GV共2001兲. 2079– 关104兴 Sih GC. 57– 63. Ramamurthy TS. Failure of delaminated interply hy- 315–335. Intralaminar frac- Poursartip and K Street 共eds兲. Nonlinear ASME/ASCE/AHS/ASC Structures. composite joints: Testing and analysis. symmetrically delaminated single leg als Conference. J. 59共1兲. Ratcliffe J. 关107兴 Rybicki EF. Analysis of composite skin/ by fiber surface treatment. Analysis of the strength of the interface between laminates. 关102兴 Davidson BD. Numerical Methods in Fracture Mechanics. DOT/FAA/AR-97/56. Compos. ASTM STP 836. J. 43. Sci. 2. 42nd AIAA/ASME/ASCE/AHS/ analysis for composite skin-stiffener debonding. 1. analysis of single lap adhesive composie joints with delaminated ad. O’Brien TK. Technol. and Poe CC 共1999兲. Compos. tolerance analysis of a hybrid composite tapered flexbeam. P Grant and CQ Rousseau 共eds兲 Am Soc modeling of bonded joints in composite structure. 56 –79. anisotropic bodies. Analysis of interlaminar stresses in ASCE/AHS/ASC Structures. 189–203. Comparison Analyses for debonding of stitched composite sandwich structures of 2D finite element modeling assumptions with results from 3D using improved constitutive models. Res. 190–210. Mater. brid composite plates under compression. tures. Scudamore R. 关127兴 Glaessgen EH. Strength and fatigue life Practice. herends. Investigation of mixed mode I/II interlaminar ington. Three dimensional 关123兴 Johnson WS. Effect of friction on the per. 170–178. 1263–1300. O’Brien TK. Flange delamination prediction in composite structures chanics Spring Conference and Exposition. Delamination and stitch Johnson 共ed兲. fracture toughness of laminated composites by using a CTS type 关133兴 Krueger R. UK and Boston. 54共4兲. Mater. AIAA 98-2022. 关126兴 Cantwell WJ. and Hachenberg D共2002兲. 关145兴 Rinderknecht S and Kroplin B共1997兲. Struct. O’Brien TK. 325–342. FG. Mater. RV 共1998兲. Y Zhang 共ed兲. LM. Sci. Southampton. Philadelphia. Kumosa M. pos. ARL-TR-2207. Sci. PA. 146 –155. and ings of the American Helicopter Society 57th Annual Forum. Mixed. 关138兴 Flanagan G 共1994兲. approach. J. J. AIAA J. Some studies on interlaminar and 关117兴 Buchholz FG. thesis. Proceed- 关110兴 Rikards R. Technol. 57共1– 4兲. Structural Dynamics and Materi- 关119兴 Li J. fects in Composite Materials. Janab MK. Nonlinear numerical and experimental 19– 47. eling and Simulation Based Engineering. analysis of stitched stiffener-skin debonding. A 关112兴 Jose S. 175–193. Johnson 共ed兲. WS 关142兴 Glaessgen EH. 关115兴 Wang H. Ding. Swansea. Eng. Paris PC. and Valentin. Knight RC. 3D fracture analysis of delaminations in laminated composites. Mech. SN Atluri and PE 关108兴 Tanzawa Y. Vol I. Ohio. Y Zhang 共ed兲. 391– 关141兴 Glaessgen EH. West Conshohocken. Buchholz FG. 关136兴 Krueger R.and four-point multi-axial loading. Helicopter 关109兴 Rhee KY 共1997兲. MH Alı́abadı́. Mater. PA. Butkus. Mech. Testing and analysis of composite skin/stringer debonding under ceived mode II delamination toughness from three. The Tenth International Conference on Composite Materials. 54共4兲.D. Fatigue life meth- Compos. Luxmorre and DRJ Owen 共eds兲. Compos. Federal 2108 –2133. DC. delamination in composite plates. Philadelphia. Hernandez TD. Compos. Analysis of double edge-cracked analysis of skin-to-stiffener debonding in composite aerospace struc- iosipescu specimens under biaxial loads. Krishnamurthy T. Characterization of mode delamination growth in compressively loaded laminates. approach to determine GIc of fiber-reinforced laminated composites. Technol. 851– 865. O’Brien TK. Krüger R.and 3D-models of a cross-ply laminated three. 关139兴 Yap JWH. Compos. A finite element model for the single lap adhesive joint. Technol. Compos. Structural Dynamics and Materials Con- thick composite laminates with and without edge delamination. On the validity of applying the elastic work factor Soc. The 35rd AIAA/ point bending specimen. Tech Science Press. Helicopter Soc. multidirectional end-notched flexure tests. Watanabe N. 61. Three dimensional 关124兴 Goswamia S and Becker W共2001兲. nol. and Davies P共1999兲. Wang H. The University of Liverpool. and Minguet PJ共2002兲. S Kalinsky. US Department of Transportation. and applications Appl Mech Rev vol 57. Wash- Richard HA共1998兲. 893– 897. ASTM STP 876. Com- single leg bending specimens. Thomson RS. Bledzki AK. chanics Publ. Struct. Compos. Paris IL. 1592–1597. The effect of facesheet/core analysis of center delaminated unidirectional and multidirectional delamination in sandwich structures under transverse loading. Bürger M. Am Soc for Testing and Delamination and Debonding of Materials. Fract. Wacker G共1998兲. 2783–2790. Applications of analysis and resulting design recommendations for unidirectional and fracture mechanics to the durability of bonded composite joints. and König M共1995兲. Sleight DW. O’Brien TK. Korjakin A. Raju IS. WS Materials. PA. Res. Delamina. Fracture mechanics 402. Encino. TK 共1995兲. Composite Materials: Theory and 关137兴 Hoyt DM. 20共3兲. Localized Damage III. Res. J. Test and analysis of com. 共1987兲. March 2004 No 94/2. Schaff JR. Dattaguru B.142 Krueger : Virtual crack closure technique: History. Deibler JE. Aviation Administration. Am Soc for Testing and Materials. Mech. Fatigue and damage Compos. 关101兴 Davidson BD. 161–168. AR Materials (ICCM13). Compos. and DJ Cartwright 共eds兲. Strain-energy release rate analysis of cyclic 关106兴 Ramakumar RL and Whitcomb JD 共1985兲. Scott ML. Davidson BD. bending specimens. 57共1– 4兲. methodology for bonded composite skin/stringer configurations. J. ASTM STP 876. 251–258. 关103兴 Krueger R and O’Brien TK 共2000兲. Ph. A shell/3D modeling technique 关125兴 Ratcliffe JGA 共2001兲. The 关116兴 Bansal A and Kumosa M共1998兲. Tech- for Testing and Materials. als Conference. Krüger R. Compos. and Vinson SS 61共11兲. and Minguet PJ共2002兲. delamination of graphite/epoxy test specimens. NASA/TM. Compos. Sci. R 共1994兲. 61共8兲. Compos. 59. J. Effect of De- I and mixed-mode delamination growth in T300/5208 graphite/epoxy. 24共2兲. odology for tapered composite flexbeam laminates. Kumar RR. Debonding of stitched 350–357. Proc of Thirteenth Int Conf on Composite Materials University of Stuttgart. ference. 32共17兲. Interlaminar fracture toughness of GFRP influenced 关134兴 Minguet PJ and O’Brien. bend end-notched flexure tests. 1115–1122. 关131兴 Murri GB. Raju IS. Pineridge Press. The 39th AIAA/ASME/ 关118兴 Whitcomb JD and Raju IS 共1985兲. Institute for Statics and Dynamics of Aerospace Structures. Fatigue life specimen. mining strain energy release rates in composites. June with ply waviness. failure in stitched composite joints. Fract. Am. and Poe CC 共1999兲. ture toughness of a cross-ply laminate and its constituent sub- 关135兴 Minguet PJ 共1997兲. and Dobyns AL 共2001兲. Computational method for the . and Minguet PJ共2002兲. PA. S. Paris IL. Wang H. als Conference. 38共5兲. The 38rd AIAA/ 关113兴 Pieracci A. Mech. Struct. and Rikards. Long Beach. Buchholz. Raju. Am Soc for Testing and Materials. 37共1兲. 7–9. and Rousseau CQ共1998兲. frame and skin in a bonded composite fuselage panel. 关129兴 Hwang S-F and Mao C-P共2001兲. Delamination and Debonding of Materials. Interfa- 2000-210287. 关105兴 Krueger R. Structural Dynamics. 21. On the use of fracture mechanics in designing a 关144兴 Rinderknecht S and Kröplin B共1995兲. Ward SH. Korjakin A. modified DCB test for 3-D orthogonal interlocked fabric composites. Proceedings of the 40th AIAA/ 69–94. 245–252. 关114兴 Schuecker C and Davidson BD 共2000兲. and Poe CC 共1998兲. MA. Struct. Mode I and mixed mode energy release rate values for 关130兴 Singh KL. Cvitkovich MK. stringer bond failures using a strain energy release rate approach. Struct. Philadelphia. intralaminar fracture toughness of multilayered composite structures. ASC Structures. Sci. A ASME/ASCE/AHS/ASC Structures. Mater. Society for Experimental Me- 关120兴 Li J 共2000兲. J. Sci. J. Technol. IS. for the analysis of delaminated composite laminates. 381–389. Cincinnati. and Eggers H 共1990兲. 515–521. 61共3– 4兲. and Minguet PJ共2000兲. Computational Me. Technol. and Irwin GR共1965兲. and Sundararaman V共1998兲. and Mangalgiri PD 共1999兲. and Rousseau CQ共1997兲. and König M共1995兲. May 9–11. 425– 435. Adhes. Structural Dynamics and Materi- analyses of homogeneous. no 2. cial fracture in sandwich laminates. Beijing. 关121兴 Wahab MMA 共2000兲. Structural Dynamics and Materi- Carpinterı́. 29共16兲. Technol. and Ishikawa T共2001兲. FEM simulation of a O’Donoghue 共eds兲. 385–394. 关122兴 Qin M and Dzenis Y 共2001兲. 关132兴 Murri GB. Beijing. and Raju IS 共2001兲. ASTM STP 1383. 1513–1527. A general sublaminate analysis method for deter- tion analysis for 2D. CA. Technol. UK. J. 334 –344. 1097–1107. 关143兴 Glaessgen EH. Bledzki AK. 关128兴 Whitcomb JD 共1984兲. 89–100. and Compos. California. 34共15兲. Am. 关140兴 Kumar RR and Rao GV 共2001兲. mode fracture analysis of orthotropic laminates by local and global Proceedings of the Thirteenth International Conference on Composite energy methods. On cracks in rectilinearly 2085. 1528 –1559. Compos. Eng. April 20–23. Int. Fract. (ICCM13). Vol II. Buchholz FG. 24共3兲. posite hat stringer pull-off test specimens. Compos. Mod- 105–123. J. 14共6兲. and Materials Conference.

PA. ASTM STP 1110. NASA/CR- titanium-graphite hybrid laminates. Ronald Krueger was a research and teaching assistant at the Institute for Statics and Dynamics of Aerospace Structures at the University of Stuttgart from 1989 until 1996. 287–311. Proc of the 42nd AIAA/ tigue and Fracture. 1855–1883. Compos. Wu XX. Comput. 96/3. Compressive behavior of multiply ture analysis of different specimens regarding 3D and mode coupling delaminated composite laminates. Scaglione A. and numerical study of delamination caused by local buckling of tional Mechanics ’95: Theory and Application. 383– 401. 19. 关177兴 Parisch H 共1995兲. 120–121. Beijing. 1015–1021. 58共1兲. J. Post-buckling behavior growth of delaminations in composite plates subjected to axial com. ing and Materials. Lee KH. Third Volume. Modeling compres- and ENF specimens. 26. and Scaramuzzino F共2000兲. Buckling of a sublaminate structures. 36共3– 4兲. Roudolff F. Tech. 55共4兲. 1523–1535. Experimental of an angle-ply laminated three-point bending specimen. Chunchu P. Edge delamination delamination in composite skin-Stringer pull-off specimens. Eng. posite laminates with an embedded elliptical delamination at the in- Technol. AIAA J. Springer. Compu- 关159兴 Ireman T. Mech. Combined effect of matrix 关173兴 Li J and Sen JK 共2001兲. and Kröplin B 共1997兲. Fracture behavior of FRP com- interlaminar fracture toughness of composite laminates. He received both his diploma and his doctorate degree in Aerospace Engineering from the University of Stuttgart in Germany in 1989 and 1996. versity of Stuttgart. J. Technomic Publishing. ments. 19共13兲. Influence of 2D finite element nol. and Ambur D 共2002兲. and Barriol R共1998兲. in laminated composites. terface. and applications 143 analysis of delamination growth in composite plates. Experimental-numerical 关161兴 Pradhan SC and Tay TE共1998兲. Analysis of a laminate with a postbuckled em- ling delamination growth in composite laminates using plate ele. 关163兴 Whitcomb JD 共1992兲. Compos. Mater. approach. Computa. Mech. quence on energy release rate distributions in multidirectional DCB 关168兴 Riccio A. Maison S. Int. Compos- (ICCM13). 47共1– 4兲. 862– 889. J. Compos. Wisnom MR. 1803–1812. Shivakumar KN. Gädke M. Compos. Am Soc for Test. Compos. Prior to joining the NIA he worked for ICASE as a Staff Scientist in Structures and Materials from August 2000 until the end of 2002. 68共4兲. Vol. Lorrain B. composite laminates. 157–171. 155–161. Computational frac. Delamination interaction in laminated 关164兴 Shivakumar KN and Whitcomb JD共1985兲. Struct. 关162兴 Whitcomb JD 共1989兲. Struct. Proc of the Thirteenth Int Conf on Composite Materials nation growth in composite panels under compressive load. Perugini P. 59共7兲. 301–309. Compos. 2323–2328. 关155兴 Salpekar SA and O’Brien TK 共1991兲. 38. Hänsel C. and Sharpe M 共2001兲. 233–242. Krüger R. Three-dimensional finite element investigation of delamination buckling and growth. J. Part 2: Finite element analysis. 209–218. Sci. 关175兴 Krueger R and Minguet PJ 共2002兲. Int. Lancaster. and König M共1996兲. CFRP-laminates: development of a predictive tool based on compu. Earlier he held a NRC post-doctoral research position at NASA Langley Research Center. Modeling delamination growth 关150兴 Davidson BD. in a quasi-isotropic laminate.Appl Mech Rev vol 57. and Buchholz F-G共2001兲. 727–733. and Rikards R 共1995兲. AIAA J. Ronald Krueger joined the National Institute of Aerospace (NIA) as a Senior Staff Scientist in January 2003. and Jones MI共1998兲. Uni- der compression loading. Eng. tational structural analysis and testing: An approach to understand Ousset Y. March 2004 Krueger : Virtual crack closure technique: History. Kissimmee. 225–240. Fract. Struct. Beijing. Numer. Composite Materials (ICCM13). 关148兴 Reeder J. Y Zhang 共ed兲. 关174兴 Li J 共2000兲. Analysis of frame-to-skin joint pull-off tests cracking and free edge on delamination. Technol. ites. Strain energy release design in finite element analysis of post-buckled delamination in rate distribution for double cantilever beam specimens. Reston. He is working in the area of damage formation in advanced composite materials. 178 –184. Mech. The tendency for free-edge 990. and Barbera AL共1996兲. Comput. 关153兴 Dhondt G. Institute for Statics and Dynamics of Aerospace Structures. Mater. Mater. 59共2兲. Sci. 关167兴 Shen F. Eng. G Yagawa. In Germany. Delamination analysis 关172兴 Lachaud F. Trans Tech Publications. Methods Eng. AIAA J. and TA Cruse 共eds兲. König M. 78. Numer. J. ASME/ASCE/AHS/ASC SSDM Conf. modeling assumptions on debonding prediction for composite skin- 关158兴 Burianek DA and Spearing SM 共2001兲. Mech. 58共5兲. WA. Sci. Split angle-ply 73– 81. embedded delamination. Ghantae S. Fract. Composite Materials. 关151兴 Shivakumar K. Heidelberg. 34共1兲. 1239–1251. Sci. and Perugini P共2001兲. Song K. Plast. Struct. ASC SSDM Conf. and Materials Conference. Embedded delami- stress state. Postbuckling and 关165兴 Gaudenzi P. and Raju IS共1991兲. 1286 –1290. 36共7兲. Part B 32共3兲. Compos. 1004 –1023. Efficient modeling of postbuck. Hänsel C. Perugini P. 359–374. Technol. 2064 –2072. of composite panels in the presence of unstable delaminations. Seattle. 关178兴 Parisch H 共1989兲. Fracture of Composites. Scaramuzzino F. Thesken JC. Interacting damage modes in stiffener specimens subjected to tension and bending. 关176兴 Krüger R. and König M 共1996兲. ISD-Report No modeling of delamination growth in notched composite laminates un. Y Zhang 共ed兲. II. 557–569. Am Society for Composites—15th Annual Technical Conference on Compos. Switzerland. and Riccio A共2001兲. Struct. Fract. Michel L. He joined the researchers at NASA’s Mechanics of Materials Branch in 1997 where he began his investigation of debond failure in composite skin/stringer configurations. 关171兴 Suemasu H and Kumagai T共1998兲. Buchholz FG. Proc of the 38th AIAA/ASME/ASCE/AHS/ dimensional non-linear contact analysis. bedded delamination. Damage propagation in delamination growth. 181–202. 51共2兲. Methods Eng. Sharp R. thermoplastic and thermoset composites. Delamination growth in plications. and prediction of the delamination failure. The 43rd AIAA/ASME/ASCE/AHS/ASC Structures. 关166兴 Tay TE. beam specimen for measurement of fracture toughness with mode III 关169兴 Riccio A. 关147兴 Zheng S and Sun CT共1998兲. 60共2兲. 209–220. A consistent tangent stiffness matrix for three- tational fracture mechanics. 关146兴 Klug J. FL. 关154兴 Wang H. 1686 –1691. composite structural elements-coupon experiments and analyses. no 2. 603– 611. Compos. Technol. Proc of the Thirteenth Int Conf on 2002-211452. An experimental method to measure the mode-III 关170兴 Pradhan B and Chakraborty D共2000兲. 983– 关157兴 Lindemann J and Becker W共2002兲. pos. 23. VA. including contact effects. Shen F. Reinf. . 2–18. and Tay TE共2001兲. Effect of stacking se. Proc of in curved (0 4 /⫾456 ) s glass-fibre/epoxy beams loaded in bending. 关152兴 Suemasu H 共1999兲. Rinderknecht S. Eng. Compos. Philadelphia. and König M 共1996兲. 关179兴 Krüger R. Com- pression. Three-dimensional effect in the prediction of flange 关156兴 Kaczmarek K. Fract. Composite Materials: Fa. Am Institute of Aeronautics and 28. and Sun CT共1996兲. Chergui A. Sci. 62共2兲. 29. and Sciuva MD共1999兲. Three-dimensional analysis of a postbuckled 64共1– 4兲. tural Dynamics. Compos. A continuum-based shell theory for non-linear ap- 关160兴 Krüger R. focusing on the investigation of debonding in composite skin/stringer configurations using finite element analysis and fracture mechanics. Astronautics. J. Lee KH. ICASE Report No 2002-4. PA. effects. sion behavior of delaminated composite panels. Mesh 关149兴 Crews JH. Greenhalgh E. EA Armanios 共ed兲. delamination in laminates and its minimization. 61共9兲. SN Atluri. Struc. respectively.