International Journal of Plasticity 27 (2011) 217–231

Contents lists available at ScienceDirect

International Journal of Plasticity

journal homepage: www.elsevier.com/locate/ijplas

On stress-state dependent plasticity modeling: Significance of the

hydrostatic stress, the third invariant of stress deviator and the
non-associated flow rule
Xiaosheng Gao a,*, Tingting Zhang a, Jun Zhou a, Stephen M. Graham b, Matthew Hayden c,
Charles Roe c
a
Department of Mechanical Engineering, The University of Akron, Akron, OH 44325, USA
b
Department of Mechanical Engineering, United States Naval Academy, Annapolis, MD 21402, USA
c
Alloy Development and Mechanics Branch, Naval Surface Warfare Center, West Bethesda, MD 20817, USA

a r t i c l e i n f o a b s t r a c t

Article history: It has been shown that the plastic response of many materials, including some metallic
Received 19 February 2010 alloys, depends on the stress state. In this paper, we describe a plasticity model for isotro-
Received in final revised form 20 April 2010 pic materials, which is a function of the hydrostatic stress as well as the second and third
Available online 7 May 2010
invariants of the stress deviator, and present its finite element implementation, including
integration of the constitutive equations using the backward Euler method and formula-
Keywords: tion of the consistent tangent moduli. Special attention is paid for the adoption of the
Plasticity modeling
non-associated flow rule. As an application, this model is calibrated and verified for a
Stress triaxiality
Lode angle
5083 aluminum alloy. Furthermore, the Gurson–Tvergaard–Needleman porous plasticity
Modified Gurson model model, which is widely used to simulate the void growth process of ductile fracture, is
Non-associated flow rule extended to include the effects of hydrostatic stress and the third invariant of stress devi-
ator on the matrix material.
Ó 2010 Elsevier Ltd. All rights reserved.

1. Introduction

Plasticity describes the deformation of a material undergoing non-reversible changes of shape in response to applied
forces. Our ancestors in ancient times already recognized the plastic behavior of metals as they attempted to make various
tools and weapons. However, the scientific study of plasticity may justly be regarded as beginning in 1864 when Tresca pub-
lished his results on punching and extrusion experiments and formulated his famous yield criterion (Tresca, 1864). This yield
criterion was then used by Saint-Venant (1870) and Levy (1870) in their development of a theory of rigid-perfectly plastic
solid. Another well known yield criterion was proposed by von Mises (1913) on the basis of purely mathematical consider-
ations and later was interpreted by Hencky (1924) as plastic yielding occurs when the elastic shear-strain energy reaches a
critical value. Von Mises also independently proposed equations similar to Levy’s for rigid-perfectly plastic materials. Other
important contributions in the early development of the plasticity theory include the works by Prandtl (1925), Reuss (1930),
among others. Subsequently, within the scope of elastic–plastic materials under small deformation, the notation of yield in
the stress space formulation was generalized to cover work-hardening materials and a unified theory of plasticity began to
emerge after World War II (Hill, 1950; Mendelson, 1968). The flow theory is the most widely known theory of plasticity,
which consists of a yield criterion, a flow rule, a hardening law and the loading–unloading conditions. The yield criterion
determines the stress-state when yielding occurs, the flow rule describes the increment of plastic strain after yielding,

* Corresponding author. Tel.: +1 330 972 2415; fax: +1 330 972 6027.
E-mail address: xgao@uakron.edu (X. Gao).

0749-6419/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.ijplas.2010.05.004
218 X. Gao et al. / International Journal of Plasticity 27 (2011) 217–231

the hardening law characterizes the evolution of the flow stress with increased plastic deformation, and the loading–unload-
ing conditions specify if the stress path moves outward from the current yield surface, inward from the current surface or
along the current surface.
The most popular continuum plasticity model is the so-called J2-flow theory. This theory assumes hydrostatic stress as
well as the third invariant of the stress deviator has no effect on plastic yielding and the flow stress and has been widely
employed to describe the plastic response of metals. However, increasing experimental evidences show that this assumption
is invalid for many materials. Inspired by the extensive experimental results reported by Spitzig et al. (1975, 1976) on the
behavior of high-strength metals undergoing uniaxial tension and compression, Brunig (1999) presented an I1–J2 yield cri-
terion, which is similar to the Drucker–Prager yield condition in soil mechanics (Drucker and Prager, 1952), to describe the
effect of the hydrostatic stress on the plastic flow properties of metals. Later, the I1–J2 plasticity model was applied by Brunig
et al. (2008) to study the effect of stress triaxiality on the onset and evolution of damage in ductile metals. In Brunig et al.
(2000), the authors added a J3 term in the yield function to study the deformation and localization behavior of hydrostatic
stress sensitive metals. Kuroda (2004) presented a phenomenological plasticity model accounting for hydrostatic stress-sen-
sitivity and vertex-type of effect. Hu and Wang (2005) proposed a stress-state dependent yield criterion for isotropic ductile
materials. Cazacu and Barlat (2003) and Soare et al. (2007) extended Drucker’s J2–J3 yield function (Drucker, 1949) to include
plastic anisotropy and applied it to simulate sheet forming. Plunkett et al. (2008) and Cazacu et al. (2010) extended the iso-
tropic yield function proposed by Cazacu et al. (2006) to account for the anisotropic plastic response of textured metals. Bai
and Wierzbicki (2008) discussed a pressure and Lode dependent metal plasticity model and its application in failure analysis.
Gao et al. (2009) noticed the plastic response of a 5083 aluminum alloy is stress-state dependent and were forced to use
different stress–strain curves to analyze specimens experiencing different stress states. In a most recent study, Mirone
and Corallo (2010) found that, for the metals they tested, the hydrostatic stress plays a significant role in accelerating failure
but has negligible effect on the stress-plastic strain relationship, while the Lode angle has a considerable effect in modifying
the stress–strain curves but does not significantly affect the failure strains. These findings agree with the findings by Gao
et al. (2009).
A common approach in the metal plasticity theory is the adoption of the so-called associated flow rule or normality con-
dition, requiring the yield function and the flow potential to be identical. Although the normality condition is a consequence
of the postulate of maximum plastic dissipation, many studies, such as Mroz (1963), Nemat-Nasser and Shokooh (1980), Ne-
mat-Nasser (1983, 1992), Runesson and Mroz (1989), Brunig et al. (2000), Stoughton (2002) and Stoughton and Yoon (2004,
2006), indicate that appropriate constitutive description of many materials can be achieved by using the less restrictive non-
associated flow rule. Brunig (1999) and Brunig et al. (2000) demonstrated that pressure sensitive yielding and non-associ-
ated flow rule remarkably influence the onset of localization and the subsequent localization behavior. To model the ductile
fracture process in solids, Ma and Kishimoto (1998) proposed a non-associated flow rule to characterize the yield and plastic
deformation of void-containing materials, where the yield function takes the form of the Gurson–Tvergaard–Needleman por-
ous plasticity model (Gurson, 1977; Tvergaard, 1981, 1982; Tvergaard and Needleman, 1984) while the flow potential takes a
slightly different form. In a recent publication, Cvitanic et al. (2008) presented detailed finite element formulations for sheet
metal forming based on non-associated plasticity.
In this paper, we describe a plasticity model for isotropic materials, which is dependent on the second and third invariants
of the stress deviator as well as the hydrostatic stress, and present its finite element implementation including integration of
the constitutive equations using the backward Euler method and the formulation of the consistent tangent moduli. The der-
ivation will be based on small-strain formulation. For finite strain plasticity, kinematic transformations are performed first so
that the constitutive equations governing finite deformation are formulated using strains–stresses and their rates defined on
an unrotated frame of reference. Once the kinematic transformations have eliminated rotation effects on rates of the tenso-
rial quantities, the stress updating procedure and the consistent tangent stiffness formulation remain the same as those for
small-strain formulation. Most commercial finite element programs adopt this kind of treatment for finite strain plasticity
thus only small-strain formulation needs to be considered in development of a user material subroutine. As an application,
the proposed plasticity model is calibrated and verified for a 5083 aluminum alloy. Furthermore, we extend the Gurson–
Tvergaard–Needleman porous plasticity model to include the effects of the stress state and present a few numerical exam-
ples. The modified porous plasticity model is expected to improve accuracy in predicting ductile fracture process of certain
materials.

2. Plasticity modeling

2.1. Influence of I1 and J3

Let rij be the stress tensor and r1, r2 and r3 be the principal stress values. The hydrostatic stress (or mean stress) can be
expressed as

1 1 1
rh ¼ I1 ¼ rii ¼ ðr1 þ r2 þ r3 Þ ð1Þ
3 3 3
 X. Gao et al. / International Journal of Plasticity 27 (2011) 217–231 219

where I1 represents the first invariant of the stress tensor and the summation convention is adopted for repeated indices. Let
r0ij be the stress deviator tensor and r01 ; r02 and r03 be its principal values, i.e.,
r0ij ¼ rij  rh dij ð2Þ

where dij represents the Kronecker delta. It is obvious that the first invariant of the stress deviator tensor is zero. The second
and third invariants of the stress deviator tensor are defined as
1 0 0   1
J2 ¼ r r ¼  r01 r02 þ r02 r03 þ r03 r01 ¼ ½ðr1  r2 Þ2 þ ðr2  r3 Þ2 þ ðr3  r1 Þ2 
2 ij ji 6 ð3Þ
  1
0 0 0 0 0 0 0
J 3 ¼ det rij ¼ rij rjk rki ¼ r1 r2 r3
3
For isotropic materials, the plastic behavior is often described by the stress invariants I1, J2 and J3 and consequently, the
general forms of the yield function (F) and the flow potential (G) are expressed as functions of I1, J2 and J3. Eq. (4) describes
the yield condition
FðI1 ; J 2 ; J 3 Þ  r
 ¼0 ð4Þ
where r
 is the hardening parameter. When material deforms plastically, the inelastic part of the deformation is defined by
the flow rule
@GðI1 ; J 2 ; J 3 Þ
e_ pij ¼ k_ ð5Þ
@ rij
where e_ pij are the rates of the plastic strain components and k_ is a positive scalar called the plastic multiplier.
A simple form of the yield function is given as follows
 1=6
F ¼ c1 a1 I61 þ 27J 32 þ b1 J23 ð6Þ

where a1, b1 and c1 are material constants. The yield function defined by Eq. (6) is a first order homogeneous function of
stress, from which an equivalent stress can be defined as
re ¼ F ð7Þ
The constant c1 can be found by substituting the uniaxial condition into Eq. (6), thus
 1=6
4
c1 ¼ a1 þ b1 þ 1 ð8Þ
729
When the material is subjected to a uniaxial stress r, the value of c1 given by Eq. (8) ensures re = r.
The flow potential takes a similar form, i.e.,
 1=6
G ¼ c2 a2 I61 þ 27J 32 þ b2 J 23 ð9Þ

where c2 = (a2 + 4b2/729 + 1)1/6 .

If the flow potential and the yield function are identical, i.e., F = G, Eq. (5) becomes the so-called associated flow rule. Fur-
thermore, if a1 = b1 = a2 = b2 = 0, the plasticity model degenerates to the formulation of classical J2-flow theory and re defined
by Eq. (7) becomes the von Mises equivalent stress.
The hardening parameter depends on the strain history. By enforcing the equivalence of plastic work, i.e.,

r e_ p ¼ rij e_ pij ð10Þ

the equivalent plastic strain increment can be defined as

e_ p ¼ rij e_ pij =r

 ð11Þ
R
 ðep Þ, where ep ¼ e_ p dt.
Therefore, the hardening behavior can be described by a stress vs. plastic strain relation r
Since the flow potential is taken to be a first order homogeneous function of stress, Euler’s homogeneous function theo-
rem results in
@G
rij e_ pij ¼ k_ rij _
¼ kG ð12Þ
@ rij
From (10) and (12), the plastic multiplier and the equivalent plastic strain rate can be related through
r F
k_ ¼ e_ p ¼ e_ p ð13Þ
G G
If the material follows the associated flow rule, F = G and k_ ¼ e_ p .
220 X. Gao et al. / International Journal of Plasticity 27 (2011) 217–231

Because of the I1 term in the flow potential G, the plastic response becomes dilatant, with the rate of volume change given
by
@G   F
e_ pkk ¼ k_ ¼ k_ 3c62 a2 I51 =G5 ¼ 3c62 a2 I51 e_ p 6 ð14Þ
@ rkk G

2.2. Numerical implementation of the I1–J2–J3 plasticity model

Following the procedures of Aravas (1987) and Kim and Gao (2005), we implement the I1–J2–J3 plasticity model outlined
in the previous section into ABAQUS (SIMULIA, 2008) via a user defined subroutine.

2.2.1. Stress update

We adopt the backward Euler method to integrate the rate constitutive equations. For the strain driven integration algo-
rithm, the stresses and state variables are known at the start of each increment and their values need to be updated at the
end of the increment corresponding to the total strain increment. The elasticity equations give
 tþDt n  o
ðrij ÞtþDt ¼ C eijkl eekl ¼ C eijkl eekl t þ Dekl  Depkl ¼ rTij  C eijkl Depkl ð15Þ

where
n  o
rTij ¼ C eijkl eekl t þ Dekl ð16Þ

is the elastic predictor, t represents the time at the start of the increment, t + Dt represents the time at the end of the incre-
ment, and the superscripts e and p denote elastic and plastic components, respectively. The total strain increment Dekl is
known, and if the linear elastic behavior is isotropic, the elastic moduli C eijkl can be expressed as
 
2
C eijkl ¼ Gðdik djl þ dil djk Þ þ K  G dij dkl ð17Þ
3
where G and K are the elastic shear and bulk moduli, respectively. Therefore, to update stresses, the plastic strain increments
need to be determined. The following outlines the procedure for computing Depij .
The yield condition and the flow rule are written as
   
F I1tþDt ; J 2tþDt ; J 3tþDt  r
 eptþDt ¼ 0 ð18Þ

and
 tþDt
@GðI1 ; J 2 ; J 3 Þ
Depij ¼ Dk ð19Þ
@ rij
Eqs. (18) and (19), when considered together with Eq. (15), result in 10 equations for Dk and nine components of Depij ,
among which seven equations are independent due to the symmetry of Depij . If the state variable ep is updated and thus
r ðeptþDt Þ is known, these equations can be solved iteratively for Depij and Dk using the Newton–Raphson method. The iterative
process follows these steps: (1) assume Depij ¼ 0 and use Eq. (15) to estimate rijtþDt ; (2) use Eqs. (18) and (19) to solve for Depij ;
(3) update stresses using Eq. (15); (4) repeat steps (2)–(3) until convergence conditions are satisfied.
To update ep , consider the hardening law and the evolution equation for ep
 
r ¼ r eptþDt ¼ r ðept þ Dep Þ ð20Þ

and

r Dep ¼ rijtþDt Depij ð21Þ

p
 
At each iteration of De ij ; D
p
e  eptþDt can be obtained by solving Eqs. (20) and (21) iteratively using the Newton–Raph-
and r
son method.

2.2.2. The consistent tangent moduli

Simo and Taylor (1985) showed that use of the consistent tangent moduli significantly improves the convergence char-
acteristics of the overall equilibrium iterations. The consistent tangent stiffness corresponding to the backward Euler inte-
gration can be obtained by linearization of Eq. (15).
 tþDt
@ rij
J ijkl ¼ ð22Þ
@ ekl
Since all quantities in calculating Jijkl are referred to time t + Dt, the superscript t + Dt will be dropped from hereafter.
Eq. (15) can be rewritten as
 X. Gao et al. / International Journal of Plasticity 27 (2011) 217–231 221

  n   o
rij ¼ C eijkl eekl ¼ C eijkl ekl  epkl ¼ C eijkl ekl  epkl t  Depkl ð23Þ

Differentiating (23) results in

 
@ rij ¼ C eijkl @ ekl  C eijkl @ Depkl ð24Þ
 
The relationship between orij and @ Depkl can be found as follows.
Let h be the hardening modulus, i.e., h ¼ Dr
 =Dep , then Eq. (21) can be rewritten as

hrij Depij
Dr
¼ ð25Þ
r
By differentiating (25) and simplifying the resulted relation, the following equation can be obtained

 Depij
hr hr
 rij  
p
@r
¼ @ rij þ p @ Deij ð26Þ
r 2 þ hrmn Depmn r þ hrmn Demn
 2

Differentiating Eq. (18) and combining the result with Eq. (26) give
!
hr rij   @FðI1 ; J 2 ; J 3 Þ  Depij
hr
p
p @ Deij ¼  2 @ rij ð27Þ
r þ hrmn Demn
2 @ rij r þ hrmn Depmn
Eliminating Dk from the nine equations given by (19) results in eight equations, such as
   
@GðI1 ; J 2 ; J 3 Þ @GðI1 ; J 2 ; J3 Þ
Dep11  Dep22 ¼ 0
@ r22 @ r11
   
@GðI1 ; J 2 ; J 3 Þ @GðI1 ; J 2 ; J3 Þ ð28Þ
Dep21  Dep22 ¼ 0
@ r22 @ r21
..
.
Differentiating (28) leads to
    !
@G   @G   @2G @2G
@ Dep11  @ Dep22 ¼ Dep22  Dep11 @ rij
@ r22 @ r11 @ r11 @ rij @ r22 @ rij
    !
@G   @G   @2G @2G ð29Þ
@ Dep21  @ Dep22 ¼ Dep22  Dep21 @ rij
@ r22 @ r21 @ r21 @ rij @ r22 @ rij
..
.
 
Eqs. (27) and (29) provide nine equations between @ Depij and orij, which can be summarized as

K@ðDep Þ ¼ D@ r ð30Þ
 T
where Dep ¼ Dep11 ; Dep21 ; Dep31 ;    ; DeP33 ; r ¼ fr11 ; r21 ; r31 ;    ; r33 gT , K is the coefficient matrix of @ðDep Þ and D is the coef-
ficient matrix of or.
From Eqs. (24) and (30) we can obtain

@ðDep Þ ¼ ðK þ DCe Þ1 DCe ð@ eÞ ð31Þ

where Ce is a 9  9 matrix representing the elasticity tensor C eijkl . Finally, the consistent tangent matrix can be obtained by
substituting (31) into (24)

J ¼ @ r=@ e ¼ Ce  Ce ðK þ DCe Þ1 DCe ð32Þ

2.3. A modified Gurson–Tvergaard–Needleman model

Ductile fracture of many structural materials is a result of void nucleation, growth and coalescence. The constitutive
description of this mechanism has received a great deal of attention in the past 30 years, which leads to various forms of
porous material models being developed to describe void growth and the associated macroscopic softening. One of the most
famous porous plasticity models was due to Gurson (1977) with modifications by Tvergaard and Needleman (Tvergaard,
1981, 1982; Tvergaard and Needleman, 1984). In the Gurson–Tvergaard–Needleman model, an extra internal variable, the
void volume fraction (f), is introduced to capture the growth of cavities and its effect on material behavior. It is important
to notice that the Gurson–Tvergaard–Needleman model reduces to that for J2-flow theory of plasticity with isotropic hard-
ening in the absence of voids (f = 0).
222 X. Gao et al. / International Journal of Plasticity 27 (2011) 217–231

Within the framework of the I1–J2–J3 plasticity theory outlined in Section 2.1, the yield function and flow potential of the
Gurson–Tvergaard–Needleman model can be modified as
 2  
F q I1
U¼ þ 2q1 f cosh 2  1  q21 f 2 ¼ 0 ð33Þ
r 2r


and
 2  
G q I1
W¼ þ 2q1 f cosh 2  1  q21 f 2 ¼ 0 ð34Þ
r 2r
 1=6  1=6
where F and G take the forms of Eqs. (6) and (9), i.e., F ¼ c1 a1 I61 þ 27J 32 þ b1 J 23 and G ¼ c2 a2 I61 þ 27J 32 þ b2 J 23 , I1, J2 and
J3 are invariants of the macroscopic stress, f is the current void volume fraction, r  is the current yield stress of the matrix
material, and q1 and q2 are parameters introduced by Tvergaard (1981, 1982) to account for void interaction and matrix
strain hardening. If a1 = b1 = a2 = b2 = 0, Eqs. (33) and (34) degenerate to the original Gurson–Tvergaard–Needleman model.
Porous material models contain an additional state variable, f. For the modified Gurson–Tvergaard–Needleman model, the
evolution equation for f can be obtained by considering the rate of the net volume change
F
f_ ¼ ð1  f Þ e_ pkk  3c62 a2 I51 e_ p 6 ð35Þ
G
By enforcing equality between the rates of macroscopic plastic work and the matrix plastic dissipation, the matrix yield
 , and the matrix plastic strain rate, e_ p , are coupled through
stress, r
 e_ p ¼ rij e_ pij
ð1  f Þr ð36Þ

where the matrix material follows a prescribed hardening function r

 ðep Þ.
Numerical implementation of the modified Gurson–Tvergaard–Needleman model follows a similar procedure as outlined
in Section 2.2.

3. Numerical examples

3.1. Modeling the plastic behavior of an aluminum 5083 alloy

In this section we apply the I1–J2–J3 plasticity model to study the plastic response of an aluminum alloy 5083-H116,
which was cold worked to achieve its strength. All specimens are machined from a 25 mm thick plate, with tensile axes ori-
ented transversely to the rolling direction. Round specimens were turned with low stress machining procedures and rectan-
gular specimens were electro-discharge machined. All tests are performed at room temperature and are considered to be
quasi-static. The test matrix includes smooth and notched round tensile bars, grooved plane strain specimens and the Lind-
holm-type specimen (Lindholm et al., 1980) subjected to different tension–torsion ratios. Fig. 1 shows a smooth round bar, a
notched round bar, a grooved plane strain specimen and a torsion specimen.
 pffiffiffi  pffiffiffi  
Two non-dimensional parameters defined as T ¼ I1 = 3 3J 1=2 2 and n ¼ 3 3J 3 = 2J 3=2
2 , together with the equivalent stress
re, are often used to characterize the stress state, where T is the stress triaxiality ratio and n can be related to the Lode param-
eter (Wierzbicki and Xue, 2005; Xue, 2007; Kim et al., 2004, 2007; Gao et al., 2005, 2009, 2010). For the smooth and notched
round tensile bars, the n-value at the center of the specimen is one while the T-value varies with the notch radius. The smal-
ler the notch radius is, the higher the stress triaxiality. On the other hand, the n-value at the center of the grooved plane

Fig. 1. Sketches of a smooth round bar, a notched round bar, a grooved plane strain specimen and a torsion specimen.
 X. Gao et al. / International Journal of Plasticity 27 (2011) 217–231 223

strain specimens is zero and the T-value varies with the groove radius. The Lindholm-type specimen can achieve various
combinations of T and n by varying the ratio between the applied tensile displacement and the applied twist angle.
The diameter of the gage section of the smooth round bar is 6.35 mm. All the notched round bars have the same diameters
of 15.2 mm in the smooth sections and 7.6 mm at their notched cross section. Three notch radii, 1.27, 2.54 and 6.35 mm, are
considered for specimens D, B and E, respectively (Table 1). The overall dimensions for the circular-grooved plane strain
specimens are 203.2  31.8  6.1 mm and the thinnest cross section is 2 mm. Three different groove radii, 2.03, 5.08 and
16.26 mm, are considered for specimens F, G and H, respectively (Table 2). The torsion specimens are hollow cylinders hav-
ing an inner diameter of 13.1 mm and outer diameter of 23.8 mm. The gage section has a length of 2.54 mm and wall thick-
ness of 0.75 mm. The torsion–tension specimens have a slightly larger outer diameter at both ends (25.4 mm) and other
dimensions are the same as the torsion specimen. Table 3 lists the ratio between the applied tensile displacement and ap-
plied twist angle for each of the tension–torsion specimens. Details of specimen drawings are given in Gao et al. (2009).
The specimens in the test matrix of this study generate a wide range of stress states. Moreover, the stress-state of a mate-
 pffiffiffi 
rial point in the test specimens evolves as plastic deformation increases. Figs. 2–5 show the variation of T ¼ I1 = 3 3J 1=2
2 and
pffiffiffi  3=2 
p
n ¼ 3 3J 3 = 2J 2 with loading history (measured by the equivalent plastic strain, e ) in the element at the specimen center
for the smooth round bar, the E-notch round bar, the G-groove plane strain specimen and the TT-16 torsion–tension spec-
imen. For the smooth round bar, n remains at 1 during the entire loading history while T increases from 1/3 to about 0.45
before specimen fractures. For the E-notch specimen, n remains at about 1 and T increases from 0.71 to 0.8. For the G-groove
specimen, n quickly decreases to 0 and remains at this level as plastic deformation increases while T increases from 0.51 to
0.8 before failure occurs. For the tension–torsion specimen TT-16, n increases from 0.4 to 0.94 and T increases from 0.1 to
0.26. For the notched and grooved specimens, changing notch (groove) radius changes the level of T in the specimen. Con-
sidering the entire loading history of each specimen and the three different notch (groove) radii, the range of T experienced
by the center element is 0.71 6 T 6 1.6 for the notched round bar tests and 0.46 6 T 6 0.97 for the plane strain tests. For the
tension–torsion tests, varying the ratio between the applied tensile displacement and the applied twist angle results in dif-
ferent combinations of the n and T histories. Considering the entire loading history and the different tensile displacement/

Table 1
Notch radii of the notched round bars.

Specimen B-notch D-notch E-notch

Notch radius (mm) 2.54 1.27 6.35

Table 2
Groove radii of the plane strain specimens.

Specimen F-groove G-groove H-groove

Groove radius (mm) 2.03 5.08 16.26

Table 3
Ratios of the applied tensile displacement and applied twist angle used in the tension–torsion tests.

Specimen TT-17 TT-19 TT-13 TT-16 TT-15 TT-14

tensile displacement/twist angle (mm/radian) 0.23 0.69 0.92 1.16 1.38 2.54

a 0.6 b 1.5

0.4 1
T ξ
0.2 0.5

0 0
0 0.1 0.2 0.3 0 0.1 0.2 0.3
ε p
ε p

 pffiffiffi  pffiffiffi  
Fig. 2. Variation of T ¼ I1 = 3 3J1=2
2 and n ¼ 3 3J3 = 2J 3=2
2 with plastic deformation in the center element of the smooth round bar.
224 X. Gao et al. / International Journal of Plasticity 27 (2011) 217–231

a 1 b 1.5
0.8
1
0.6
T ξ
0.4
0.5
0.2

0 0
0 0.05 0.1 0.15 0.2 0 0.05 0.1 0.15 0.2
ε p
ε p

 pffiffiffi  pffiffiffi  
Fig. 3. Variation of T ¼ I1 = 3 3J 1=2
2 and n ¼ 3 3J3 = 2J3=2
2 with plastic deformation in the center element of the E-notch specimen.

a 1
b 0.6

0.8
0.3
0.6
T ξ 0
0.4

0.2 -0.3

0 -0.6
0 0.05 0.1 0.15 0.2 0 0.05 0.1 0.15 0.2
ε p
ε p

 pffiffiffi  pffiffiffi  3=2 

1=2
Fig. 4. Variation of T ¼ I1 = 3 3J 2 and n ¼ 3 3J 3 = 2J2 with plastic deformation in the center element of the G-groove specimen.

a 0.4 b 1

0.8
0.3
0.6
T 0.2 ξ
0.4
0.1 0.2

0 0
0 0.05 0.1 0.15 0 0.05 0.1 0.15
ε p
ε p

 pffiffiffi  pffiffiffi  
Fig. 5. Variation of T ¼ I1 = 3 3J 1=2
2 and n ¼ 3 3J3 = 2J3=2
2 with plastic deformation in the center element of the tension–torsion specimen TT-16.

twist angle ratios listed in Table 3, the range of T and n experienced by the tension–torsion specimens in the test matrix are
in the range of 0.018 6 T 6 0.365 and 0.08 6 n 6 0.99.
Numerical analyses of the specimens are carried out using the general purpose finite element software ABAQUS (SIMULIA,
2008), which employs an updated Lagrangian formulation to handle finite deformation. The material models outlined in pre-
vious sections are implemented in ABAQUS via user defined subroutines. For round tensile bars, axisymmetric conditions are
considered and the 4-node, axisymmetric solid elements with reduced integration (CAX4R) are used. For torsion–tension
specimens, an additional degree of freedom needs to be added to the axisymmetric element to handle the twist and the
CGAX4R element in ABAQUS is developed for this purpose. For grooved plane strain specimens, the 3D 8-node brick elements
with reduced integration (C3D8R) are used. Usually the symmetry conditions allows for only 1/4 or 1/8 of the specimen
being modeled. Fig. 6 shows two typical finite element meshes. A typical axisymmetric model has 700 elements and a typical
1/8-symmetric 3D model has 20,000 elements. Since reduced integration is used, the hourglass control is employed by using
*HOURGLASS STIFFNESS in the ABAQUS input files to provide increased resistance to hourglassing.
Fig. 7 shows the true (von Mises) stress vs. plastic strain curves obtained using the smooth tensile bar data and the pure
torsion test data. Gao et al. (2009) provides the details of how these two curves were obtained from the measured load–dis-
placement and torque–twist angle curves, respectively. The two curves clearly show significant difference, suggesting that
the stress-state has a strong effect on the plastic response of this material.
 X. Gao et al. / International Journal of Plasticity 27 (2011) 217–231 225

Fig. 6. Typical finite element meshes: (a) an axisymmetric model for a notched round bar, (b) a 1/8-symmetric model for a grooved plane strain specimen.

Fig. 7. Comparison of the true (von Mises) stress vs. true plastic strain curves obtained using the smooth tensile bar data and the torsion test data.

Now consider the I1–J2–J3 plasticity model presented in previous sections and conduct finite element analyses of these
two specimens. The uniaxial tension stress–strain curve can be described by a power-law hardening relation

e ¼ rE when r  r0
 1=N ð37Þ
r0 r
e¼ E r0 when r > r0

with E = 68.4 GPa, r0 = 198.6 MPa, m = 0.3, and N = 0.155. The plasticity model requires a relation of the current yield stress as
a function of the effective plastic strain. This is achieved by using a UHARD subroutine. Using the uniaxial tension stress–
strain curve and adjusting material parameters (a1, b1) and (a2, b2), it is found that a set of parameters, a1 = a2 = 0,
b1 = 60.75 and b2 = 25, result in a best match between the numerically predicted and experimentally measured torque
vs. twist angle response of the pure torsion test. This indicates no I1 effect but significant J3 effect on the plastic response.
Fig. 8 compares the numerical predictions and the experimental measurements for the smooth tensile specimen and the
pure torsion specimen, respectively. The tensile tests have seven specimens and the torsion tests have 10 specimens. The
experimental data are represented by thinner lines while the finite element result is represented by a thicker line. In both
cases, excellent comparison between the numerical prediction and the experimental measurements is observed.
With the parameters of the plasticity model being calibrated, the next question needs to be answered is whether this
model correctly predicts the material response under complex stress states. To this end, notched round bars with different
226 X. Gao et al. / International Journal of Plasticity 27 (2011) 217–231

a 12
b 75

10

Torque (kN mm)

8 50
Load (kN)

6
Experiments
4 25 Experiments
FEA
2
FEA
0 0
0 2 4 6 0 0.1 0.2 0.3 0.4
Displacement (mm) Twist Angle (radian)
Fig. 8. Comparison of the numerical and experimental load vs. displacement curves for (a) the smooth tensile specimen and (b) the pure torsion specimen.

notch radii, plane strain specimens with different groove radii and torsion specimens subjected to different tension–torsion
ratios are tested and analyzed and the numerical predictions are compared with the experimental records. For notched
round bars and grooved plane strain specimens, the applied tensile force vs. extensometer gage displacement response is
monitored. For tension–torsion specimens, both applied axial force vs. axial displacement and applied torque vs. twist angle
responses are monitored. Fig. 9 shows typical comparisons – again, model predictions agree with experimental measure-
ments very well. Similar comparisons between experimental results and numerical predictions are observed for all other
specimens in the test matrix.
To further justify the proposed plasticity model, Fig. 10a and b compares the numerical predictions using the classical J2-
flow theory and the proposed I1–J2–J3 model for the tension–torsion test TT-16 with experimental results. In these figures,
we also included the comparison between the non-associated flow rule (Non-AFR) and the associated flow rule (AFR). As can
be seen from the plots, only the proposed I1–J2–J3 model with the calibrated parameters (non-associated flow rule) leads to
good agreement with experimental records for both torque vs. twist angle and axial force vs. axial displacement. The asso-

a 20 b 30
25
15
Load (kN)
Load (kN)

20
10 15 Experiments
Experiments
10
5 FEA
FEA
5
0 0
0 0.2 0.4 0.6 0.8 1 0 0.1 0.2 0.3 0.4
Displacement (mm) Displacement (mm)

c d
50 6
Torque (kN·mm)

40
Axial Force (kN)

4
30

20
2
10

0 0
0 0.05 0.1 0.15 0 0.05 0.1 0.15
Twist Angle (radian) Axial Displacement (mm)
Fig. 9. Comparisons of the numerical predictions and experimental records: (a) notched round bar (E-Notch); (b) plane strain specimen (G-Groove); (c)
torque vs. twist angle response for torsion–tension test (TT-16); (d) axial force vs. axial displacement response for torsion–tension test (TT-16).
 X. Gao et al. / International Journal of Plasticity 27 (2011) 217–231 227

a b 6
40

Axial Force (kN)

Torque (kN·mm) 30 4

20
2
10

0 0
0 0.05 0.1 0.15 0 0.05 0.1 0.15
Twist Angle (radian) Axial Displacement (mm)

Fig. 10. Comparisons of the numerical predictions using the classical J2-flow theory and the proposed I1–J2–J3 model for the tension–torsion test TT-16 with
experimental results: (a) torque vs. twist angle; (b) axial force vs. axial displacement.

ciated flow rule gives acceptable torque vs. twist angle prediction but unsatisfying axial force vs. axial displacement
prediction.
Fig. 11 shows the difference between the locus of the plastic flow potential (G) and the locus of the yield function (F) given
by the I1–J2–J3 model with a1 = a2 = 0, b1 = 60.75 and b2 = 25 on the r1–r2 plane, where the locus of the flow potential is
represented by the solid line and the locus of yield function is represented by the dashed line.
In summary, the plasticity behavior of the 5083 aluminum alloy considered in this study depends on the stress state and
can be described by an I1–J2–J3 plasticity model described in Section 2. The calibrated model constants are a1 = a2 = 0,
b1 = 60.75 and b2 = 25, indicating no I1 effect but significant J3 effect on the plastic response of this material.

3.2. Numerical examples using the Gurson–Tvergaard–Needleman model

In this section we conduct a series of parametric studies to illustrate the effect of the modified Gurson–Tvergaard–Nee-
dleman model on predicted material response. Fig. 12 shows a cubic element with dimensions of D0  D0  D0. Displacement
boundary conditions are imposed on the element surfaces such that the macroscopic stress ratios, q1 = r11/r22 and q2 = r33/
r22, are kept constants during the entire deformation history (which results in a constant stress triaxiality ratio and Lode
angle). Faleskog et al. (1998) and Kim et al. (2004) provide details of how to prescribe this kind of boundary conditions.
The material parameters used in the analyses are E = 68.4 GPa, t = 0.3, r0 = 207 MPa, N = 0.125 and f0 = 0.003, where E, t,
r0, n and f0 represent the Young’s modulus, the Poisson’s ratio, the yield stress, the strain hardening exponent and the initial
void volume fraction, respectively. The q1 and q2 parameters in the Gurson–Tvergaard–Needleman model are taken as
q1 = 1.5 and q2 = 1.

Fig. 11. Comparison between the locus of the plastic flow potential (G) and the locus of the yield function (F) given by the I1–J2–J3 model with a1 = a2 = 0,
b1 = 60.75 and b2 = 25 on the r1–r2 plane.
228 X. Gao et al. / International Journal of Plasticity 27 (2011) 217–231

u 2 ; σ 22

u1 ; σ 11
u 3 ;σ 33

Fig. 12. A cubic element.

For the first set of analyses, the boundary conditions are imposed such that q1 = 0.268 and q2 = 0.634, corresponding to a
stress triaxiality ratio of 1 and Lode angle of 0°. When a1 = a2 = b1 = b2 = 0, the analysis results using our user subroutine are
the same as those obtained using the original Gurson–Tvergaard–Needleman model implemented in ABAQUS, serving as a
check of our numerical implementation.
Fig. 13 illustrates the effect of I1, where the dotted lines represent the results obtained using a1 = a2 = b1 = b2 = 0 and the
solid lines represent the results obtained using a1 = a2 = 6  104 and b1 = b2 = 0. At the same applied displacement in the x2-
direction (u2), both r22 and the void growth rate are lower when the effect of I1 is taken into account.

a b
σ 22 / σ 0 f
3 0.03

2 0.02

a1 = a2 = b1 = b2 = 0
1 a1 = a2 = 6×10-4 0.01
b1 = b2 = 0

0 0
0 0.1 0.2 0.3 0.4 0.5 0 0.1 0.2 0.3 0.4 0.5
u 2 / D0 u2 / D 0
Fig. 13. Comparison of the r22/r0 vs. u2/D0 response and f vs. u2/D0 response predicted using a1 = a2 = b1 = b2 = 0 (dotted lines) and a1 = a2 = 6  104 and
b1 = b2 = 0 (solid lines).

a b
σ 22 / σ 0 f
3 0.02

0.01
a1 = a2 = b1 = b2 = 0
1 a1 = a2 = 0
b1 = b2 = -60.75

0 0
0 0.1 0.2 0.3 0.4 0.5 0 0.1 0.2 0.3 0.4 0.5
u 2 / D0 u 2 / D0
Fig. 14. Comparison of the r22/r0 vs. u2/D0 response and f vs. u2/D0 response predicted using a1 = a2 = b1 = b2 = 0 (dotted lines) and a1 = a2 = 0 and
b1 = b2 = 60.75 (solid lines).
 X. Gao et al. / International Journal of Plasticity 27 (2011) 217–231 229

a b
σ 22 / σ 0 f
3 0.03

2 0.02

a1 = a2 = 6×10-4
b1 = b2 = 0
1 0.01
a1 = 6×10-4
a2 = b1 = b2 = 0

0 0
0 0.1 0.2 0.3 0.4 0.5 0 0.1 0.2 0.3 0.4 0.5
u 2 / D0 u 2 / D0
Fig. 15. Comparison of the r22/r0 vs. u2/D0 response and f vs. u2/D0 response predicted using a1 = a2 = 6  104 and b1 = b2 = 0 (associated flow rule; dotted
lines) and a1 = 6  104 and a2 = b1 = b2 = 0 (non-associated flow rule; solid lines).

Fig. 14 demonstrates the effect of J3 by setting non-zero values for b1 and b2. Here the dotted lines show the results of
a1 = a2 = b1 = b2 = 0 and the solid lines are obtained by using a1 = a2 = 0 and b1 = b2 = 60.75. The analysis results show that
negative values of b1 and b2 lead to lower value of r22 and slower void growth rate.
The results shown in Figs. 13 and 14 are obtained using the associated flow rule. Figs. 15 and 16 compare the results of the
non-associated flow rule with those of the associated flow rule. As is shown in Fig. 15, where the dotted lines are the results
of an associated flow rule (a1 = a2 = 6  104 and b1 = b2 = 0) and the solid lines are the results of a non-associated flow rule
(a1 = 6  104 and a2 = b1 = b2 = 0), increasing a2 from 0 to 6  104 leads to an increase of r22 and decrease of void growth
rate.
Fig. 16 compares the results when b2 takes a different value than b1. In these analyses, a1 and a2 are taken as zero and the
results using b1 = b2 = 60.75 (associated flow rule) are denoted by the dotted lines while the results using b1 = 60.75 and
b2 = 0 (non-associated flow rule) are denoted by the solid lines. Varying b2 from 0 to 60.75 results in negligible effect on r22
but a higher void growth rate.
In the analyses presented above, the boundary conditions are imposed to keep q1 = 0.268 and q2 = 0.634. The stress-state
experienced by the material can be altered by changing the boundary conditions so that different values of q1 and q2 are
achieved. For example, q1 = 0.4 and q2 = 0.4 corresponds to a stress triaxiality ratio of 1 and Lode angle of 30°. Fig. 17 illus-
trates how this change of stress state (Lode angle changes from 0° to 30° while stress triaxiality ratio remains 1) affects r22
and the void growth rate, where a1 = a2 = 6  104 and b1 = b2 = 60.75 are used in the calculations. In Fig. 17, the dotted
lines represent the results of q1 = 0.268 and q2 = 0.634 (Lode angle equal to 0°) and the solid lines represent the results of

a b
σ 22 / σ 0 f
3 0.02

a1 = a2 = 0 0.01
b1 = b2 = -60.75
1 a1 = a2 = 0
b1 = -60.75, b2 = 0

0 0
0 0.1 0.2 0.3 0.4 0.5 0 0.1 0.2 0.3 0.4 0.5
u 2 / D0 u 2 / D0
Fig. 16. Comparison of the r22/r0 vs. u2/D0 response and f vs. u2/D0 response predicted using b1 = b2 = 60.75 (dotted lines) and b1 = 60.75 and b2 = 0 (solid
lines), where a1 and a2 are taken as zero.
230 X. Gao et al. / International Journal of Plasticity 27 (2011) 217–231

a b
σ 22 / σ 0 f
3 0.02

0.01
ρ1 = 0.268, ρ2 = 0.634
1
ρ1 = ρ2 = 0.4

0 0
0 0.1 0.2 0.3 0.4
. 0.5 0 0.1 0.2 0.3 0.4 0.5
u 2 / D0 u 2 / D0
Fig. 17. Comparison of the r22/r0 vs. u2/D0 response and f vs. u2/D0 response when q1 = 0.268 and q2 = 0.634 (dotted lines) with those when q1 = 0.4 and
q2 = 0.4 (solid lines), where a1 = a2 = 6  104 and b1 = b2 = 60.75.

q1 = 0.4 and q2 = 0.4 (Lode angle equal to 30°). The void growth rate is slightly higher and r22 becomes noticeably larger
when the Lode angle changes from 0° to 30°.
In summary, the dependence of the matrix plasticity behavior on I1 and J3 results in significant changes in void growth
and the macroscopic stress vs. deformation response of porous materials. The modified Gurson–Tvergaard–Needleman mod-
el provides a means to describe these changes.

4. Concluding remarks

In this paper, we describe a plasticity model for isotropic materials, which is dependent on the second and third invariants
of the stress deviator as well as the hydrostatic stress, and present its finite element implementation including integration of
the constitutive equations using the backward Euler method and the formulation of the consistent tangent moduli. As an
application, this model is calibrated and verified for a 5083 aluminum alloy. The model captures the strong stress-state effect
on the plastic behavior of this material and accurately predicts the plastic responses of specimens experiencing a wide range
of stress states. Furthermore, the Gurson–Tvergaard–Needleman porous plasticity model, which is widely used to simulate
the void growth process of ductile fracture, is extended to include the effects of the third invariant of the stress deviator and
the hydrostatic stress on the matrix material. A series of parametric studies illustrate the effects of model parameters on the
predicted material response. This simple modification enriches the Gurson–Tvergaard–Needleman model and expands its
applicability as a micromechanical model for ductile fracture.

Acknowledgements

This research is made possible by the funding from the Office of Naval Research N00014-09-1-0553 (Program Manager:
Dr. Paul Hess).

References

ABAQUS/Standard User’s Manual (Version 6.9), 2008. SIMULIA, Providence, RI.

Aravas, N., 1987. On the numerical integration of a class of pressure-dependent plasticity models. Int. J. Numer. Meth. Eng. 24, 1391–1416.
Bai, Y., Wierzbicki, T., 2008. A new model of metal plasticity and fracture with pressure and Lode dependence. Int. J. Plast. 24, 1071–1096.
Brunig, M., 1999. Numerical simulation of the large elastic–plastic deformation behavior of hydrostatic stress-sensitive solids. Int. J. Plast. 15, 1237–1264.
Brunig, M., Berger, S., Obrecht, H., 2000. Numerical simulation of the localization behavior of hydrostatic-stress-sensitive metals. Int. J. Mech. Sci. 42, 2147–
2166.
Brunig, M., Chyra, O., Albrecht, D., Driemeier, L., Alves, M., 2008. A ductile damage criterion at various stress triaxialities. Int. J. Plast. 24, 1731–1755.
Cazacu, O., Barlat, F., 2003. Application of representation theory to describe yielding of anisotropic aluminum alloys. Int. J. Eng. Sci. 41, 1367–1385.
Cazacu, O., Plunkett, B., Barlat, F., 2006. Orthotropic yield criterion for hexagonal close packed metals. Int. J. Plast. 22, 1171–1194.
Cazacu, O., Ionescu, I.R., Yoon, J.W., 2010. Orthotropic strain rate potential for the description of anisotropy in tension and compression of metals. Int. J. Plast.
26, 887–904.
Cvitanic, V., Vlak, F., Lozina, Z., 2008. A finite element formulation based on non-associated plasticity for sheet metal forming. Int. J. Plast. 24, 646–687.
Drucker, D.C., 1949. Relations of experiments to mathematical theories of plasticity. Trans. ASME J. Appl. Mech. 16, 349–357.
Drucker, D.C., Prager, W., 1952. Soil mechanics and plastic analysis of limit design. Quart. Appl. Math. 10, 157–165.
Faleskog, J., Gao, X., Shih, C.F., 1998. Cell model for nonlinear fracture analysis-I. Micromechanics calibration. Int. J. Fract. 89, 355–373.
Gao, X., Wang, T., Kim, J., 2005. On ductile fracture initiation toughness: effects of void volume fraction, void shape and void distribution. Int. J. Solids Struct.
42, 5097–5117.
 X. Gao et al. / International Journal of Plasticity 27 (2011) 217–231 231

Gao, X., Zhang, T., Hayden, M., Roe, C., 2009. Effects of the stress state on plasticity and ductile failure of an aluminum 5083 alloy. Int. J. Plast. 25, 2366–2382.
Gao, X., Zhang, G., Roe, C., 2010. A Study on the effect of the stress state on ductile fracture. Int. J. Damage Mech. 19, 75–94.
Gurson, A.L., 1977. Continuum of ductile rupture by void nucleation and growth: part I-yield criteria and flow rules for porous ductile media. J. Eng. Mater.
Tech. 99, 2–55.
Hencky, H., 1924. Zur Theorie plastischer Deformationen und der hierdurch im Material hervorgerufenen Nachspannungen. ZaMM 4, 323–335.
Hill, R., 1950. The Mathematical Theory of Plasticity. Oxford University Press, London.
Hu, W., Wang, Z.R., 2005. Multiple-factor dependence of the yielding behavior to isotropic ductile materials. Comput. Mat. Sci. 32, 31–46.
Kim, J., Gao, X., Srivatsan, T.S., 2004. Modeling of void growth in ductile solids: effects of stress triaxiality and initial porosity. Eng. Fract. Mech. 71, 379–400.
Kim, J., Gao, X., 2005. A generalized approach to formulate the consistent tangent stiffness in plasticity with application to the GLD porous material model.
Int. J. Solids Struct. 42, 103–122.
Kim, J., Zhang, G., Gao, X., 2007. Modeling of ductile fracture: application of the mechanism-based concepts. Int. J. Solids Struct. 44, 1844–1862.
Kuroda, M., 2004. A phenomenological plasticity model accounting for hydrostatic stress-sensitivity and vertex-type of effect. Mech. Mater. 36, 285–297.
Levy, M., 1870. Mémoire sur les équations générales des mouvements intérieurs des corps ductiles au delà des limites en élasticité pourrait les ramener à
leur premier état. C. R. Acad. Sci. 70, 1323–1325.
Lindholm, U.S., Nagy, A., Johnson, G.R., Hoegfeldt, J.M., 1980. Large strain, high strain rate testing of copper. J. Eng. Mater. Tech. 102, 376–381.
Ma, F., Kishimoto, K., 1998. On yielding and deformation of porous plastic materials. Mech. Mater. 30, 55–68.
Mendelson, A., 1968. Plasticity: Theory and Application. Macmillan Publishing Company, New York.
Mirone, G., Corallo, D., 2010. A local viewpoint for evaluating the influence of stress triaxiality and Lode angle on ductile failure and hardening. Int. J. Plast.
26, 348–371.
Mroz, Z., 1963. Non-associated flow-laws in plasticity. J. Méc. Théo. Appl. 2, 21–42.
Nemat-Nasser, S., Shokooh, A., 1980. On finite plastic flows of compressible materials with internal friction. Int. J. Solids Struct. 16, 495–514.
Nemat-Nasser, S., 1983. On finite plastic flow of crystalline solids and geomaterials. ASME J. Appl. Mech. 50, 1114–1126.
Nemat-Nasser, S., 1992. Phenomenological theories of elasto-plasticity and strain localization at high strain rates. Appl. Mech. Rev. 45, s19–s45.
Plunkett, B., Cazacu, O., Barlat, F., 2008. Orthotropic yield criteria for description of the anisotropy in tension and compression of sheet metals. Int. J. Plast.
24, 847–866.
Prandtl, L., 1925. Spannungsverteilung in plastischen Koerpern. In: Waltman Jr., J. (Ed.), Proceedings of the 1st Intrnational Congress on Applied Mechanics.
Delft, Technische Boekhandel en Druckerij, pp. 43–54.
Reuss, E., 1930. Beruecksichtigung der elastischen Formaenderungen in der Plastizitaetstheorie. Z. Angew. Math. Mech. 10, 266–274.
Runesson, K., Mroz, Z., 1989. A note on non-associated plastic flow rules. Int. J. Plast. 5, 639–658.
Saint-Venant, B.de., 1870. Sur l’établissement des équations des mouvements intérieurs opérés dans les corps solides ductiles au-delà des limites où
l’élasticité pourrait les ramener à leur premier état. Comptes Rendus des Séances de 70, 473–480.
Simo, J.C., Taylor, R.L., 1985. Consistent tangent operators for rate-independent elasto-plasticity. Comput. Methods Appl. Mech. Eng. 48, 101–118.
Soare, S., Yoon, J.W., Cazacu, O., barlat, F., 2007. Applications of a recently proposed anisotropic yield function to sheet forming. In: Banabic, D. (Ed.),
Applications of a Recently Proposed Anisotropic Yield Function to Sheet Forming. Springer, Berlin, Heidelberg, pp. 131–149.
Spitzig, W.A., Sober, R.J., Richmond, O., 1975. Pressure dependence of yielding and associated volume expansion in tempered martensite. Acta Metall. 23,
885–893.
Spitzig, W.A., Sober, R.J., Richmond, O., 1976. The effect of hydrostatic pressure on the deformation behavior of maraging and HY-80 steels and its
implications for plasticity theory. Metall. Trans. 7A, 1703–1710.
Stoughton, T.B., 2002. A non-associated flow rule for sheet metal forming. Int. J. Plast. 18, 687–714.
Stoughton, T.B., Yoon, J.W., 2004. A pressure-sensitive yield criterion under a non-associated flow rule for sheet metal forming. Int. J. Plast. 20, 705–731.
Stoughton, T.B., Yoon, J.W., 2006. Review of Drucker’s postulate and the issue of plastic stability in metal forming. Int. J. Plast. 22, 391–433.
Tresca, H., 1864. Memoir on the flow of solid bodies under strong pressure. Comptes-rendus de l’académie des sciences 59, 754.
Tvergaard, V., 1981. Influence of voids on shear band instabilities under plane strain conditions. Int. J. Fract. 17, 389–407.
Tvergaard, V., 1982. On Localization in ductile materials containing spherical voids. Int. J. Fract. 18, 237–252.
Tvergaard, V., Needleman, A., 1984. Analysis of the cup–cone fracture in a round tensile bar. Acta Metall. 32, 157–169.
von Mises, R., 1913. Mechanik der festen Körpern im plastisch-deformablen Zustand. Nachrichten von der königlichen Gesellschaft der Wissenschaften zu
Göttingen. Mathematisch-Physikalische Klasse, 582–592.
Wierzbicki, T., Xue, L., 2005. On the effect of the third invariant of the stress deviator on ductile fracture. Technical Report, Impact and Crashworthiness Lab,
MIT.
Xue, L., 2007. Damage accumulation and fracture initiation in uncracked ductile solids subject to triaxial loading. Int. J. Solids Struct. 44, 5163–5181.