Chp.12.Elasticity and Plasticity 2019

12
Elasticity and plasticity
Theory: Elastic rheology. Rotation of elastic stresses. Maxwell visco-elastic

rheology. Plastic rheology. Plastic yielding criterion. Plastic flow potential.
Plastic flow rule. Visco-elasto-plastic rheology.
Exercises: Stress buildup/relaxation with a visco-elastic Maxwell rheology,
elastic stress rotation programming.
12.1 Why should we care about elasticity and plasticity?

As mentioned in the Introduction, rocks behave elastically on a relatively short time scale
(<104 years) and, therefore, modelling of relatively fast processes within the Earth’s crust
and mantle (e.g. magma intrusion) should take into account the elastic properties of rocks.
On the other hand, rocks at cold temperatures can also be subjected to localized brittle (at
low pressure) and plastic (at higher pressure) deformation, which leads to shear zones and
fracture zones in natural rock complexes. Therefore, if we want to account for this broad
range of geodynamic conditions in our models, we should generally consider the visco-
elasto-plastic rheology of rocks and be able to model such a complex rheology with our
thermomechanical numerical codes. This chapter discusses elastic and plastic rheological
behaviours and compares them to the viscous rheology. Please note that in geology viscous
and plastic deformation are often used as synonyms, which is thus different from the
continuum mechanics convention used here.
12.2 Elastic rheology

The elastic rheology assumes proportionality of stress and strain (Fig. 12.1). This is
expressed by Hooke’s law
τ ¼ Eγ; ð12:1Þ
171
Downloaded from https://www.cambridge.org/core. Nottingham Trent University, on 27 Aug 2019 at 19:38:46, subject to the Cambridge Core terms of use, available
at https://www.cambridge.org/core/terms. https://doi.org/10.1017/9781316534243.013
172 Elasticity and plasticity
Fig. 12.1. Relationship between applied stress τ and deformation ΔL, of an elastic body with initial
length L.
Fig. 12.2. Reversible deformation of initially unstressed (a) elastic slab surrounded by a weak
viscous medium (Gerya and Yuen, 2007). Deformation of the slab in (b) is caused by a vertical
gravity field. When gravity is ‘removed’, the deformed slab recovers its original shape (c) while the
surrounding medium remains deformed since viscous deformation is irreversible.
where τ is applied stress, γ ¼ ΔL=L is elastic strain (i.e. displacement ΔL normalized to the
initial length L of the deforming body) and E is the proportionality (elasticity) coefficient.
In contrast to viscous deformation, elastic deformation is reversible: if the load applied on
an elastic body is removed, the body recovers its original state (Fig. 12.2). This shape
recovery effect is reproduced with a spring.
For an isotropic body in 3D, the elastic relationship is written in tensorial form as
12.2 Elastic rheology 173
σ ij ¼ λδij εkk þ 2μεij ; ð12:2Þ

1 ∂ui ∂uj
εij ¼ þ ; ð12:3Þ
2 ∂xj ∂xi
εkk ¼ εxx þ εyy þ εzz ; ð12:4Þ
where xi and xj are the spatial coordinates (x, y, z), σij are components of stress, ui and uj are
components of the material displacement vector ~ u ¼ ðux ; uy ; uz Þ so that vi ¼ Dui =Dt,
where vi are components of the velocity vector ~ v ¼ ðvx ; vy ; vz Þ; εij are components of the
strain tensor so that ε_ ij ¼ Dεij =Dt, where ε_ ij are components of the strain rate tensor
(Chapter 4), εkk is volumetric strain (cubical dilatation) and λ and μ are two elastic
parameters termed Lamé’s constants (which depend on pressure, temperature and composi-
tion). The material displacement characterizes the absolute amount of movement (i.e. is
similar to ΔL in Fig. 12.1), while the strain tensor εij reflects the relative intensity of the
deformation (i.e. is similar to ΔL=L in Fig. 12.1).
Formulating pressure through mean normal stress (Chapter 4), we can obtain the
following relations
σ kk σ xx þ σyy þ σ zz
P¼ ¼ ; ð12:5Þ
3 3
3λ þ 2μ
P¼ εxx þ εyy þ εzz ¼ Bεkk ; ð12:6Þ
3
2
B ¼ λ þ μ; ð12:7Þ
3
where B is the bulk modulus (see Chapter 2) also called incompressibility; it establishes the
relation between mean stress (pressure) and volumetric strain. Accordingly, the deviatoric
0
stresses σij can be formulated as
0 σ kk
σ ij ¼ σ ij δij ¼ σ ij þ Pδij ; ð12:8Þ
3

0 1
σ ij ¼ σ ij Bεkk δij ¼ 2μ εij δij εkk ; ð12:9Þ
3
0 0
σ ij ¼ 2μεij ; ð12:10Þ
0 1
εij ¼ εij δij εkk ; ð12:11Þ
3
0
where εij are components of the deviatoric strain tensor and μ is the shear modulus or
rigidity, which is one of the Lamé constants. The elastic shear modulus establishes the
relationship between the deviatoric stress and deviatoric strain for the elastic rheology (μ is
Fig. 12.3. Force balance for a small triangle ABC used for computing σxx, σyx (a), (c) and σyy, σxy (b),
(d) stress components after a clockwise solid body rotation of the triangle by an angle θ around
point A.
thus somewhat similar to the shear viscosity η that defines the analogous relationship
between the deviatoric stress and the deviatoric strain rate for a viscous rheology, see
Eq. 5.11).
12.3 Rotation of elastic stresses

An important peculiarity of stress behaviour in a deforming elastic medium consists in
changes in local stress orientation due to the rotation of material points. This rotation is
caused by a rigid body rotation component, which is present in the velocity field and
changes the orientation of principal stress axes for moving Lagrangian points. Stress
rotation changes stress tensor components but not the stress invariants P and σ II .
The method of computing various elastic stress components after rotation of
a Lagrangian point by an angle θ can be derived on the basis of analysing the force balance
for a small triangle ABC, as shown in Fig. 12.3. Let us take a small triangle with two sides
12.3 Rotation of elastic stresses 175
parallel to the x and y axes oriented as shown in Fig. 12.3a. The force equilibrium condition
requires that forces acting on the outside of the triangle balance each other and that the
resulting force is zero
fx ¼ fx1 þ fx2 þ fx3 þ fx4 ¼ 0; ð12:12Þ

fy ¼ fy1 þ fy2 þ fy3 þ fy4 ¼ 0; ð12:13Þ
where f are forces acting on different sides of the triangle from the outside. These forces can
be computed from shear and normal stresses as discussed in Chapter 4 (note that arrows
shown on side AC correspond to the counterforce part of σxx and σyx which therefore have
a minus sign)
jACj ¼ jABj cosðθÞ;

jBCj ¼ jABj sinðθÞ;
fx1 ¼ σ x0 x0 jABj cosðθÞ;
fx2 ¼ σy0 x0 jABj sinðθÞ;
fx3 ¼ σ xx jACj ¼ σ xx jABj cosðθÞ;

fx4 ¼ σ xy jBCj ¼ σ xy jABj sinðθÞ;
fy1 ¼ σ x0 x0 jABj sinðθÞ;
fy2 ¼ σ y0 x0 jABj cosðθÞ;
fy3 ¼ σ yx jACj ¼ σ yx jABj cosðθÞ;
fy4 ¼ σ yy jBCj ¼ σ yy jABj sinðθÞ;
where jABj, jACj and jBCj are the lengths of the respective triangle sides. Then, Eqs.
(12.12) and (12.13) can be converted to yield
fx
¼ σ x0 x0 cosðθÞ þ σ y0 x0 sinðθÞ σ xx cosðθÞ þ σ xy sinðθÞ ¼ 0; ð12:14Þ
jABj
fy
¼ σ x0 x0 sinðθÞ þ σ y0 x0 cosðθÞ σ yx cosðθÞ þ σ yy sinðθÞ ¼ 0: ð12:15Þ
jABj
By multiplying Eq. (12.14) by cos(θ) and Eq. (12.15) by sin(θ), we obtain
σ x0 x0 cos2 ðθÞ þ σ y0 x0 sinðθÞcosðθÞ σ xx cos2 ðθÞ þ σ xy sinðθÞcosðθÞ ¼ 0; ð12:16Þ
σ x0 x0 sin2 ðθÞ þ σ y0 x0 cosðθÞsinðθÞ σ yx cosðθÞsinðθÞ þ σ yy sin2 ðθÞ ¼ 0: ð12:17Þ
By subtracting Eq. (12.17) from Eq. (12.16) we get

σ x0 x0 cos2 ðθÞ þ sin2 ðθÞ σ xx cos2 ðθÞ
þ σxy cosðθÞsinðθÞ þ σ yx cosðθÞsinðθÞ σ yy sin2 ðθÞ ¼ 0;
which can be further simplified to
σ x0 x0 ¼ σ xx cos2 ðθÞ þ σ yy sin2 ðθÞ σ xy sinð2θÞ ð12:18Þ
by using σ yx ¼ σ xy and the trigonometric relations sin2 ðθÞ þ cos2 ðθÞ ¼ 1 and
2 sinðθÞcosðθÞ ¼ sinð2θÞ.
Similarly, by multiplying Eq. (12.14) by sin(θ) and Eq. (12.15) by cos(θ), and adding
them to each other, we obtain the following expression for σ y0 x0
1
σ y0 x0 ¼ σ xx σ yy sinð2θÞ þ σ xy cosð2θÞ ð12:19Þ
2
using the trigonometric relation cos2 ðθÞ sin2 ðθÞ ¼ cosð2θÞ (verify as an exercise).
Obviously, after the triangle has been rotated clockwise by an angle θ around point
A (Fig. 12.3c), the AB side becomes parallel to the y axis and σ y0 x0 ; σ y0 x0 , will correspond to
the respective stress components σ yxðrÞ ; σ yxðrÞ , for the rotated system.
Similarly, by analysing Fig. 12.3b,d, the following expressions for the corrected stress
components σ y0 y0 and σ x0 y0 can be obtained (verify as an exercise):
σ y0 y0 ¼ σ xx sin2 ðθÞ þ σ yy cos2 ðθÞ þ σxy sinð2θÞ; ð12:20Þ
1
σ x0 y0 ¼ σ xx σ yy sinð2θÞ þ σ xy cosð2θÞ: ð12:21Þ
2
Equations (12.19) and (12.21) are obviously equivalent since σ yx ¼ σ xy . Note that
rotation does not change the first stress invariant (mean normal stress, pressure) and thus
σ x0 x0 þ σ y0 y0 ¼ σ xx þ σ yy ¼ 2P:
Equations for rotating deviatoric normal stresses are similar to Eqs. (12.18) and (12.20)
since subtracting mean stress does not change the form of these expressions, hence
0 0 0 0
σ x0 x0 ¼ σ xx cos2 ðθÞ þ σyy sin2 ðθÞ σxy sinð2θÞ; ð12:22Þ
0 0 0 0
σ y0 y0 ¼ σ xx sin2 ðθÞ þ σ yy cos2 ðθÞ þ σxy sinð2θÞ: ð12:23Þ
0 0 0 0
The condition σ x0 x0 þ σ y0 y0 ¼ 0 is also satisfied due to σ xx þ σ yy ¼ 0.
If the angle θ is very small and tends to 0, then cos(θ) tends to 1, sin2(θ) tends to 0, and
sin(θ) tends to θ. In this case, Eqs. (12.18)–(12.23) can be simplified to the Jaumann
co-rotation formulas, which are often used in numerical modelling of elastic problems to
account for effects of rigid body rotation of stresses:
0 0 0
σ x0 x0 ¼ σ xx σ xy 2θ; ð12:24Þ
0 0 0
σ y0 y0 ¼ σ yy þ σ xy 2θ; ð12:25Þ
0 0
0 0

σ x0 y0 ¼ σ xy þ σ xx σ yy θ: ð12:26Þ
In a complex velocity field, the intensity of rotation is defined by the local rotation rate ω,
which can be computed from the local velocity field as

∂θ 1 ∂vy ∂vx
ω¼ ¼ : ð12:27Þ
∂t 2 ∂x ∂y
Note that the expressions for stress rotation and formulation of ω depend on
• the convention for normal stresses – whether they are taken to be positive (here) or
negative (e.g. Turcotte and Schubert, 2002) under extension,
• orientation of the x and y axes,
• whether the clockwise rotation direction is taken to be positive (here) or negative (e.g.
Turcotte and Shubert, 2002).
In particular, Eq. (12.27) may become invalid if the definitions of stresses, axes and
rotation are different from those used here. In this case, Eqs. (12.18)–(12.27) should not
be used ‘automatically’, but should rather be re-derived on the basis of a similar
analysis.
In 3D, the rotation rate is represented by a rotation rate tensor ω with components
defined as

1 ∂vi ∂vj
ωij ¼ ; ð12:28Þ
2 ∂xj ∂xi
where i and j are coordinate indexes; xi and xj are spatial coordinates and the view at
the ij-plane should be taken in the direction of the third axis. In contrast to symmetric
stress and strain rate tensors i:e: σ ij ¼ σ ji ; ε_ ij ¼ ε_ ji Þ, the rotation rate tensor is anti-
symmetric i:e:; ωij ¼ ωji Þ and the diagonal components of this tensor are always
equal to zero ði:e:; ωxx ¼ ωyy ¼ ωzz ¼ 0Þ. One possible way of computing both total and
deviatoric stress rotation in 3D is to use the general form of the Jaumann stress rate:
total stress
σ_ ijðJaumannÞ ¼ ωik σ kj σ ik ωkj ; ð12:29Þ
deviatoric stress
0 0 0
σ_ ijðJaumannÞ ¼ ωik σ kj σ ik ωkj ; ð12:30Þ
where σ_ ijðJaumannÞ is the rate of change for the rotating σ ij stress component in the moving
non-rotating reference frame and the repeated index k indicates a summation. Using Eq.
0
(12.30) in 3D for example, the σ xx deviatoric stress component yields
0 0 0 0 0 0 0
σ_ xxðJaumannÞ ¼ ωxx σ xx þ ωxy σ yx þ ωxz σ zx σ xx ωxx σ xy ωyx σ xz ωzx ; ð12:31aÞ
0 0 0 0 0
σ_ xxðJaumannÞ ¼ ωxy σyx þ ωxz σ zx σ xy ωyx σ xz ωzx ; ð12:31bÞ
0 0 0
σ_ xxðJaumannÞ ¼ 2σxy ωxy þ 2σ xz ωxz ; ð12:31cÞ
where

1 ∂vx ∂vx 1 ∂vx ∂vy
ωxx ¼ ¼ 0; ωxy ¼ ¼ ωyx
2 ∂x ∂x 2 ∂y ∂x
and

1 ∂vx ∂vz
ωxz ¼ ¼ ωzx ;
2 ∂z ∂x
according to Eq. (12.28). Similar derivations can also be done for other deviatoric stress
components (verify these as an exercise):
0 0 0
σ_ yyðJaumannÞ ¼ 2σ yx ωyx þ 2σyz ωyz ; ð12:32Þ
0 0 0
σ_ zzðJaumannÞ ¼ 2σ zx ωzx þ 2σ zy ωzy ; ð12:33Þ
0 0
0 0
0 0
σ_ xyðJaumannÞ ¼ σ_ yxðJaumannÞ ¼ σ xx σ yy ωyx þ ωxz σzy σ xz ωzy ; ð12:34Þ
0 0
0 0
0 0
σ_ xzðJaumannÞ ¼ σ_ zxðJaumannÞ ¼ σ xx σ zz ωzx þ ωxy σ yz σ xy ωyz ; ð12:35Þ
0 0
0 0
0 0
σ_ yzðJaumannÞ ¼ σ_ zyðJaumannÞ ¼ σ yy σ zz ωzy þ ωyx σ xz σ yx ωxz : ð12:36Þ
In 2D, Eqs. (12.31)–(12.33) are equivalent to previously derived Eqs. (12.24)–(12.27)

since ω ¼ ωyx ¼ ωxy .
It is worth mentioning that in continuum mechanics, in addition to the Jaumann stress
rate, typically used in geodynamic modelling, there are a large variety of other objective
stress rate formulations such as the Truesdell rate, the Green–Naghdi rate, the Oldroyd rate,
the convective rate etc. (e.g. Shabana, 2008). However, the other objective derivatives
(beside Jaumann) do not preserve the deviatoric property of a tensor. Hence, using them in
our case is not straightforward as our formulations assume a splitting of stress into
a deviatoric and a homogeneous (pressure) part.
It should also be mentioned that numerical calculations of stress rotation based on the
Jaumann stress rate are less accurate than the analytical formulas (12.18)–(12.21) for 2D
finite angle rotation. Accuracy of the calculation can be improved by applying a smaller
time step and/or using higher order Runge–Kutta integration schemes in stress space (e.g.
Farrington et al., 2014; Popov et al., 2014a; Kaus et al., 2016; Popov, personal commu-
nication). Another efficient way to improve the accuracy of 3D stress rotation calculations
is to extend the 2D analytical finite angle stress rotation approach to 3D (e.g. Rubinstein and
Atluri, 1983; Popov et al., 2014a; Kaus et al., 2016). One optimal numerical 3D finite angle
stress rotation algorithm has been recently proposed and tested by Popov et al. (2014a,
personal communication), which is based on the work of Rubinstein and Atluri (1983).
According to this approach, 3D rotation is represented by a vorticity pseudo-vector (~ ω ),
which has three components:

1 ∂vz ∂vy 1 ∂vx ∂vz 1 ∂vy ∂vx
ωx ¼ ; ωy ¼ ; ωz ¼ : ð12:37Þ
2 ∂y ∂z 2 ∂z ∂x 2 ∂x ∂y
The 3D algorithm can then be summarized as follows (Popov et al., 2014a, personal
communication).
(1) Compute the vorticity vector magnitude:
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
ωmag ¼ ω2x þ ω2y þ ω2z : ð12:38Þ
(2) Compute the unit rotation vector ~

n, which also has three components:
ωx ωy ωz
nx ¼ ; ny ¼ ; nz ¼ : ð12:39Þ
ωmag ωmag ωmag
(3) Integrate the incremental rotation angle:
θ ¼ ωmag Δt: ð12:40Þ
(4) Evaluate the rotation matrix using the Euler–Rodrigues formula:

0 1 0 1 0 1
1 0 0 0 nz ny nx nx nx ny nx nz
Rmat ¼ cosðθÞ@ 0 1 0 A þ sinðθÞ@ nz 0 nx A þ 1 cosðθÞ @ ny nx ny ny ny nz A:
0 0 1 ny nx 0 nz nx nz ny nz nz
ð12:41Þ
Fig. 12.4. Schematic representation of the Maxwell visco-elastic rheology. The solid line shows
a typical pattern of visco-elastic stress buildup under the condition of a linearly growing deformation
(dashed line). Deformation initially starts in an elastic mode (see shortening of the spring) but with the
growing stress, viscous deformation activates and becomes dominant (see movement of the dashpot)
and stress stabilizes. The length of the black arrows reflects the magnitude of the applied stress at
different moments in time.
0
(5) Compute the rotated stress matrix by multiplying Rmat, σ and RmatT matrices:
0 0
σrotated ¼ Rmat σ Rmat T ; ð12:42Þ
where RmatT is the transpose of the matrix Rmat (RmatT is obtained by reflecting the elements
of the matrix Rmat along its main diagonal).
This simple 3D stress rotation algorithm is very efficient and has superior accuracy
compared to other methods (Popov et al., 2014a; Kaus et al., 2016).
12.4 Maxwell visco-elastic rheology

A visco-elastic rheology is obtained by combining viscous (Eq. 5.11) and elastic (Eq. 12.10)
rheological relations under certain physical assumptions (e.g. Turcotte and Schubert, 2002,
Chapter 7, Section 7–10). In numerical geodynamic modelling, Maxwell visco-elastic
rheology is the most commonly used type; it is based on the assumption that both viscous
0
and elastic deformations are happening under the same applied deviatoric stress σij such
0 0
that the bulk deviatoric strain rate ε_ ij can be represented as a sum of viscous ε_ ijðviscousÞ and
0
elastic ε_ ijðelasticÞ strain rates (see Fig. 12.4 for the relationship between the viscous and elastic
deformations of a Maxwell body)
0 0 0
ε_ ij ¼ ε_ ijðviscousÞ þ ε_ ijðelasticÞ ; ð12:43Þ
12.5 Plastic rheology 181
which can then be obtained from the rheological relations (5.11) and (12.10) under the
assumption that the shear modulus in Eq. (12.10) is constant:
0 1 0
ε_ ijðviscousÞ ¼ σ ; ð12:44Þ
2η ij
0 0
! 0
0 ⌢ εijðelasticÞ
D D
⌢
σ ij 1 D⌢ σ ij
ε_ ijðelasticÞ ¼ ¼ ¼ ; ð12:45Þ
Dt Dt 2μ 2μ Dt
0
0 1 0 ⌢ σ ij
1 D
ε_ ij ¼ σ ij þ ; ð12:46Þ
2η 2μ Dt
0
where D ⌢ σ ij =Dt is the objective co-rotational time derivative of the deviatoric stress
0
component σ ij in a moving non-rotating Lagrangian reference frame which accounts for
the effects of stress rotation discussed above.
It should be noted that Eq. (12.45) in fact re-defines the shear modulus μ as
0
⌢ σ ij
D 0
¼ 2με_ ij ; ð12:47Þ
Dt
which (at variable μ) differs from its original definition by Eq. (12.10). Similarly, the bulk
modulus can be re-defined compared to Eq. (12.6) as
DP
¼ Bε_ kk ; ð12:48Þ
Dt
which also directly relates (through the continuity equation, Chapter 1) the bulk modulus
B and the compressibility β (Chapter 2)
1 DP 1 DP 1 DP 1 DP 1
B¼ ¼ ¼ DlnðρÞ ¼ ∂lnðρÞ ¼ : ð12:49Þ
ε_ kk Dt divð~
vÞ Dt Dt DP Dt β
Dt ∂P Dt
The modified definitions of shear and bulk moduli by Eqs. (12.47)–(12.49) are consistent
with the Biot non-linear elasticity theory of incremental deformation (Biot, 1965). These
definitions are more convenient for modelling deformation of complex visco-elastic and
visco-elasto-plastic materials, such as rocks, under conditions of broad variations of
pressure and temperature causing variations of elastic moduli.
12.5 Plastic rheology

The plastic rheology assumes that an absolute shear stress limit σ yield exists for a body and
after reaching this limit plastic yielding occurs (Fig. 12.5). Like viscous deformation,
plastic yielding is irreversible, but the pattern of deformation is notably different
Fig. 12.5. Relationship between the applied stress σ and deformation ΔL, of an elastic-plastic body
with initial length L. Elastic deformation changes to plastic yielding after reaching a stress limit
σyield.
Fig. 12.6. Plastic deformation of sand in a numerical sandbox experiment (Buiter et al., 2006; Gerya
and Yuen, 2007). Irreversible localized plastic deformation forms multiple, highly deformed shear
zones separating relatively undeformed blocks.
(Fig. 12.6): plastic creep is localized and forms multiple highly deformed shear zones
separating relatively undeformed blocks.
The plastic strength σ yield of a rock generally depends on the mean stress of the solid (Ps)
and on the pore fluid pressure (Pf, we will discuss this in more detail in Chapter 16). Based
on the simplified Griffith theory (Griffith, 1924; Cai, 2010; Keller et al., 2013) it could be
assumed that:
12.5 Plastic rheology 183
for dry rocks
σc σt
σyield ¼ σ c þ γint P when P > ðconfined fracturesÞ; ð12:50aÞ
1 γint
σc σt
σ yield ¼ σ t þ P when P < ðtensile fracturesÞ; ð12:50bÞ
1 γint
and for fluid-bearing rocks

σc σt
σ yield ¼ σ c þ γint Pt Pf when Pt Pf > ðconfined fracturesÞ ð12:51aÞ
1 γint
σc σt
σyield ¼ σ t þ Pt Pf when Pt Pf < ðtensile fracturesÞ ð12:51bÞ
1 γint
Pt ¼ ð1 ϕÞPs þ ϕPf ; ð12:52Þ
where σyield ≥ 0 (non-negative strength requirement), γint ¼ sinðφÞ is the internal friction
coefficient (φ is the angle of internal friction), Pt = P is total pressure, ϕ is fluid volume
fraction (porosity), σc and σt are respectively compressive strength and tensile strength
(2 ≤ ðσ c =σ t Þ ≤ 8, Cai, 2010; Keller et al., 2013). The compressive strength σc is related to the
material cohesion C ¼ σ c =cosðφÞ. For the majority of dry fractured crystalline rocks, γint is
independent of composition and varies from 0.85 at P < 200 MPa to 0.60 at higher pressure
(Byerlee’s law, Byerlee, 1978; Brace and Kohlstedt, 1980). The plastic strength of dry rocks
thus strongly increases with pressure to a limit of several GPa. The strength is limited by the
Peierls mechanism of plastic deformation (Evans and Goetze, 1979; Kameyama et al.,
1999; Karato, 2008).
The Peierls mechanism is a temperature-dependent mode of plastic deformation (also
called exponential creep), which takes over from the dislocation creep mechanism at
elevated stresses (typically above 0.1 GPa). Rheological relationships (flow law) for
Peierls creep are commonly represented as (Katayama and Karato, 2008)
( " k #q )
Ea þ PVa σ II
ε_ II ¼ APeierls σ II exp
2
1 ; ð12:53Þ
RT σ Peierls
where σPeierls, APeierls, Ea, Va, k and q are experimentally determined parameters
(Chapter 6): σPeierls is the Peierls stress that limits the strength of the material and is similar
to σ yield in Eq. (12.50), APeierls is a material constant for Peierls creep (Pa−2 s−1), Ea is the
activation energy (J/mol) and Va is the activation volume (J/Pa), the exponents k and
q depend on the shape and geometry of obstacles that limit the dislocation motion.
Microscopic models show that k and q should have the following range 0 < k ≤ 1, 1 ≤ q ≤
2 (Kocks et al., 1975). In contrast to other types of plasticity, Peierls creep is already
activated at stresses that are notably lower than the actual strength of material given by
σPeierls. This deformation mechanism is very important, in particular for deformation of

subducting slabs characterized by lowered temperature and elevated stresses compared to
the surrounding mantle (e.g. Duretz et al., 2011b; Karato et al., 2001), or for lithospheric-
scale shear localization (Kaus and Podladchikov, 2006).
Further information about various types of plasticity used in geosciences such as
Mohr–Coulomb, Von Mises, Drucker–Prager and Treska models can be found in the
books of Turcotte and Schubert (2002) and Ranalli (1995).
12.6 Visco-elasto-plastic rheology

In nature, the general behaviour of rocks is visco-elasto-plastic, which can be formulated by
0
decomposing the bulk deviatoric strain rate ε_ ij into the three respective components:
0 0 0 0
ε_ ij ¼ ε_ ijðviscousÞ þ ε_ ijðelasticÞ þ ε_ ijðplasticÞ ; ð12:54Þ
where
0 1 0
ε_ ijðviscousÞ ¼ σ ; ð12:55Þ
2η ij
0
0 1 D⌢ σ ij
ε_ ijðelasticÞ ¼ ; ð12:56Þ
2μ Dt
0
ε_ ijð plasticÞ ¼ 0 for σ II < σ yield ;
0
0 ∂Gplastic σ ij ð12:57Þ
ε_ ijð plasticÞ ¼ χ 0 ¼χ for σ II ¼ σ yield ;
∂σ ij 2σII
Gplastic ¼ σ II ; ð12:58Þ
rffiffiffiffiffiffiffiffiffiffi
1 02
σ II ¼ σ ; ð12:59Þ
2 ij
0
where D ⌢ σ ij =Dt
0
is the objective co-rotational time derivative of the deviatoric stress
component σ ij , Eq. (12.57) is the plastic flow rule, σyield is the plastic yield strength for
a given rock, σII is the second invariant of the deviatoric stress tensor (Eq. 4.5), Gplastic is the
plastic flow potential, which reflects the amount of mechanical energy per unit volume that
supports plastic deformation, Pa ¼ N=m2 ¼ J=m3 , and χ is the plastic multiplier, which
satisfies the Drucker–Prager plastic yielding condition
σ II ¼ σ yield : ð12:60Þ
The plastic multiplier is a variable scaling coefficient, which connects, in a uniform way,
0
components of the plastic strain rate ε_ ijð plasticÞ with the deviatoric stress components when
Programming exercises 185
the yielding condition (12.60) is reached. This coefficient is unknown a priori and should be
determined locally at each moment of time by solving Eqs. (12.54)–(12.60), based on local
0 0
values of stresses σ ij , strain rates ε_ ij , viscosity η and shear modulus μ. Based on Eqs.
(12.57)–(12.60), we can conclude that the plastic multiplier χ is equal to the double of
the second invariant of the deviatoric plastic strain rate tensor ε_ IIðplasticÞ (derive as an
exercise):
χ ¼ 2ε_ IIðplasticÞ ; ð12:61Þ

rffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
1 02
ε_ IIð plasticÞ ¼ ε_ : ð12:62Þ
2 ijðplasticÞ
Plastic flow rule formulation (Eq. 12.57) includes deviatoric stress and strain rate
components only and, consequently, the plastic potential formulation (Eq. 12.58) is the
same for both dilatant (i.e. increasing volume during plastic deformation) and non-dilatant
materials. For plastic deformation of dilatant materials, this formulation is, therefore,
combined with the equation describing volumetric changes:
vÞ ¼ 2 sinðψÞε_ IIð plasticÞ ;

Γplastic ¼ divð~ ð12:63Þ
where ψ is the dilatation angle, which generally depends on total plastic strain.
Analytical exercise
Exercise 12.1.
Derive the equation for visco-elastic stress buildup/relaxation with time, using the Maxwell
0
model (Eq. 12.46) under conditions of constant strain rate ε_ ij , viscosity η, shear modulus μ,
0 0
and no stress rotation involved such that D
0
⌢ σ ij =Dt ¼ Dσ ij =Dt: Take the initial state of stress
0
to be given by σ 0ij and integrate Eq. (12.46) (now reformulated in terms of stress σ ij ) to
obtain the analytical solution. Reformulate the resulting equation in terms of Maxwell
relaxation time
η
tMaxwell ¼ ; ð12:64Þ
μ
which defines the characteristic time scale for visco-elastic stress relaxation.
Programming exercises
Exercise 12.2.
Use the analytical formula from the previous example to compute and compare stress-time
0 0
curves for the following parameters: (1) σ 0ij = 0 Pa, ε_ ij = 10−14 1/s, η = 1021 Pa s, μ = 1010 Pa;
0 0 0 0
(2) σ0ij = 108 Pa, ε_ ij = 10−14 1/s, η = 1021 Pa s, μ = 1010 Pa; (3) σ 0ij = 0 Pa, ε_ ij = 10−15 1/s,
0 0
η = 1021 Pa s, μ = 1010 Pa; (4) σ 0ij = 0 Pa, ε_ ij = 10−14 1/s, η = 1022 Pa s, μ =1010 Pa; (5)
0 0 0 0
σ0ij = 0 Pa, ε_ ij = 10−14 1/s, η = 1021 Pa s, μ = 1011 Pa; (6) σ 0ij = 0 Pa, ε_ ij = 10−14 1/s, η = 1022
Pa s, μ = 1011 Pa. Try to understand how the different parameters control the stress buildup/
relaxation. An example is in Viscoelastic_stress.m.
Exercise 12.3.
Compute the visco-elasto-plastic stress buildup and observe the changes in the viscous,
elastic and plastic strain rates with time when using the parameters from case (4) of the
0
previous example. Assume the condition that the visco-elastic stress σ ij must not exceed the
yield stress limit of 1.5 × 108 Pa. Use Eq. (12.55) to compute the viscous strain rate. Use
Eqs. (12.54), (12.56) and (12.57) to compute elastic and plastic strain rates. Consider that
after reaching the yielding limit, the stress in the visco-elasto-plastic material should
0 0
not change anymore and therefore D ⌢ σ ij =Dt ¼ Dσ ij =Dt ¼ 0: An example is in
Viscoelastoplastic_strain_rate.m.
Exercise 12.4.
Modify Exercise 6.3 by adding Peierls creep for the high stress region (>108 Pa). Compute
the effective viscosity for this region by analogy to Eqs. (6.16)–(6.18) as follows
1 1 1 1
¼ þ þ ; ð12:65Þ
ηeff ηdiff ηdisl ηPeierls
where ηPeierls is Peierls creep viscosity defined on the basis of Eqs. (12.53) and (6.4) as
( " k #q )
1 Ea þ PVa σ II
ηPeierls ¼ exp 1 : ð12:66Þ
2APeierls σ II RT σPeierls
Use the following Peierls creep parameters (Evans and Goetze, 1979; Katayama
and Karato, 2008): dry olivine, k = 1, q = 2, APeierls = 10−4.2 Pa−2 s−1, Ea = 540 000 J/mol,
σPeierls = 9.1 × 109 Pa; wet olivine, k = 1, q = 2, APeierls = 10−4.2 Pa−2 s−1, σPeierls = 2.9 × 109
Pa; Ea = 4 300 000 J/mol. Note that the activation energy Ea for Peierls creep is the same as
the activation energy for respective dislocation creep (Table 6.1). Note that the stress σII
used for computing effective viscosity with Eq. (12.66) should always be limited by σPeierls,
which corresponds to the upper strength limit. An example is in Peierls_creep.m.
Exercise 12.5.
Compare different formulas for the case of 2D stress rotation with constant angular velocity
ω ¼ 1. Use different numbers of time steps per one revolution time interval (t = 2π/ω) and
0 0 0
compare differences between the initial ðσxx ¼ 106 Pa; σ yy ¼ 106 Pa; σ xy ¼ 0Þ and final
stress states for the following approaches: (A) analytical (Eqs. 12.21–12.23), (B) Jaumann
(Eqs. 12.24–12.26), (C) Jaumann with effective stress rate (Eqs. 12.31, 12.32, 12.34 with
0 0 0
ωyx = −ωxy = ω, ωxz = ωzx = ωyz = ωzy = 0, σ xz ¼ σ yz ¼ σ zz ¼ 0) computed with a fourth-
order (in stress space) Runge–Kutta method (Chapter 8) and (D) 3D finite angle rotation
Programming exercises 187
0 0 0
(Eqs. 12.37–12.42 with ωx = ωy = 0, ωz = ω, σ xz ¼ σ yz ¼ σ zz ¼ 0). In the last case use
matrix multiplication (*) and transpose (’) to compute the rotated stress matrix
SIGMArotated ¼ Rmat SIGMA Rmat’:
Use the analytical solution as a reference to evaluate error for the other stress rotation
methods. An example is in Stress_rotation.m.
Exercise 12.6.
Combine the previous exercise with stress advection and rotation on markers in a squared
1 × 1 m2 model area. Define a rotational vx = ω(0.5 − y), vy = ω(x − 0.5), ω ¼ 1 (Exercise 4.1,
Fig. 4.4) velocity field around the model centre on a regular staggered 51×51 gridwith 4 × 4
randomly distributed markers per cell. Compute ωyxði;jÞ ¼ 12 ∂vy =∂x ∂vx =∂y for basic
nodal points using finite differences (i.e., similarly to ε_ xyði;jÞ in Eq. (9.18), Fig. 9.3a). Define
deviatoric stress components on the markers. Move markers located within 0.5 m distance
from the model centre by the prescribed nodal velocity field (Exercise 8.6) and rotate their
stresses using ωyx values interpolated from the grid at each time step. Trace the evolution of
the changes in coordinates and stresses for a selected marker. Test different stress rotation
approaches. An example is in Stress_rotation_markers.m.

Chp.12.Elasticity and Plasticity 2019

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Chp.12.Elasticity and Plasticity 2019

Uploaded by

Copyright:

Available Formats

12

Elasticity and plasticity

Theory: Elastic rheology. Rotation of elastic stresses. Maxwell visco-elastic

12.1 Why should we care about elasticity and plasticity?

12.2 Elastic rheology

σ ij ¼ λδij εkk þ 2μεij ; ð12:2Þ

εkk ¼ εxx þ εyy þ εzz ; ð12:4Þ

12.3 Rotation of elastic stresses

fx ¼ fx1 þ fx2 þ fx3 þ fx4 ¼ 0; ð12:12Þ

jACj ¼ jABj cosðθÞ;

fx1 ¼ σ x0 x0 jABj cosðθÞ;

fx2 ¼ σy0 x0 jABj sinðθÞ;

fx3 ¼ σ xx jACj ¼ σ xx jABj cosðθÞ;

fy1 ¼ σ x0 x0 jABj sinðθÞ;

fy2 ¼ σ y0 x0 jABj cosðθÞ;

fy3 ¼ σ yx jACj ¼ σ yx jABj cosðθÞ;

fy4 ¼ σ yy jBCj ¼ σ yy jABj sinðθÞ;

By multiplying Eq. (12.14) by cos(θ) and Eq. (12.15) by sin(θ), we obtain

σ x0 x0 cos2 ðθÞ þ σ y0 x0 sinðθÞcosðθÞ σ xx cos2 ðθÞ þ σ xy sinðθÞcosðθÞ ¼ 0; ð12:16Þ

σ x0 x0 sin2 ðθÞ þ σ y0 x0 cosðθÞsinðθÞ σ yx cosðθÞsinðθÞ þ σ yy sin2 ðθÞ ¼ 0: ð12:17Þ

By subtracting Eq. (12.17) from Eq. (12.16) we get

which can be further simpliﬁed to

σ x0 x0 ¼ σ xx cos2 ðθÞ þ σ yy sin2 ðθÞ σ xy sinð2θÞ ð12:18Þ

σ y0 y0 ¼ σ xx sin2 ðθÞ þ σ yy cos2 ðθÞ þ σxy sinð2θÞ; ð12:20Þ

σ_ ijðJaumannÞ ¼ ωik σ kj σ ik ωkj ; ð12:29Þ

In 2D, Eqs. (12.31)–(12.33) are equivalent to previously derived Eqs. (12.24)–(12.27)

(2) Compute the unit rotation vector ~

(3) Integrate the incremental rotation angle:

θ ¼ ωmag Δt: ð12:40Þ

(4) Evaluate the rotation matrix using the Euler–Rodrigues formula:

12.4 Maxwell visco-elastic rheology

12.5 Plastic rheology

for dry rocks

and for ﬂuid-bearing rocks

Pt ¼ ð1 ϕÞPs þ ϕPf ; ð12:52Þ

σPeierls. This deformation mechanism is very important, in particular for deformation of

12.6 Visco-elasto-plastic rheology

χ ¼ 2ε_ IIðplasticÞ ; ð12:61Þ

vÞ ¼ 2 sinðψÞε_ IIð plasticÞ ;

SIGMArotated ¼ Rmat  SIGMA  Rmat’:

You might also like

SIGMArotated ¼ Rmat SIGMA Rmat’: