Introduction to Rock Mechanics

CE - 425

Lecture 6
3 dimensional state of stress
Tri-axial properties
Failure Theories
Failure Criteria

Dr. Rana Muhammad Asad Khan

masadkhan87@gmail.com
03314438163
1
To learn
Tri-axial properties and 3-Dimensional stress state
Elastic Constants of a material especially rocks
Failure theories
Failure criteria in rocks

2
Stress Tensor
In a matrix
Rows representing
the components
on any plane

Columns representing the

components acting in any given
direction

3
Stress Tensor
• 9 components of which 6 are independent
• Values which are point properties
• Values which depend on orientation relative to a set
of reference axes
• 6 of the 9 components becoming zero at a particular
orientation
• Three principal components
• Complex data reduction requirements because two or
more tensors cannot, in general, be averaged by
averaging the respective principal stresses

4
Stress as a point Property
• Because the acting forces will vary
according to the orientation of ΔA within
the slice
• It is most useful to consider the normal
stress (ΔN/ΔA) and shear stress
(ΔS/ΔA) as the area ΔA become very
small, eventually approaching zero.
• Limitations in reducing the
size of the area to zero
∆𝑁
𝑛𝑜𝑟𝑚𝑎𝑙 𝑠𝑡𝑟𝑒𝑠𝑠 𝜎𝑛 = lim
• It is important to realize that ∆𝐴→0 ∆𝐴
the stress components are
defined as mathematical ∆𝑆
𝑆ℎ𝑒𝑎𝑟 𝑠𝑡𝑟𝑒𝑠𝑠 𝜏 = lim
quantities, with the result that ∆𝐴→0 ∆𝐴
stress is a point property
5
Stress components on a
infinitesimal cube
• For convenience, the shear and normal components of
stress may be resolved with reference to a given set of
axes
• A rectangular Cartesian X-Y-Z system.
• The body can be considered to be cut at three orientations
corresponding to the visible faces of a cube.

6
Stress tensor convention
• 9 stress components comprising of 3
normal and 6 shear.
• Standard convention for denoting
these components is that the first
subscript refers the plane on which
stress component acts Negative face is the
• Second subscript denotes the one in which the
direction in which it acts outward normal to the
• Normal stress, compression is face points in the
positive negative direction
• For shear stress, positive stress acts
in positive direction on negative face

In geotechnical engineering: compression positive and tension negative

7
Symmetry in stress matrix
• Although there are 9 stress components in the stress
matrix, assumption is that body is in equilibrium
• By inspecting, a resultant moment of zero
• Then shear stresses opposite form one another must be
equal in magnitude
• Thus, considering the equilibrium around x, y, and z axes
𝜏𝑥𝑦 = 𝜏𝑦𝑥
𝜏𝑦𝑧 = 𝜏𝑧𝑦
𝜏𝑥𝑧 = 𝜏𝑧𝑥

Stress matrix is symmetrical about the leading diagonal

8
Symmetry in stress matrix
• Although there are 9 stress components in the stress
matrix, assumption is that body is in equilibrium
• By inspecting, a resultant moment of zero
• Then shear stresses opposite form one another must be
equal in magnitude
• Thus, considering the equilibrium around x, y, and z axes
𝜏𝑥𝑦 = 𝜏𝑦𝑥
𝜏𝑦𝑧 = 𝜏𝑧𝑦
𝜏𝑥𝑧 = 𝜏𝑧𝑥

Stress matrix is symmetrical about the leading diagonal

State of stress at a point is defined completely by six independent
components (3 normal and 3 Shear)

9
Symmetry in stress matrix
Scalar quantity can be specified
completely by 1 value
Vector by 3 values
Tensor requires 6 values
Always there exist 6 independent pieces
of information
6 stress components in stress matrix for a given body
subjected to loading will depend on the orientation of the cube
in the body itself

If cube is rotated, it should be possible to find the directions in

which the normal stress components take on maximum and
minimum values

10
Symmetry in stress matrix
Shear component on all faces of
cube becomes zero

The Principal stresses are

defined as those normal
components of stress that act on
planes that have shear stress
components with zero magnitude

11
Stresses transformation
σ‘x=(σx+σy)/2+Cos(2θ)(σx-σy)/2+ τ xySin(2θ)

σ‘y=(σx+σy)/2-Cos(2θ)(σx-σy)/2- τ xySin(2θ)

τ'xy=-Sin(2θ)(σx-σy)/2+τxyCos(2θ)

12
Example #1
The state of stress on a soil
specimen is σv = 100 kPa, σh
= 50 kPa.
A shear stress of 30 kPa is
applied to the vertical and
horizontal planes, as shown
in the figure.
Calculate the following:
• Major and minor principal stresses and the orientation of
the planes on which they act. Determine the angle of
rotation of the principal planes.
• Maximum shear stress
• Stresses on a plane inclined 30 degrees clockwise from the
horizontal plane.
13
 Example #1
𝜎𝑥 + 𝜎𝑦 𝜎𝑥 − 𝜎𝑦 2
𝜎 𝑚𝑎𝑥 = + ( + 𝜏𝑥𝑦 2)
2 2

𝜎 𝑚𝑎𝑥 =114.05 KPa

𝜎𝑥 + 𝜎𝑦 𝜎𝑥 − 𝜎𝑦 2
𝜎 𝑚𝑖𝑛 = − + 𝜏𝑥𝑦 2)
2 2

𝜎 𝑚𝑖𝑛 =35.95 KPa

𝜏𝑥𝑦
tan 2∅ = 2 ∗
𝜎𝑥 − 𝜎𝑦
This negative sign is indication for the anti-
∅= -25.1o clock wise rotation in Mohr Column graph.14
Example #1
𝜎 𝑥 − 𝜎𝑦 2 2
𝜏𝑚𝑎𝑥 = + 𝜏𝑥𝑦
2

𝜏𝑚𝑎𝑥=39.05 KPa

𝜎𝑥 + 𝜎𝑦 𝜎𝑥 − 𝜎𝑦
𝜎 ′x = + 𝑐𝑜𝑠2𝜃 ∗ + 𝜏𝑥𝑦 ∗ 𝑆𝑖𝑛2𝜃
2 2

𝜎 ′ 𝑥 = 36.52 KPa
𝜎𝑥 + 𝜎𝑦 𝜎𝑥 − 𝜎𝑦
𝜎 ′y = − 𝑐𝑜𝑠2𝜃 ∗ − 𝜏𝑥𝑦 ∗ 𝑆𝑖𝑛2𝜃
2 2

113.48 Kpa
15
Generalized Hook’s Law
𝜎
𝐸=
𝜀

If a stress σx acts, it
will cause an increase
in x direction whereas
the dimensions in y
and z will decrease
𝜎𝑥
Ɛx =
𝐸
𝜎𝑥
Ɛy = -μ Ɛx 𝜀 y=-μ
𝐸
Ɛz = -μ Ɛx 𝜀z=-μ
𝜎𝑥
𝐸
16
Generalized Hook’s Law
If a stress σy acts, it will cause an increase
in y direction whereas the dimensions in x
and z will decrease
𝜎𝑦
Ɛy = 𝜎𝑦
𝐸
Ɛx = -μ Ɛy 𝜀x=-μ
𝐸
𝜎𝑦
Ɛz = -μ Ɛy 𝜀z=-μ
𝐸
If a stress σz acts, it will cause an increase
in z direction whereas the dimensions in x
and y will decrease
𝜎𝑧
Ɛz =
𝐸
Ɛx = -μ Ɛz
Ɛy = -μ Ɛz
17
Generalized Hook’s Law
Total strain in any one direction Hook’s law for the individual
will be shear strains
𝜎𝑥 𝜇
𝜀𝑥 = − 𝜎𝑦 + 𝜎𝑧 𝝉𝒙𝒚
𝐸 𝐸 𝛾𝑥𝑦 =
𝐺
𝜎𝑦 𝜇
𝜀𝑦 = − 𝜎𝑥 + 𝜎𝑧 𝝉𝒚𝒛
𝐸 𝐸
𝛾𝑦𝑧 =
𝜎𝑧 𝜇 𝐺
𝜀𝑧 = − 𝜎𝑥 + 𝜎𝑦
𝐸 𝐸 𝝉𝒛𝒙
𝛾𝑧𝑥 =
𝐺
These equations are the generalized Hook’s Law and are
constitutive equations for linear elastic isotropic materials

18
 Generalized Hook’s Law
To an Engineer Plane strain conditions are of much relevance
𝜎𝑧 = 𝜏𝑥𝑧 = 𝜏𝑦𝑧 = 0
𝜎𝑥 𝜇𝜎𝑦
𝜀𝑥 = −
𝐸 𝐸
𝜎𝑦 𝜇𝜎𝑥
𝜀𝑦 = −
𝐸 𝐸
𝜇𝜎𝑥 𝜇𝜎𝑦
𝜀𝑧 = − −
𝐸 𝐸
All the engineering materials are not linear elastic isotropic.

19
 Elastic Constants and their
relationships
E Young’s modulus of rigidity = stress/strain

G= Shear Modulus or Modulus of rigidity = Shear stress / Shear strain

μ = Possion's ratio = lateral strain / longitudinal strain

K = Bulk Modulus of elasticity = Volumetric stress / Volumetric strain

𝐸 = 2𝐺(1 + 𝜇) 𝐸 = 3𝐾(1 − 2𝜇)

9𝐺𝐾
𝐸= 𝜇 = 0.5(3𝐾 − 2𝐺)(𝐺 + 3𝐾)
(3𝐾 + 𝐺)
20
THEORIES OF ELASTIC FAILURE
Stress

a. Maximum Principal stress theory (due to Rankine)

b. Maximum shear stress theory (Guest - Tresca)

Strain
c. Maximum Principal strain (Saint - venant) Theory

d. Total strain energy per unit volume (Haigh) Theory

e. Shear strain energy per unit volume Theory (Von –

Mises & Hencky)

21
Maximum Principal stress theory
When the maximum principal stress in a
complex stress system reaches the elastic
limit stress (max normal strength) in a
simple tension, failure will occur
Good for brittle materials
Therefore, the criterion for failure would be
σ1 = σyp (σyp is the yield point)
For a two dimensional complex stress
system σ1 is expressed as
𝜎𝑥 + 𝜎𝑦 1 2
𝜎𝑦𝑝 = 𝜎1 = + 𝜎𝑥 − 𝜎𝑦 + 4𝜏𝑥𝑦2
2 2
22
Maximum shear stress theory
Failure can be assumed to occur when the maximum
shear stress in the complex stress system is equal to the
value of maximum shear stress in simple tension
(maximum shear strength exceeds maximum allowable
shear stress)
𝜎𝜃 = 𝜎𝑦 sin2 𝜃
1
𝜏𝜃 = 𝜎𝑦 sin 2𝜃
2
𝑚
1
𝜏𝜃|𝑚𝑎𝑥 = 𝜎𝑦
2
For two dimensional complex stress system

𝑚
(𝜎1 − 𝜎2)
𝜏𝑚𝑎𝑥 = σ1 is the maximum principle stress
2
σ3 is the minimum principle stress
23
Maximum shear stress theory
𝜎1 − 𝜎2 1 2
= 𝜎𝑥 − 𝜎𝑦 + 4𝜏2𝑥𝑦
2 2
𝜎1 − 𝜎2 1
= 𝜎𝑦𝑝 = 𝜎1 − 𝜎2 = 𝜎𝑦
2 2

𝜎𝑦𝑝 = 𝜎𝑥 − 𝜎𝑦 2 + 4𝜏2𝑥𝑦

Becomes the criterion for failure

24
Maximum Principal strain theory
Failure occurs when the maximum strain for a complex
state of stress system becomes equals to the strain at
yield point in the tensile test
Not recommended

For a 3 - dimensional state of stress system the total

strain energy Ut per unit volume in equal to the total
work done by the system and given by the equation

1 1 1
𝑈𝑡 = 𝜎𝜖 + 𝜎2𝜖2+ 𝜎3𝜖3
2 1 1 2 2

25
Maximum Principal strain theory
1
𝜖1 = 𝜎1 − 𝛾(𝜎2 + 𝜎3)
𝐸
1
𝜖2 = 𝜎2 − 𝛾(𝜎1 + 𝜎3)
𝐸
1
𝜖3 = 𝜎3 − 𝛾(𝜎1 + 𝜎2)
𝐸

Thus the failure criterion becomes

𝜎1 𝜎2 𝜎3 𝜎𝑦𝑝
−𝛾 −𝛾 =
𝐸 𝐸 𝐸 𝐸

26
Total strain energy per unit volume
theory
Failure occurs when the total strain energy for a
complex state of stress system is equal to that at the
yield point a tensile test.
when the strain energy per unit volume of any portion
of the stressed member reaches the failure value of
strain energy per unit volume
1 𝜎2𝑓
𝜎12 + 𝜎22 + 𝜎32 − 2𝛾(𝜎1𝜎2 + 𝜎2𝜎3 + 𝜎3𝜎1 =
2𝐸 2𝐸

This theory gives fair good results for ductile material

(rocks)

27
Maximum shear strain energy per unit
volume theory
Failure occurs when the maximum shear strain energy
component for the complex state of stress system is
equal to that at the yield point in the tensile test
when the strain energy per unit volume due to shear of
any portion of the stressed member reaches the failure
value of strain energy per unit volume
2
1 𝜎
𝜎1 − 𝜎2)2 + (𝜎2 − 𝜎3)2 + (𝜎3 − 𝜎1)2 = 𝑓
12𝐺 6𝐺

G is the modulus of rigidity, failure criteria becomes

𝜎1 − 𝜎2)2 + (𝜎2 − 𝜎3)2 + (𝜎3 − 𝜎1)2 = 2𝜎2𝑓

28
Maximum shear strain energy per unit
volume theory
General state of stress
i) Hydrostatic (volumetric strain)
ii) Distortional /deviatoric stress (shear strain energy)
Strian energy due to distortion is
1
Udistortion= 𝜎1 − 𝜎2)2 + (𝜎2 − 𝜎3)2 + (𝜎3 − 𝜎1)2
12𝐺

Distortion energy should be equal to maximum distortion

energy in simple tension test, this happens when one
principal stress reaches yield point (σyp)of material
𝜎1 − 𝜎2)2 + (𝜎2 − 𝜎3)2 + (𝜎3 − 𝜎1)2 = 2𝜎2𝑓

29
Maximum shear strain energy per unit
volume theory
Thus, σ2=σ3=0 distortion energy for simple test
2𝜎12
𝑈𝑑 =
12𝐺
σ1- σyp
𝜎𝑦𝑝2
𝑈𝑑 =
𝐺

30
Conclusions
• Materials does not fail under hydrostatic stress
system i.e σ1 = σ2 = σ3
• None of the theories agrees with the test
perform for all types of materials and
combinations of loads.
• There is a good agreement between the
maximum distortion energy theory and
experimental result for ductile materials.
• The max. principal stress theory appears to be
the best for brittle materials
31
Conclusions
• Max. shear stress or max. strain energy
theories give the good approximation for
ductile materials but the max. shear stress
criterion is somewhat more conservative.
• The max. strain theory should not be used in
general as it only gives the reliable results in
particular cases.
• If the brittle material has a stress strain
diagram, that is different in tension and
compression, then the MOHR’S Failure
Criterion may be used to predict the failure.

32
Mohr-coulomb criterion
The simplest and best-known failure criterion is
given
𝜏 = 𝑐 ′ + 𝜎𝑛 tan ∅′
where, τ is shear strength, c’ is cohesion; σ’ is
internal friction.
′ 2
∅
𝜎1 = 𝜎𝑐 + 𝜎3′ tan 45 +
2
In terms of effective principal stresses, this leads to
a linear relationship,
where, σc is uniaxial compressive strength, and 'φ is
angle of internal friction.
The classical Mohr-Coulomb theory can't be used to
predict the non-linear response of rocks. 33
 Mohr-coulomb criterion
Keeping in mind this limitation, a number of investigators suggested
empirical strength of various forms suitable to predict the non-linear
response of jointed rocks.

Tension cut-off
is the limit of
load carrying
capacity of the
Mohr-Coulomb
plasticity model
near the tensile
region

34
 Mohr-coulomb criterion
τ Failure envelop τ Failure envelop

K=tanφ 1 + sin 𝜑
𝑘=
Tension Tension 1 + sin 𝜑
Cut off Cut off σz

σ σ
σt

35
Mohr-coulomb criterion

36
 Mohr-coulomb criterion
Effect of water pressure and principle stress ratio

Tarzaghi's effective stress law

“a pressure of U as the pore
water in rock will reduce the
peak (Total) normal stress by an
amount U”
σ’ = σ –u

• Reduction in strength upto 15 % in the presence of water

in sandstone
• Complete loss of strength in Clay shale
• If drainage is blocked, pore water and fissure water exert
pressure 37
Hoek and Brown criterion
Hoek and Brown gave an empirical failure criterion
capable of modelling the highly non-linear relationship
between the minor and major principal stresses and also
predicting the influence of rock mass quality on the
strength, this criterion is given as
𝜎 ′ 1𝑓 = 𝜎 ′ 3𝑓 + 𝑚𝜎𝑐 𝜎 ′ 3𝑓 + 𝑠𝜎𝑐2

Where, σ’1f and σ’3f are the major and minor effective
principal stresses at failure.
σc is the uniaxial compressive strength of the intact rock
material,
and m and s are material constants,
where s = 1 for intact rock.
38
Hoek and Brown criterion
Later, Hoek and Brown (2002) have modified the
equation to give a generalized criterion in which
the shape of the principal stress plot or the Mohr
envelope could be adjusted by means of a
variable coefficient ‘a’ in place of the square root
term, which is as given below.

𝜎 ′3
𝜎 ′ 1𝑓 = 𝜎 ′ 3𝑓 + 𝜎𝑐 𝑚 +𝑠 𝑎
𝜎𝑐

39
Comparison of Hoek-Brown and Mohr-Coulomb
criteria

40
Griffith Failure criterion
• A crystalline material (like rock) contain randomly
oriented zones of potential failure in the form of
grain boundaries (microfractures).
• Stress concentrations at the end of these cracks
causing the crack to propagate and ultimately
contribute to the macroscopic failure.
1. The flaw (gap) which is elliptical in shape can be treated
as single ellipse in a semi-infinite elastic medium.
2. Adjacent flaws don't interact.
3. The material is assumed to be homogeneous.
4. Ellipse and stress system are taken to be two
dimensional.
41
Griffith Failure criterion
For a thin elastic strip of unit thickness containing a
elliptical hole oriented with its long axis perpendicular to an
applied tensile stress σo, the maximum stress σmax at the
apex of the ellipse depends on radius of curvature of the
apex (ρ) and the length of the crack 2c

42
Griffith Failure
criterion

43
Griffith Failure criterion
𝑐
𝜎𝑚𝑎𝑥 = 2𝜎𝑜
𝑝

Difference in energy with and without hole to be Total

energy
𝜋𝑐2𝜎𝑜
𝑊𝑒 =
𝐸
E is the young modulus
The surface energy resulting from the formation of crack is
𝑊𝑠 = 4𝑐𝑇
T is te surface tension
Hence the eliptocal hole has decreased the total energy
𝜋𝑐2𝜎𝑜
W= 𝑊𝑒 − 𝑊𝑠 = − 4𝑐𝑇
𝐸

44
Griffith Failure criterion

45
Empirical Rock failure criterion
More precise criterion of failure may be determined for any
rock by fitting an envelope to Mohr's circles representing
values of the principal stresses at peak conditions in
laboratory tests.

46