Unit 2

Orthogonal Matrix
To transform the stress tensor from one Cartesian coordinate system to another, we use a mathematical tool
called an orthogonal transformation. Let's consider an original coordinate system with axes x1,x2,x3 (equivalent
to x, y, z) and a new, rotated coordinate system with axes x1′,x2′,x3′ (equivalent to x', y', z'). The orientation of the
new system relative to the old one can be defined by a set of direction cosines.
A direction cosine, denoted as lmi, is the cosine of the angle between the m-th new axis (xm′) and the i-th old axis
(xi). For example, l12 is the cosine of the angle between the new x1′ axis and the old x2 axis. These nine direction
cosines can be arranged in a 3x3 matrix.
Transformation of Stress: The Direction Cosine Matrix

The stress tensor, being a second-order tensor, transforms according to a specific rule. If [σ] is the stress
tensor in the original coordinates and [σ′] is the stress tensor in the new, rotated coordinates, the
transformation law is given by :
[σ′]=[L][σ][L]T
2D Transformation (Plane Stress):

In many engineering problems, the state of stress is simplified. A very common and important case is plane
stress. This condition occurs in thin plates loaded only in the plane of the plate. For a plate in the x-y plane,
the stresses acting in the z-direction are assumed to be zero :

σzz=τxz=τyz=0

Under these conditions, the 3D transformation problem simplifies significantly. We are typically interested in
transforming stresses within the x-y plane. The transformation is defined by a single angle of rotation, θ,
measured counter-clockwise from the x-axis to the new x'-axis.
3D Stress Transformation Equations :
Principal Stresses & Stress Invariants
principal stresses-denoted as σ1,σ2, and σ3. By convention, they are ordered such that σ1≥σ2≥σ3.
These represent the maximum, intermediate, and minimum normal stresses at the point.
The physical condition for a principal plane can be stated as follows: we are searching for a specific
plane, defined by its unit normal vector n, where the traction vector T acting on that plane is purely
normal. This means the traction vector T must be parallel to the normal vector n. We can express this
condition mathematically as:
T=λn

Here, λ is a scalar magnitude representing the principal stress on that plane.

From Cauchy's stress theorem, we know that the traction vector is related to the stress tensor σ by T=σn.
Combining these two equations gives:
σn=λn
This equation is a cornerstone of linear algebra and represents an eigenvalue problem. It can be
rearranged into its standard homogeneous form by introducing the identity tensor I :
σn−λIn=0

(σ−λI)n=0
This formulation reveals a profound connection:
1.The principal stresses (σ1,σ2,σ3) are the eigenvalues (λ) of the stress tensor σ.
2. The principal directions (the unit normal vectors n to the principal planes) are the corresponding eigen
vectors of the stress tensor σ.
3.Because the stress tensor is a real and symmetric matrix, it is guaranteed to have three real eigenvalues
(the principal stresses) and three mutually orthogonal eigenvectors (the principal directions).

The Characteristic Equation and Stress Invariants

The eigenvalue problem (σ−λI)n= 0 is a system of three linear homogeneous equations for the components of
the eigenvector n. For this system to have a non-trivial solution (i.e., for n not to be a zero vector), the
determinant of the coefficient matrix must be zero :
det(σ−λI)=0
In matrix form, this is written as:

The three roots of this cubic equation are the three

principal stresses, σ1,σ2,σ3. The coefficients I1,I2,
and I3 are known as the stress invariants.
First Stress Invariant (I1): The sum of the normal stresses on the diagonal, also known as the trace of the stress
tensor.

Second Stress Invariant (I2): The sum of the principal minors of the stress matrix.

Third Stress Invariant (I3): The determinant of the stress tensor.

The concept of invariance is profound. It means that while the six numbers we use to describe stress (σxx,τxy,
etc.) are just a convenient but arbitrary representation, the three invariants (I1,I2,I3) are fundamental, intrinsic
properties of the stress state at that point. No matter how we rotate our axes, these three values remain constant.
Special Stress States and Decomposition
Now, lets break down the general stress tensor into physically meaningful components that are directly
used in theories of plasticity and material failure.
Hydrostatic and Deviatoric Stress Tensors
A general state of stress can be thought of as a superposition of two simpler, physically distinct states: one that
causes a change in volume and one that causes a change in shape (distortion). This conceptual split is formalized
by decomposing the stress tensor into a hydrostatic (or spherical) component and a deviatoric component.

Deviatoric Stress: The deviatoric stress tensor, denoted sij or σij′, is what remains after the hydrostatic
component is subtracted from the total stress tensor :
Octahedral Stresses
Another way to characterize the stress state using invariant quantities is through octahedral stresses. These are
the normal and shear stresses that act on a specific set of planes called the octahedral planes.
An octahedral plane is defined as a plane whose normal vector makes equal angles with each of the three principal
stress axes. If we align our coordinate system with the principal directions (1, 2, 3), there are eight such planes.

The unit normal vector n to any of these planes has components whose absolute values are equal: ∣n1∣=∣n2∣=∣n3∣.
Since the vector must have unit length (n1^2+n2^2+n3^2=1), the direction cosines are:

Octahedral Normal Stress (σoct): The normal stress on an octahedral plane can be derived using the general
formula for normal stress on an arbitrary plane. In the principal coordinate system, this simplifies significantly.
The resulting octahedral normal stress is found to be exactly equal to the mean (hydrostatic) stress :
Strain, Constitutive Laws, and Compatibility
This part transitions from the study of forces (stress) to the study of deformation (strain) and connects the two via
the material's constitutive law.

The Concept of Strain at a Point

When a body is subjected to stress, it deforms. Strain is the geometric measure of this deformation. It is a
dimensionless quantity that describes the relative change in size or shape of the material. Just as stress
describes the internal force state at a point, strain describes the local deformation state at a point.

There are two fundamental types of strain :

Normal Strain (ϵ): This measures the change in length per unit length of a line element. It represents stretching
(tensile strain, positive) or contraction (compressive strain, negative). For a line element of original length L0 that
deforms to a new length Lf, the engineering normal strain is:
• Shear Strain (γ): This measures the change in the angle between two line elements that were initially
perpendicular. It represents distortion or a change in shape. It is defined as the change in the right
angle, measured in radians

The diagonal terms (ϵxx,ϵyy,ϵzz) are the normal strains in the x, y, and z directions. The off-diagonal terms
(ϵxy,ϵyz,ϵzx) are the tensorial shear strains.

A crucial point of convention must be addressed here. The engineering shear strain, γij, which represents
the total change in angle, is related to the tensorial shear strain, ϵij, by a factor of two :
Generalized Hooke's Law for Isotropic Materials
The relationship between stress and strain is a material property, described by a constitutive law. For many
common engineering materials operating within their elastic limit, this relationship is linear. For materials that are
also isotropic (meaning their properties are the same in all directions), this linear relationship is known as the
Generalized Hooke's Law.

We can derive this law using the principle of superposition. Consider a normal stress σxx applied in the x-
direction. It produces a primary normal strain in the x-direction and, due to the Poisson's effect, secondary
contractile strains in the y and z directions :

Here, E is Young's Modulus (Modulus of Elasticity), and ν is Poisson's Ratio.

If we now consider a general 3D state of stress with σxx, σyy, and σzz acting simultaneously, the total strain
in each direction is the sum of the strains caused by each stress component. This superposition gives the
normal strain equations :

For an isotropic material, the shear components are uncoupled from the normal components. A shear stress only
produces its corresponding shear strain :