In continuum mechanics an eigenstrain is any mechanical deformation in a material that is not caused by an external mechanical stress, with thermal expansion often given as a familiar example. The term was coined in the 1970s by Toshio Mura, who worked extensively on generalizing their mathematical treatment.[1] A non-uniform distribution of eigenstrains in a material (e.g., in a composite material) leads to corresponding eigenstresses, which affect the mechanical properties of the material.[2]
Many distinct physical causes for eigenstrains exist, such as crystallographic defects, thermal expansion, the inclusion of additional phases in a material, and previous plastic strains.[3] All of these result from internal material characteristics, not from the application of an external mechanical load. As such, eigenstrains have also been referred to as “stress-free strains”[4] and “inherent strains”.[5] When one region of material experiences a different eigenstrain than its surroundings, the restraining effect of the surroundings leads to a stress state on both regions.[6] Analyzing the distribution of this residual stress for a known eigenstrain distribution or inferring the total eigenstrain distribution from a partial data set are both two broad goals of eigenstrain theory.
Eigenstrain analysis usually relies on the assumption of linear elasticity, such that different contributions to the total strain
\epsilon
\epsilon*
\epsilonij=eij+
* | |
\epsilon | |
ij |
where
i
j
Another assumption of linear elasticity is that the stress
\sigma
e
Cijkl
\sigmaij=Cijklekl
In this form, the eigenstrain is not in the equation for stress, hence the term "stress-free strain". However, a non-uniform distribution of eigenstrain alone will cause elastic strains to form in response, and therefore a corresponding elastic stress. When performing these calculations, closed-form expressions for
e
\epsilon*
One of the earliest examples providing such a closed-form solution analyzed a ellipsoidal inclusion of material
\Omega0
\Omega
\Omega0
\Omega0
\Omega
\Omega0
\Omega
\Omega0
The solutions for the total stress and strain within
\Omega0
\epsilonij=Sijkl
* | |
\epsilon | |
kl |
\sigmaij=Cijkl(\epsilonij-
*) | |
\epsilon | |
kl |
Where
S
\Omega0
\Omega0
S
Eigenstrains and the residual stresses that accompany them are difficult to measure (see:Residual stress). Engineers can usually only acquire partial information about the eigenstrain distribution in a material. Methods to fully map out the eigenstrain, called the inverse problem of eigenstrain, are an active area of research. Understanding the total residual stress state, based on knowledge of the eigenstrains, informs the design process in many fields.
Residual stresses, e.g. introduced by manufacturing processes or by welding of structural members, reflect the eigenstrain state of the material. This can be unintentional or by design, e.g. shot peening. In either case, the final stress state can affect the fatigue, wear, and corrosion behavior of structural components.[7] Eigenstrain analysis is one way to model these residual stresses.
Since composite materials have large variations in the thermal and mechanical properties of their components, eigenstrains are particularly relevant to their study. Local stresses and strains can cause decohesion between composite phases or cracking in the matrix. These may be driven by changes in temperature, moisture content, piezoelectric effects, or phase transformations. Particular solutions and approximations to the stress fields taking into account the periodic or statistical character of the composite material's eigenstrain have been developed.
Lattice misfit strains are also a class of eigenstrains, caused by growing a crystal of one lattice parameter on top of a crystal with a different lattice parameter.[8] Controlling these strains can improve the electronic properties of an epitaxially grown semiconductor.[9] See: strain engineering.