On the inelastic behavior of solids. I: Twinning (Q1905451)

To model the ``inelastic'' behaviour of certain materials like metals, polymers, and shape memory alloys, when they are subjected to large deformations, it is necessary to take into account several physical mechanisms determining the corresponding microstructures and to allow for the possibility of distinct preferred configurations associated with them. A general theory of the ``inelastic'' behaviour of materials is based on the idea that the stress of the material is determined by a functional depending on the history of a one-parameter family of deformation gradients computed with respect to a one-parameter family of preferred reference configurations. The authors consider here the simplified theory, which in fact captures the essential features of the general theory capable of describing the microstructural changes, when the Cauchy stress tensor \(\underset \sim T\) of the material is given by a constitutive relation of the form \((*)\) \({\underset \sim T} = \widehat {{\underset \sim T}} {_{\kappa_0, \kappa_n}} ({\underset \sim F}{_{\kappa_0}}, {\underset \sim F} {_{\kappa_n} })\). Here \(\kappa_0\) is the reference configuration associated with the initial microstructure of the material, \(\kappa_n\) is new reference configuration corresponding to a new possible microstructure of the material appearing at a certain stage of its deformation, and \({\underset \sim F}_{\kappa_0}\), \({\underset \sim F}_{\kappa_n}\) are the deformation gradients with respect to the initial and the new configurations. A constitutive assumption of this theory is the existence of an a priori given indexed set \(\underset \sim S\) containing all the possible reference configurations of the material, and which is in a one-to-one correspondence with all of its possible microstructures. According to this theory, the material response with respect to each configuration in \(\underset \sim S\) is that of a solid, while its overall response cannot be classified either as that of a simple solid or a simple fluid. The temporal evolution of the material is well determined as soon as the temporal evolution of \(\kappa_n\) is given or, equivalently, the mapping from \(\kappa_0\) to \(\kappa_n\) is specified. The presence of subscripts \(\kappa_0\) and \(\kappa_n\) in \((*)\) highlights the dependence of the stress tensor upon the choice of the set \(\underset \sim S\) of reference configurations. To delineate the range of deformation for which there is no change in microstructure of the material and to quantify the boundary of the elastic region corresponding to each microstructure \(\kappa_n\), the authors introduce the concept of the activation function of the material which is denoted by \(g({\underset \sim F} {_{\kappa_0}}, {\underset \sim F}{_{\kappa_n}})\). The so-called activation criterion, stipulating that \(\kappa_n\) remains unchanged as long as \(g({\underset \sim F} {_{\kappa_0} }, {\underset \sim F} {_{\kappa_n}}) \leq 0\), determines the instant at which the material undergoes a change in microstructure. For any fixed \(\kappa_n\), the equation \(g( {\underset \sim F}{_{\kappa_0}}, {\underset \sim F}{_{\kappa_n}}) = 0\) represents a hypersurface in the nine-dimensional space of deformation gradients which is called the activation surface. The closed region inside the activation surface represents the elastic domain associated with the microstructure \(\kappa_n\), and no microstructural changes take place as long as the deformation gradient is within activation surface. An analysis is made of the dependence of the functional form of the stress \(\widehat {{\underset \sim T}} {_{\kappa_0, \kappa_n}}\) and of the activation function on the choice of the set \(\underset \sim S\) of the reference configurations representing the possibile microstructures of the material, and the consequences of the invariance requirements under superposed rigid motions that must be satisfied by \(\widehat {{\underset \sim T}} {_{\kappa_0, \kappa_n}}\) and \(g\) are derived. A minute examination of the special constitutive equations for transitions between two microstructure is made, and the problem of discontinuous twinning is studied in detail. This very interesting and important paper may be considered as an extension and refinement of the results regarding the ``inelastic'' behaviour of polymeric materials, obtained by \textit{K. R. Rajagopal} and \textit{A. S. Wineman} [Int. J. Plast. 8, 385-395 (1992; Zbl 0765.73005)].