Second gradient theory of thermopiezoelectricity (Q6597692)

A theory of multipolar continuum mechanics mainly developed by \textit{A. E. Green} and \textit{R. S. Rivlin} [Arch. Ration. Mech. Anal. 17, 113--147 (1964; Zbl 0133.17604)] from contributions due to Toupin, Mindlin and Ericksen, is extended to the coupled three-dimensional dynamic theory of thermopiezoelectricity. The electic fields are assumed quasi-static while the second gradients of displacement, the electric potential and the temperature are included as independent constitutive variables. The other fundamental postulates are a rate of work equation, an entropy production inequality, and invariance under superposed rigid body motions. This enables the balance laws to be derived in the manner of \textit{A. E. Green} and \textit{R. S. Rivlin} [loc. cit.]. The discussion, conducted with respect to fixed Cartesian coordinates, requires the introduction of numerous quantities represented by an extensive notation that is somewhat simplified by a judicious grouping of terms. Practical applications are cited to motivate the investigation.\N\NSubject to the displacement, the temperature and the electric potential being small, the linear constitutive relations are introduced which imply that the enthalpy function may be expressed as a quadratic form. The stipulation of initial and boundary conditions complete the formulation of the linear initial boundary value problem. The uniqueness of the classical solution to this problem is established by a careful generalisation of Neumann's proof (1885). Consequently, the dissipation inequality as well as certain positive-definite assumptions used in the proof become sufficient for the uniqueness.\N\NElements of spectral theory and of algebraic linear equations are combined to construct the solution for a stationary free electric point charge in infinite space under the assumption that the appropriate variables vanish asymptotically at large spatial distance. A similar treatment is applied to the problem of steady vibrations caused by a point heat source. These results should be useful in future applications of the theory.\N\NThe Lagrange identity method pioneered by \textit{L. Brun} [J. Méc., Paris 8, 167--192 (1969; Zbl 0194.26104)] was apparently not considered for the uniqueness proof, nor were integral transform techniques explored in the treatment of point singularities.