Igor Sevostianov

RESSOURCES21 accueille de son 1^er Chercheur Invité, le P^r Igor SEVOSTIANOV.

Igor Sevostianov est professeur au département Mechanical and Aerospace Engineering de New Mexico State University, USA. Ses recherches sont consacrées à la modélisation micromécanique des milieux poreux hétérogènes, notamment :
 identification des paramètres microstructuraux pertinents pour la caractérisation de milieux hétérogènes,
 analyse de propriétés croisées : expression des propriétés élastiques en fonction des conductivités électriques ou thermiques,
 caractérisation des relations entre microstructure et propriétés effectives. Il est co-auteur de nombreux papiers dans le domaine de la physique et mécanique des matériaux, il est membre du comité de lecture de International Journal of Engineering Sciences et éditeur associé de International Journal of Theoretical and Applied Multiscale Mechanics.

Sujet et Résumé de la conférence :

RELATING MATERIAL’S PERFORMANCE TO MICROSTRUCTURE : CHALLENGES AND OPPORTUNITIES

The presentation covers two interconnected topics I. QUANTITATIVE CHARACTERIZATION OF MICROSTRUCTURES Proper quantitative characterization of microstructures, for the purpose of modeling the effective properties, is discussed. This is a broad subject that covers different physical properties (elastic, conductive, transport, etc), as well as various types of microstructures. The presentation focuses on microstructures that can be characterized as continuous matrices containing isolated inhomogeneities of diverse shapes, properties and orientations. We address their proper quantitative characterization in the context of elastic and conductive properties (transport and fracture-related properties are also briefly discussed). Proper microstructural parameters must correctly represent the individual inhomogeneity contributions to the considered property. They may differ for different physical properties. The key problem is to identify the mentioned individual contributions.

For the elastic properties, we demonstrate, on a number of microstructures, how the proper parameters are implied by the elastic potential. Relative importance of various “irregularity factors” (shape irregularities, orientation scatter) is analyzed. We discuss similarities and differences between microstructural parameters intended for different physical properties. The possibility of explicit cross-property connections between two physical properties depends on whether the proper microstructural parameters for these two properties are sufficiently similar.

II. EXPLICIT CONNECTIONS BETWEEN ELASTIC AND CONDUCTIVE PROPERTIES : THEORY AND EXPERIMENTAL VERIFICATION

Cross-property connections for heterogeneous materials belong to the realm of fundamental problems of engineering science. Whereas connections between properties governed by mathematically similar laws are straightforward (say, electric and thermal conductivities), the ones between the elastic properties and conductivities constitute a much more complex problem. Moreover, their very existence is not obvious : besides being governed by different differential equations, they are characterized by tensors of different ranks. Their practical usefulness lies in the fact that one physical property (say, electric conductivity) may be easier to measure than the other (say, full set of anisotropic elastic constants

In the presentation, approximate connections between conductivities and elastic compliances are discussed and specified for several heterogeneous anisotropic microstructures and verified by comparison with experimental data. The tensor of elastic compliances is expressed in terms of the conductivity tensor in the closed form. The cross-property connections are derived in the framework of non-interaction approximation. In the practical cases, when the interactions between inhomogeneities cannot be neglected, we hypothesize that the interactions affect both groups of properties – elastic and conductive – in a similar way, so that the cross-property correlations continue to hold, although this approximation may yield substantial errors for each of the properties separately (this idea was first suggested by Bristow, 1960 for a material with randomly oriented microcracks). This assumption is confirmed by comparison with experimental data on various materials : plasma sprayed ceramic coatings, short glass fiber reinforced thermoplastics and aluminum foams. The similar type of cross- property connection is also obtained for granular materials.