Of the many techniques that have been applied to the study of crystal defects, probably no single technique has contributed more to our understanding of their nature, properties, and influence on the physical and chemical properties of crystalline materials than transmission electron microscopy (ТЕМ). Although the importance of crystal defects and the use of ТЕМ for their direct observation were recognized by physical metallurgists in the early 1950s, it was at least a decade later that earth scientists responded to many of the new ideas of the defect solid state and to the power of ТЕМ. However, ТЕМ is now used extensively for the direct observation of defect microstructures in minerals and rocks, and there appears to be an increasing number of earth scientists who want to use the technique or to become more familiar with the interpretation of ТЕМ observations. This book is written for such people. However, it makes no attempt to be a practical manual of ТЕМ or a definitive text, but rather an introduction to the basic principles of the technique and of the interpretation of electron micrographs and electron diffraction patterns. As such, I hope the book will also be useful to students of materials science.

The physical-chemical properties of kimberlite melt govern the transport and eruption behaviour of kimberlite magmas. However, the physical properties of kimberlitic melts remain unknown, in part, because the composition of the melt phase is poorly constrained (Price et al. 2000; Kopylova et al. 2006; Sparks et al. 2006). As well, the experimental techniques needed to probe these extreme melts for viscosity (!), glass transition temperatures (Tg), heat capacity (Cp), and volatile solubilities are not yet available. Furthermore, the physical properties of kimberlite melts need not be simple linear extrapolations from the properties of other silicate melts (Russell & Giordano 2005). Computational models, calibrated on high quality experimental data, provide a means of exploring the physical properties of silicate melts for which there are no data.

In this presentation we use the current estimates for the range of kimberlite melt compositions (Table 1 & 2; after Price et al. 2000; Sparks et al. 2006; Kopylova et al. 2006). Secondly, we introduce and apply two multicomponent chemical models for predicting specific physical properties of melts in general. These models are designed to predict the calorimetric glass transition temperature and the viscosity of silicate melts as a function of melt composition and are used to explore the corresponding properties of kimberlite melts.

Cette étude a été commencée en Egypte à l'automne 1970 à la Faculté des Sciences de l'Université Ain Shams, au Caire, sous la direction du Dr. M.I. Youssef. Depuis, et malgré mon départ pour la France quelques années plus tard pour y poursuivre mes recherches, le Dr. Youssef n'a cessé de s'intéresser à mes travaux et de me prodiguer les encouragements dont j'avais besoin. Je suis heureux de lui exprimer ici ma profonde reconnaissance.

Geochemistry has deep roots. Its beginnings can be traced back to antiquity, but many of the discoveries that are basic to the science were made between 1800 and 1910. The periodic table of elements was assembled, radioactivity was discovered, and the thermodynamics of heterogeneous systems was developed. The solar spectrum was used to determine the composition of the Sun. This information, together with chemical analyses of meteorites, provided an entry to a larger view of the universe.