Published:January 01, 2001
The use of computer simulations, or “computer experiments” as I prefer to call them, is becoming one routine tool among others, and is opening new doors to research in many fields including mineralogy. I shall be concerned here purely with simulations at the atomic level, though the general remarks apply also to macroscopic simulations in other connections.
From the point of view of scientific methodology, a computer experiment really is analogous to a laboratory experiment, although the actual techniques are very different. One has a system consisting of a number of atoms in some structure, which may be crystalline or glassy or molten, which may or may not have impurities and other types of defects, and which may have a free surface or represent pure bulk. One also puts into the computer the interaction forces between the atoms, such as the electrostatic attractions and repulsions between ions, and especially the chemical bonding between atoms and their repulsion if one tries to push them too close together. This can be done at two different levels of sophistication. Either one represents the bonding by some empirical interatomic interaction potentials including forces depending on the angle between two bonds at an atom (Dove, 2001), or else one solves the quantum mechanical equations that govern all the electrons in the system which are ultimately responsible for the chemical bonding and all ordinary properties of matter. laws of motion to give the correct movements of the atoms, using the forces on them calculated from the classical model or quantum mechanically, whichever route one is taking.
Figures & Tables
Solid Solutions in Silicate and Oxide Systems
The EMU book series or notes, as they are called, were introduced to provide university teachers with up-to-date reviews in important, rapidly evolving areas of mineralogy, petrology and geochemistry. They are also meant to introduce scientists into special and often interdisciplinary fields of research. In this regard, a volume on solid solutions is current and sorely needed. The solid Earth, as well as many meteorites and the other solid planets, consists for the most part of mineral solid solutions. Research on solid solutions is extremely broad encompassing work in physics and chemistry, metallurgy, materials science and, last but not least, mineralogy and petrology. Hence, because the theme is so strongly interdisciplinary in nature, the workshop was organised to include solid state physicists, physical chemists, crystallographers, mineralogists and petrologists. The various chapters reflect some of this diversity and show what mineralogy has become. Experimental investigations in mineralogy now routinely include different types of spectroscopies along with more traditional phase equilibrium, X-ray diffraction, calorimetry, and TEM methods. There have also been new and impressive developments in theory and computation. Many computational approaches relating to the study of solid solutions, for example, the Cluster Variation Method or Monte Carlo simulations, have been brought in from materials science, chemistry and physics. It can be concluded that the traditional or historical, and perhaps artificial, boundaries between the various disciplines are disappearing. Many current research efforts in mineralogy are similar to those in chemistry, materials science and physics.