One hundred years of inorganic crystal chemistry - a personal view

It was suspected by Leucippus (fifth century BCE) that atoms had to be ordered, but no relationship between atomic order and crystals has been handed down to us from antiquity. In modern times, beginning with Kepler, 1611, crystals were thought to be composed of ordered small particles, but not necessarily atoms. This changed with Dalton, ca. 1800. Barlow made in 1883 a successful prediction of alkali halide structures, while Alfred Werner understood in 1893 already coordination compounds and coordination numbers. By the time of the experiments of the Braggs, 1913, atomism was mostly accepted. The first important crystal chemistry paper was published in 1920 by W. Lawrence Bragg, the son: he found that interatomic distances in crystals obeyed an additivity rule. In 1926, Goldschmidt had amassed a large amount of crystal structure data and deduced from that what he called the laws of crystal chemistry, thus he became the founder of crystal chemistry. On the basis of W. L. Bragg's crystal structure determinations and of Goldschmidt's laws Machatschki unravelled in 1927, the principles of the constitution of the silicates, a problem that had vexed mineralogists and inorganic chemists during the nineteenth century. In 1928/1929, Pauling published his famous rules concerning the principles determining the structures of complex ionic crystals. His second rule, the electrostatic valence principle, which essentially says that the charges of the ions in a crystal structure are balanced locally, was particularly useful. The first book on crystal chemistry was published by Hassel in 1934. Wells showed beginning in 1954 in a series of papers and books that one could systematize crystal structures on the basis of three-dimensional periodic nets of bonds. After the 1950s, the increasing accuracy of crystal structure determinations made it possible to look for interpretations of bond length variations and distortions of coordination polyhedra. Barnighausen constructed family trees of the group-subgroup relationships of topologically related crystal structures. A general method for predicting crystal structures of inorganic compounds from a knowledge of the chemical composition alone is not yet available. Empirical crystal chemistry remains a valuable tool for searching for methods of representing and classifying structures.