The crystal structure of barytes, celestine and anglesite

Abstract

1. Barytes, BaS0 4 , Celestine, SrS0 4 , and Anglesite, PbS0 4 , crystallise in the di-digonal equatorial class of the orthorhombic system, and are members of a very interesting isomorphous series, which includes, among other substances, the selenates and chromates of strontium and barium, the perchlorates and permanganates of potassium, rubidium and ammonium, and potassium fluoborate, KBF 4 . Although these compounds belong to several chemically very dissimilar classes, they are all of the type XRO 4 , where X is a kation and RO 4 an anion, and there can be little doubt that, when they crystallise, the size and shape of the RO 4 ion, which is probably very much the same whatever the nature of R, determines the arrangement of the atoms. This paper describes an attempt to analyse the structures of three crystals of this series. The structures of such crystals must be determined by a process of trial and error. Atomic arrangements must be assumed, and then tested by comparing the intensities of the X-ray spectra predicted from them with those actually observed. The crystals are orthorhombic, and therefore, so far as symmetry is concerned, the atoms have considerable freedom of position. The process of selecting the right structure will be simplified if we can get any idea as to the nature of the SO 4 group. The most reasonable assumption we can make is that the centres of the oxygen atoms lie at the corners of a regular tetrahedron with the sulphur atoms at its centre. The work of Bradley on the structure of LiKSO 4 shows that this assumption is well founded : the ion in this crystal certainly has a trigonal axis and cannot depart greatly from the regular tetrahedral form, with the distance S to O about 1.5 Å. If we suppose the ion has the same shape and size in other sulphates, we have to deal only with the orientation of the group, considered as a whole, and with the position of its centre. A number of types of arrangement quite consistent with the symmetry of the crystals are thus excluded, and the choice of arrangements to be tested by comparing the observed and calculated intensities is much narrowed down. This method has been followed in the present work, and it has been found possible to assign structures to the three crystals investigated which are practically identical, and which account completely for the observed intensities of the X-ray spectra given by them.