SUMMARY. The effect of heat treatment in air at temperatures from 200 ~ to iooo ~ of two Cornish biotites has been studied using the M6ssbauer effect in 57Fe. One of the micas was also studied after heating in vacuo.Progressive changes in the M6ssbauer spectra of the materials after heat treatment can be correlated with the following processes: Oxidation of Fe 2+ ions in isolated octahedral sites to give Fe 3+ in the co-ordination Fe3+(OsOH); oxidation of Fe ~+ ions in adjacent a-octahedral (cis OH-) sites to give FeS+(O~); and oxidation of Fe z+ in adjacent a and b (trans OH-) sites to give either Fe3+(O~OH) or Fe3+(O6) depending on whether a shared hydroxyl group remains intact or otherwise (these processes occur at temperatures below 5oo ~ Also, at higher temperatures, dehydroxylation of Fe3+(O~OH) to give more Fe 3+ in the Fez+(Os) configuration. Finally, in the temperature range 9o0 to IOOO ~ structural breakdown yielding a-Fe~Oz as the iron-containing phase.The results and interpretation are in good agreement with thermogravimetric data and with a previous infra-red study. The mechanism of charge diffusion in the lattice during oxidation is discussed in the light of the results.ALTHOUGH much work has beend one on the thermal decomposition of hydrous silicates in general, there has been little published on biotites in particular, most of the mica work having been confined to muscovite. Brindley (1963) has summarized much of the work prior to that date and has emphasized the importance of X-ray techniques in elucidating the broader outlines of the various changes, but the more detailed mechanisms require other techniques for their elucidation. Hodgson, Freeman, and Taylor (I965a) have studied the thermal decomposition of crocidolite and amosite (I965b) by differential thermal analysis and thermogravimetry, and have interpreted their results in terms of the original detailed investigation of amphiboles by Addison et al. (1962). Gibb and Greenwood (I965) have also investigated these two amphiboles using the MSssbauer effect and their results generally support the results of previous workers. Vedder and Wilkins (1969) have studied the thermal decomposition of muscovites and biotites by infra-red spectroscopy, and the interpretation of the results to be presented here has been made along the lines of their arguments.A M~Sssbauer study of high-temperature reactions in kaolinite and halloysite has been undertaken by MacKenzie (1969) and M~Sssbauer work on oxidation of biotite by weathering has been published by Rice and Williams (I969).

No abstract

SUMMARY. Two clay samples originating in the kaolinized granite area near St. Austell, Cornwall, have been examined by MSssbauer spectroscopy. The iron-containing impurity in the sample with a red discoloration (sample A) has, by measurement of the hyperfine magnetic field, been identified as hematite. In the case of sample B, which was yellow in colour, preliminary MSssbauer work indicated fi-FeOOH as the impurity, but detailed measurements of hyperfine field and N6el temperature, when compared with similar results for a synthetic/3-FeOOH sample, suggested otherwise. X-ray and chemical work suggested that the iron-containing phase in sample B is goethite, ~-FeOOH in which some substitution of Al+8 for Fe z+ has occurred; the MSssbauer data are consistent with this conclusion. Both clay samples exhibited superparamagnetic behaviour and this has been utilized to obtain rough estimates of the mean radius of the particles of the iron-containing impurity. These were 117/~ for the ~-Fe~O8 in sanaple A and 270/~ for aluminian c~-FeOOH in sample B.IT has been established by Malden and Meads (I967) by Mtissbauer exp ~'riments on a natural kaolinite purified by magnetic separation that Fe z+ substitutes, almost certainly in the octahedral cation site, in the kaolinite lattice. It is important to distinguish this situation from another that commonly exists, namely the occurrence of iron in separable impurities in the clay, usually in mica, or as oxides and oxy-hydroxides of iron. The experiments here described illustrate the usefulness of M6ssbauer work in the identification of impurities of the latter type. Some recent Mossbauer experiments by Janot et al. (I973) on kaolinites of tropical soil origin have yielded results that are generally similar to those reported here, but the measurements were restricted to two temperatures, 85 K and 3o0 K. As will be shown, more information regarding the iron-rich impurity phase can be obtained by experiments at a number of intermediate temperatures.The association of iron hydroxides with the surfaces of kaolinite particles has been studied by chemical, X-ray, and electron microscopic methods by Greenland and Oades (I968). Under certain circumstances the iron compound may be rather poorly crystallized.Sample description and preparation. Two samples of kaolinite from a clay pit in the St. Austell area of Cornwall were studied: both showed coloration, sample A having a red tint and sample B a yellow tint. Each clay contained about 2 % by 9

SUMMARY.Well-resolved M6ssbauer spectra of several dioctahedral and trioctahedral micas have been obtained and subjected to detailed computer analysis. Most of the spectra can be resolved into two ferrous quadrupole doublets and one ferric quadrupole doublet. In dioctahedral micas, ferrous iron is seen to occupy the larger, more symmetric octahedral site in preference to the smaller, less symmetric site, confirming predictions made on the basis of structure alone. In trioctahedral micas there is still a tendency for ferrous iron to occupy preferentially the more symmetric octahedral site even though the two octahedral sites are no longer distinguished by size. A lithium-rich biotite gives a spectrum typical of that expected from a mica with a zinnwaldite structure, but the spectrum of a more authenticated zinnwaldite could not be resolved sufficiently for detailed structural interpretation. The spectra of pyrophyllite and talc give results that can be related to the corresponding micas by considering their respective structural differences. In none of the micas studied was there evidence of ferric iron in tetrahedral coordination. The ferrous:ferric ratios obtained from the M6ssbauer spectra do not always agree with the chemical values. In several cases the spectra show more ferrous iron than found chemically. It is suggested that the spectral values are more accurate, oxidation on chemical analysis being the most likely source of error. THE MSssbauer spectra of various micas have been obtained by several authors (Pollak et al., I962; Herzenberg and Toms, t966; Weaver et al., I967; Malden and Meads, I967; Herzenberg et al., I968; Bowen et al., I969) but nowhere has a detailed analysis of the spectra in terms of mica structures been presented. This is probably due to the low iron content of many micas coupled with orientation effects, which give confusing spectra. Bowen et al. (I969) have obtained well-resolved spectra of several micas, but they found no evidence for more than one ferrous and one ferric quadrupole doublet, and have concentrated on the change in oxidation state of the iron in the micas under artificial weathering conditions. Bancroft et al. 0967, 1968) have shown how detailed computer analysis of M6ss-bauer spectra in minerals can lead to accurate interpretations of the spectra in terms of crystal structure. Similar techniques have been applied here to well-resolved spectra of a number of micas and related sheet silicates, and the spectra interpreted in terms of structure and site occupation.Sample description and preparation. Of the twelve sheet silicates investigated, five were