Clays: Controlling the Environment

Clays: Controlling the Environment

Proceedings of the 10th International Clays Conference, Adelaide, Australia, July 18 to 23, 1993.

Clays have provided us with the most active ingredients in soils, with building materials, with pottery and ceramics for both utility and decoration, and with coatings and fillers for paper, among other uses. The unique properties of these apparently everyday materials are being studied and used in an increasing range of industrial and environmental applications.

Clays: Controlling the Environment provides a valuable compendium of the latest results from the complete range of clay-related scientific research. It includes coverage of the economic and environmental issues as well as directions for further research and development in many vital and expanding industries. All papers in these proceedings were subject to peer review.

The topics discussed are:

*Clays in industry and the environment

*Surface and interlayer reactions

*Clay mineral structures and chemistry

*Methods of investigation

*Clays in geology

*Soil mineralogy

The emphasis of this book reflects the vital role that clays play in controlling natural, polluted and technological environments.

  1. Page xiii
  2. Page xv
  3. Page xvii
  4. Page xviii
    1. Page 3
      Abstract

      The use of clay minerals for improving environmental quality can take two approaches. One strategy is to tailor the clay either as an adsorbent or as a catalyst for remediation processes to correct already polluted environments. The second, and more desirable, strategy is to utilise the clay for ‘green chemical processing, wherein reagents are converted to products in essentially quantitative yields with little or no polluting byproducts. The present paper focuses on the properties of cationic (smectitic) and anionic (hydrotalcite-like) clays as catalysts for potentially benign chemical processes.

  5. Page 9
    Abstract

    All papers under this Theme were given by invitation. Dr J.B. Dixon, Convenor of the Symposium on Teaching at the Conference, writes: ‘Teaching clay mineralogy continues to grow as a challenging and important professional pursuit. The current economic stresses make staffing and student levels modest. Yet the need for information on clays to address environmental needs where clays play a major role in mining and reclaiming mined lands, in controlling pollution of water supplies, and in assuring the continued abundance of food and fibre for a growing population all indicate the need for a good supply of well trained and well educated students.’

    University teaching programmes involving day mineralogy should include instruction on short-range order (SRO) clays and specifically allophane, which forms primarily in volcanic terrains through andisolisation or in podzolising situations on various materials. This paper describes, for teaching purposes, a summary of the different types of allophanes and their structures, their main conditions of formation, and their influence on soils (Andisols). It includes several purposely simplified figures suitable for slide reproduction aimed at helping student understanding of relatively complex structures and processes. The work is based chiefly on experience from New Zealand.

    A review of the phyllosilicate-structure literature is given and the following trends noted: (1) fewer new structures are reported, (2) research involving the effect of environmental conditions on structures is increasingly common, and (3) new techniques are becoming routine. The research area of structural mineralogy appears to be evolving: as availability of new structures suitable for single crystal study declines, research is increasingly directed towards geologically-relevant questions, such as how structures respond to environmental conditions. Spectroscopic studies are common also, with the emphasis in establishing empirical approaches to interpret the spectra and not theoretical methods. Thus, much of the spectroscopic research involves the examination of a set of phyllosilicates with a systematic variation of chemistry.

    The distribution of calcite in a hot spring microbial mat from Iceland has been analyzed chemically, structually and crystallographically. Optical microscopy, scanning and transmission electron microscopies with energy-dispersive X-ray spectroscopy of thin sectioned sample revealed a variety of calcite-encrusted microbial forms. Strong 3.01, 2.47 and 2.20 Å d-spacings of X-ray powder diffraction pattern in the corresponding electron diffraction patterns indicated that well crystallized calcite was formed within the microbial mat. Many microorganisms, particulary cyanobacteria such as Oscillatoria, Synechocystis, Gloeothece and Chamaesiphon, were encrusted by calcite microcrystals which were deposited extensively around cell walls or colonies and within sheaths or cells. Individual bacterial cells and their colonies were prominent as nucleation sites for CaCO3

    Teaching is easier and learning is more thorough when visual concepts can be employed. A short field trip can contribute to the visual perception of clays and tie the facts learned later in the course to the environments of soils and rocks where interpretations and applications later may be made. Five environments are presented: soil profile, siderite weathering, seasonally formed and dissolved iron oxides, tonstein formation, and residual iron oxide and silicate minerals. A short field trip has proven to be a great help in illustrating several visual properties of 20 minerals and the environments in which they can occur early in the clay mineralogy course. This approach has been used in our soil mineralogy course for many years and it continues to be improved.

    1. Page 11
    2. Page 15
    3. Page 19
      Abstract

      University teaching programmes involving day mineralogy should include instruction on short-range order (SRO) clays and specifically allophane, which forms primarily in volcanic terrains through andisolisation or in podzolising situations on various materials. This paper describes, for teaching purposes, a summary of the different types of allophanes and their structures, their main conditions of formation, and their influence on soils (Andisols). It includes several purposely simplified figures suitable for slide reproduction aimed at helping student understanding of relatively complex structures and processes. The work is based chiefly on experience from New Zealand.

    4. Page 24
      Abstract

      A review of the phyllosilicate-structure literature is given and the following trends noted: (1) fewer new structures are reported, (2) research involving the effect of environmental conditions on structures is increasingly common, and (3) new techniques are becoming routine. The research area of structural mineralogy appears to be evolving: as availability of new structures suitable for single crystal study declines, research is increasingly directed towards geologically-relevant questions, such as how structures respond to environmental conditions. Spectroscopic studies are common also, with the emphasis in establishing empirical approaches to interpret the spectra and not theoretical methods. Thus, much of the spectroscopic research involves the examination of a set of phyllosilicates with a systematic variation of chemistry.

    5. Page 30
      Abstract

      The distribution of calcite in a hot spring microbial mat from Iceland has been analyzed chemically, structually and crystallographically. Optical microscopy, scanning and transmission electron microscopies with energy-dispersive X-ray spectroscopy of thin sectioned sample revealed a variety of calcite-encrusted microbial forms. Strong 3.01, 2.47 and 2.20 Å d-spacings of X-ray powder diffraction pattern in the corresponding electron diffraction patterns indicated that well crystallized calcite was formed within the microbial mat. Many microorganisms, particulary cyanobacteria such as Oscillatoria, Synechocystis, Gloeothece and Chamaesiphon, were encrusted by calcite microcrystals which were deposited extensively around cell walls or colonies and within sheaths or cells. Individual bacterial cells and their colonies were prominent as nucleation sites for CaCO3

    6. Page 38
      Abstract

      Teaching is easier and learning is more thorough when visual concepts can be employed. A short field trip can contribute to the visual perception of clays and tie the facts learned later in the course to the environments of soils and rocks where interpretations and applications later may be made. Five environments are presented: soil profile, siderite weathering, seasonally formed and dissolved iron oxides, tonstein formation, and residual iron oxide and silicate minerals. A short field trip has proven to be a great help in illustrating several visual properties of 20 minerals and the environments in which they can occur early in the clay mineralogy course. This approach has been used in our soil mineralogy course for many years and it continues to be improved.

  6. Page 47
    Abstract

    The Keynote paper covers both aspects of this Theme, while the first seven papers deal largely with environmental aspects of clays and the last seven with clays in industry. However, several of the papers address both aspects. Among these is the paper by Butruille and Pinnavaia. With this paper, Dr Jean-Rémi Butruille was the winner of the Bradley Award for a young clay scientist.

    Clays are important industrial minerals used throughout the world in construction, agriculture, engineering and environmental related applications and in the ceramic, paint, paper, plastics, rubber, foundry, glass, petroleum, fertiliser, chemical and other industries. Clay minerals used in industry include kaolinite, halloysite, sodium montmorillonite, calcium montmorillonite, hectorite, palygorskite, sepiolite, and to a lesser extent, illite.

    The applications of clay minerals are intimately related to their structure and composition. Important characteristics are particle size and shape, surface area and chemistry, and other physical and chemical properties specific to particular applications effecting brightness and colour; viscosity; plasticity; green dry, and fired strength; adsorption and adsorption; abrasion and others. Differences in structure and composition of the various clay minerals make for diverse industrial and environmental applications. Kaolinite, which has a low surface area (8–12 m2gl) and low charge has very different applications than sodium montmorillonite, which has a high surface area (150–200 m2g1) and high charge. Clay minerals with high sorptive capacities such as calcium montmorillonite and palygorskitelsepiolite are used in increasingly important environmental applications such as clay liners for landfills and hazardous waste sites, and in geosynthetic membranes.

    Clays now have a well-established role in landfill technology. The mineral sealing liner often provides the final barrier for preventing or retarding pollutant emissions in a landfill basal lining system. Soil physical attributes are important, as well as clay mineralogical and geochemical characteristics, which are not discussed here. Mechanisms of diffusion and sorption, as well as the mechanism of convection, are required for permeability calculations. Yet only limited values are available for convective transport, while no useful instruments exist for the determination of diffusion and sorption parameters.

    A special laboratory test to investigate diffusion and sorption parameters is described. The test was developed following theoretical considerations. The method involves a flow of solution with a constant gradient parallel to the surface of a soil sample. Distilled water is flowed past the opposite side of the sample. Using this arrangement, one obtains a constant concentration gradient through the sample so that diffusion results. The concentrations of chemicals in liquids on both sides of the sample are determined frequently so that the mass of pollutants moving through the sample can be calculated. Consequently, diffusion and sorption parameters can be calculated. The effect of an overburden pressure on a soil can also be considered. An example is given of the calculation of the emission of zinc through a classified mineral base liner.

    Compacted sodium bentonite has been proposed as a buffer material in the final disposal of spent nuclear fuel in many countries. Safety analysis requires the diffusion mechanisms and the corresponding parameters to be known. In this paper, the results of our studies on diffusion mechanisms in compacted bentonite have been compiled and evaluated in relation to the microstructure of bentonite. The diffusivities, effective porosities and sorption factors were determined by tracer experiments. Exclusion of chloride and anionic uranium complexes could be closely explained by the use of the Stem-Gouy model when the differences in the charge of the ions were taken into account. An exponential expression was developed for anions to describe the relation between effective porosity and apparent and effective diffusivity. Diffusivities of Cs+ and Sr2+ in compacted bentonite were studied as a function of sorption factor, which was adjusted by electrolyte concentration in solution. The apparent diffusivities were independent of the sorption, and the effective diffusivities were found to increase with increasing sorption factor, thus indicating that sorbed cations would diffuse.

    Selective sorption and fixation of Cs+ in smectites has been examined by chemical analysis and X-ray powder diffractotnetry (XRD).

    Distribution coefficients (Kd) for cations measured on initially Na* -saturated smectites in aqueous solutions were generally in the order Cs+>Cd2+>Sr2>Na+. The Kd values for Ca2+, Sr2+ and Na+ were substantially constant for almost all specimens, whereas values for Cs* decreased with reaction temperature and depended on the sites of isomorpbous substitution in the silicate layers; the larger the octahedral charge and the lower the reaction temperature, the greater was the Kd value.

    The amount of fixed Cs+ also increased with increasing octahedral charge but not necessarily with tetrahedral charge. This result appears to contradict the fact that layer silicates with large total charge, especially with large tetrahedral charge, tend to fix many kinds of cations readily. Cs+ occupancies in collapsed interlayers after Ba2+ -exchange were estimated by examining the interstratified structure of Cs+-fixed specimens. Such occupancies however show a poor correspondence with the amount of fixed Cs+, probably because the swelling of interlayers occurs at values of occupancy by Cs+ which differ according to the specimen. Cs+ occupancy was large for specimens with a high layer charge which included a high octahedral charge. In such smectites, the interlayers do not collapse easily even after large amounts of Cs+ have been fixed. The height of the interlayer is large enough for more Cs+ to penetrate into the interactive sites. Hence Kd values were high for specimens with large octahedral charge.

    Isotherms for sorption and desorption and distribution coefficients of radioiodide (KI labelled with 125 I) were measured on different particle size fractions (clay and silt) of Cretaceous clay, of loess soil and of Vermiculite each treated with hexadecylpyridiniumchloride (HDPY+). Isotherms were linear up to iodide fractional loadings of about 5×101 mmolg1. At higher iodide concentrations a hysteresis occurred indicating more than one type of sorption mechanism. Depending on the clay type and on the size fraction the distribution coefficients (RD-values) varied between 3 and 15 000 mlg1.

    Highest iodide sorption was observed in the medium size clay fractions of the materials investigated. In the case of Cretaceous clay and clay from loess the sorption behaviour was related to the HDPY+ incorporation levels, whereas with the fractions of vermiculite no relation between Revalues for radioiodide and HDPY+-uptake was observed.

    The firing process of brick clays generally gives rise to the release of pollutant elements (F, S, CL) into the flue gas. The amount of stack emissions depends on several factors, firstly the content of pollutant elements in the raw materials. 250 samples of clays used by Italian brickworks were analyzed by X-ray fluorescence spectrometry. Fluorine contents ranged from 500 to 1200 ppm (average 870 ppm); sulfur 100–6200 ppm (average 670 ppm); chlorine 10–1000 ppm (average 100 ppm). The fluorine release is the most serious problem, since its content in the clays is quite high and similar in the various geological units. Actual emissions should depend to a large extent on the technological process. The sulfur contribution from clays to stack emissions appears to be less than that due to S-rich combustibles and limited in comparison with other industries. The risk of pollution seems to be minimal for chlorine, due to the predominantly low concentrations.

    The Hot Water Extraction process is presently used in Alberta to extract bitumen from oil sands ore; during the process an aqueous dispersion of fine solids is generated. Eventually the solid-liquid system stabilises to a sludge with a gel-like structure having very poor settling and compaction rates. The gel-forming nature and, consequently, high water holding capacity of this sludge have been attributed to the presence of colloidal solids fractions identified as ultrafine (≤0.2 μm) aluminosilicates. The gel-forming propensity appears to depend on concentration and particle size of the solids as well as salinity of the water.

    On the basis of their surface characteristics, two major types of ultrafine solids have been recognised in both oil sands and sludge, these are: preferentially hydrophilic solids (AS) and biwetted solids (BUS) associated with various amounts of so-called ‘strongly bound organic matter’. This organic matter could be the source of the attractive forces responsible for the observed accelerating rate of gelation in the presence of BUS.

    From the results obtained we elucidate the colloidal properties of these ultrafines and discuss the implications for environmentally acceptable disposition of the sludge.

    Abstract:

    The catalytic performance of a Co(II) phthalocyanine tetrasulfonate complex, [CoPcTs]4, intercalated in [Mg1_x 2+ Alx3+(OH)2][An]x/n. yH2O layered double hydroxides (LDHs) has been evaluated for the dioxygen oxidation of 2,6-di-tert-butyl phenol and 1-decanethiol. Regardless of the LDH layer charge, the complex intercalates ‘edge on’, i.e. with the plane of the macrocycle perpendicular to the LDH layers. This orientation allows the metal centres at edge sites to be accessible for reaction with reagents from solution. Although intracrystal microporosity sufficient for the interlayer adsorption of organic reagents is absent, a low charge density LDH is more active and, hence, preferred as a support for the immobilisation of the metallomacrocyclic molecules for the purpose of remediating contaminated ground water and industrial effluents.

    The molecular sieving properties of alumina pillared clays have not been demonstrated as clearly as with zeolites. The isopropylation of biphenyl has been studied as a test reaction for pillared clay catalysts. Clays of various charge densities, charge localisation and extent of layer stacking, and also an acid washed montmorillonite and two zeolites were investigated. Under liquid phase conditions at 250°C, the reaction was under diffusion control. Thus, the activities and selectivities of the catalysts were mainly dependent on their pore structure. In particular, the mesoporous volume was critical for catalytic activity. When calcined at 350°C, alumina pillared fluorohectorite exhibited greatly enhanced catalytic activity and acidity, whereas calcination at 500°C resulted in a relatively inactive clay with greatly diminished base chemisorption properties. Using27Al, 29Si and 19FMAS-NMR, FTIR spectroscopy and mass spectrometric analysis of volatile by-products, it was shown that between 30°C and 500°C, specific lattice fluorine atoms adjacent to charged sites in the octahedral layers are replaced by hydroxyl groups. The Brensted acidity of hydroxyl groups is enhanced by the near-neighbor fluorine at neutral sites in the octahedral sheet. Between 350°C and 500°C, a second process occurs that causes dehydroxylation of the layers, and this results in a sharp decrease in the acidity and catalytic activity of the pillared clay.

    A series of 48 clays was studied to find relationships between materials data and rheological behaviour. The results emphasise the significance of the type of expandable minerals in clays for their ability to form slips of high solid concentration: montmorillonite prevents deflocculation even when present only in small amounts. Expandable layers in mixed-layer minerals are less critical; indeed they just cause increased consumption of deflocculant for the formation of castable slips. Ion exchange data of clays give additional information on their rheological properties.

    Extrusion through a range of dies shows that volume adjustment in deforming plastic clay takes place by sliding on a network of boundary surfaces which surround less sheared domains of clay. During extrusion, accommodation to the die geometry proceeds by adjustment between domains and reshaping of domain boundaries, rather than by individual clay particles responding to pressure gradients in a stepless plastic deformation.

    A freezing technique was used for visualising domain structure in a group of structural clay bodies, whose particle size distribution, plasticity and clay mineralogy were characterised. Domain structure helps to explain the deformation phenomena of clay bodies with different characteristics, and aspects of their manufacturing defects and durability in service that are texture dependent.

    The acid treatment of clays has long been studied with the intention of modifing their texture and acidity for catalytic applications. Here we have investigated the effect of the kaolin calcination temperature on the final textural properties of the acid dealuminated metakaolins. Calcination temperatures rangingfrom 550 to 925° C were used, and dealumination was carried out with sulfuric acid. Textural properties were studied by nitrogen adsorption and mercury porosimetry. Only pores wider than 200 Å are unaffected by different calcination temperatures. Furthermore, the increase in such temperature follows the increase in microporosity up to 925° C. It is to be supposed that the resulting microporosity is due to the removal of gibbsite layers from the metakaolin structure. The micropores generated by the acid attack are approximately 11 Å wide. Mesopores may be formed by partial destruction of the structure.

    Brazil is a major producer of processed kaolin with an annual production of approximately 0.8 million tonnes. The largest mining operations export paper coating kaolin from sedimentary deposits in the Amazon basin. Brazilian deposits are of diverse origins and in addition to sedimentary kaolin, include kaolin derived from pegmatites, granite, volcanic rocks and anorthosites. Amazon sedimentary kaolin comprises euhedral kaolinite with high iron and titania (lattice-held), and low alkali contents. South-eastern pegmatites are a mixture of kaolinite and 7 Å/10 Å-halloysitt and can be very low in iron and titania. North-eastern pegmatites contain predominantly euhedral kaolinite. Kaolins from granites generally have higher iron contents than those from pegmatites and invariably are mixtures of kaolinite and 7 Å halloysite. Volcanic rocks produce a fine siliceous kaolin while anorthosites give kaolin with iron and titania levels similar to those derived from granitic rocks. Kaolins formed from basement rocks are utilised mainly for ceramics and some for paper filling.

    Within the framework of research aimed at the characterisation of the raw materials used by the Italian brick industry, 52 clays from Tuscany and Umbria (central Italy) were studied from the chemical, mineralogical and granulometric viewpoints and related to the main properties of heavy-clay products. The corresponding 55 heavy-clay products (masonry bricks, hollow products, roofing tiles and paving bricks) are produced using both carbonate-bearing and carbonate-free clays. Porosity and pore size distribution are clearly correlated to both clay mineral and carbonate content. Efflorescences are discussed with reference to both soluble salts and pore size distribution.

    Over 230 kaolin mines are located in the lower valleys of the Chubut River, Chubut Province, Argentina. Kaolin formed by alteration of a volcanic rhyolitic complex of Jurassic age and has been extensively explored and tested for suitability as a filler for use in the paper industry. The <2 μm fraction shows good response to bleaching to give a high brightness filler of medium to low abrasion. Coarser kaolin fractions have higher quartz content and are suitable only for ceramics. Mineralogy is dominantly kaolinite and 7 Å halloysite, with minor quartz. At depth, illite, mixed layer clays and smectite are also present. Isotopic composition of structural water in kaolin supports formation by weathering, although some parasitic exhalative hydrothermal activity in areas of structural control may have increased the depth of kaolinisation.

    1. Page 49
      Abstract

      Clays are important industrial minerals used throughout the world in construction, agriculture, engineering and environmental related applications and in the ceramic, paint, paper, plastics, rubber, foundry, glass, petroleum, fertiliser, chemical and other industries. Clay minerals used in industry include kaolinite, halloysite, sodium montmorillonite, calcium montmorillonite, hectorite, palygorskite, sepiolite, and to a lesser extent, illite.

      The applications of clay minerals are intimately related to their structure and composition. Important characteristics are particle size and shape, surface area and chemistry, and other physical and chemical properties specific to particular applications effecting brightness and colour; viscosity; plasticity; green dry, and fired strength; adsorption and adsorption; abrasion and others. Differences in structure and composition of the various clay minerals make for diverse industrial and environmental applications. Kaolinite, which has a low surface area (8–12 m2gl) and low charge has very different applications than sodium montmorillonite, which has a high surface area (150–200 m2g1) and high charge. Clay minerals with high sorptive capacities such as calcium montmorillonite and palygorskitelsepiolite are used in increasingly important environmental applications such as clay liners for landfills and hazardous waste sites, and in geosynthetic membranes.

    2. Page 56
      Abstract

      Clays now have a well-established role in landfill technology. The mineral sealing liner often provides the final barrier for preventing or retarding pollutant emissions in a landfill basal lining system. Soil physical attributes are important, as well as clay mineralogical and geochemical characteristics, which are not discussed here. Mechanisms of diffusion and sorption, as well as the mechanism of convection, are required for permeability calculations. Yet only limited values are available for convective transport, while no useful instruments exist for the determination of diffusion and sorption parameters.

      A special laboratory test to investigate diffusion and sorption parameters is described. The test was developed following theoretical considerations. The method involves a flow of solution with a constant gradient parallel to the surface of a soil sample. Distilled water is flowed past the opposite side of the sample. Using this arrangement, one obtains a constant concentration gradient through the sample so that diffusion results. The concentrations of chemicals in liquids on both sides of the sample are determined frequently so that the mass of pollutants moving through the sample can be calculated. Consequently, diffusion and sorption parameters can be calculated. The effect of an overburden pressure on a soil can also be considered. An example is given of the calculation of the emission of zinc through a classified mineral base liner.

    3. Page 61
      Abstract

      Compacted sodium bentonite has been proposed as a buffer material in the final disposal of spent nuclear fuel in many countries. Safety analysis requires the diffusion mechanisms and the corresponding parameters to be known. In this paper, the results of our studies on diffusion mechanisms in compacted bentonite have been compiled and evaluated in relation to the microstructure of bentonite. The diffusivities, effective porosities and sorption factors were determined by tracer experiments. Exclusion of chloride and anionic uranium complexes could be closely explained by the use of the Stem-Gouy model when the differences in the charge of the ions were taken into account. An exponential expression was developed for anions to describe the relation between effective porosity and apparent and effective diffusivity. Diffusivities of Cs+ and Sr2+ in compacted bentonite were studied as a function of sorption factor, which was adjusted by electrolyte concentration in solution. The apparent diffusivities were independent of the sorption, and the effective diffusivities were found to increase with increasing sorption factor, thus indicating that sorbed cations would diffuse.

    4. Page 67
      Abstract

      Selective sorption and fixation of Cs+ in smectites has been examined by chemical analysis and X-ray powder diffractotnetry (XRD).

      Distribution coefficients (Kd) for cations measured on initially Na* -saturated smectites in aqueous solutions were generally in the order Cs+>Cd2+>Sr2>Na+. The Kd values for Ca2+, Sr2+ and Na+ were substantially constant for almost all specimens, whereas values for Cs* decreased with reaction temperature and depended on the sites of isomorpbous substitution in the silicate layers; the larger the octahedral charge and the lower the reaction temperature, the greater was the Kd value.

      The amount of fixed Cs+ also increased with increasing octahedral charge but not necessarily with tetrahedral charge. This result appears to contradict the fact that layer silicates with large total charge, especially with large tetrahedral charge, tend to fix many kinds of cations readily. Cs+ occupancies in collapsed interlayers after Ba2+ -exchange were estimated by examining the interstratified structure of Cs+-fixed specimens. Such occupancies however show a poor correspondence with the amount of fixed Cs+, probably because the swelling of interlayers occurs at values of occupancy by Cs+ which differ according to the specimen. Cs+ occupancy was large for specimens with a high layer charge which included a high octahedral charge. In such smectites, the interlayers do not collapse easily even after large amounts of Cs+ have been fixed. The height of the interlayer is large enough for more Cs+ to penetrate into the interactive sites. Hence Kd values were high for specimens with large octahedral charge.

    5. Page 74
      Abstract

      Isotherms for sorption and desorption and distribution coefficients of radioiodide (KI labelled with 125 I) were measured on different particle size fractions (clay and silt) of Cretaceous clay, of loess soil and of Vermiculite each treated with hexadecylpyridiniumchloride (HDPY+). Isotherms were linear up to iodide fractional loadings of about 5×101 mmolg1. At higher iodide concentrations a hysteresis occurred indicating more than one type of sorption mechanism. Depending on the clay type and on the size fraction the distribution coefficients (RD-values) varied between 3 and 15 000 mlg1.

      Highest iodide sorption was observed in the medium size clay fractions of the materials investigated. In the case of Cretaceous clay and clay from loess the sorption behaviour was related to the HDPY+ incorporation levels, whereas with the fractions of vermiculite no relation between Revalues for radioiodide and HDPY+-uptake was observed.

    6. Page 79
      Abstract

      The firing process of brick clays generally gives rise to the release of pollutant elements (F, S, CL) into the flue gas. The amount of stack emissions depends on several factors, firstly the content of pollutant elements in the raw materials. 250 samples of clays used by Italian brickworks were analyzed by X-ray fluorescence spectrometry. Fluorine contents ranged from 500 to 1200 ppm (average 870 ppm); sulfur 100–6200 ppm (average 670 ppm); chlorine 10–1000 ppm (average 100 ppm). The fluorine release is the most serious problem, since its content in the clays is quite high and similar in the various geological units. Actual emissions should depend to a large extent on the technological process. The sulfur contribution from clays to stack emissions appears to be less than that due to S-rich combustibles and limited in comparison with other industries. The risk of pollution seems to be minimal for chlorine, due to the predominantly low concentrations.

    7. Page 84
      Abstract

      The Hot Water Extraction process is presently used in Alberta to extract bitumen from oil sands ore; during the process an aqueous dispersion of fine solids is generated. Eventually the solid-liquid system stabilises to a sludge with a gel-like structure having very poor settling and compaction rates. The gel-forming nature and, consequently, high water holding capacity of this sludge have been attributed to the presence of colloidal solids fractions identified as ultrafine (≤0.2 μm) aluminosilicates. The gel-forming propensity appears to depend on concentration and particle size of the solids as well as salinity of the water.

      On the basis of their surface characteristics, two major types of ultrafine solids have been recognised in both oil sands and sludge, these are: preferentially hydrophilic solids (AS) and biwetted solids (BUS) associated with various amounts of so-called ‘strongly bound organic matter’. This organic matter could be the source of the attractive forces responsible for the observed accelerating rate of gelation in the presence of BUS.

      From the results obtained we elucidate the colloidal properties of these ultrafines and discuss the implications for environmentally acceptable disposition of the sludge.

    8. Page 87
      Abstract

      Abstract:

      The catalytic performance of a Co(II) phthalocyanine tetrasulfonate complex, [CoPcTs]4, intercalated in [Mg1_x 2+ Alx3+(OH)2][An]x/n. yH2O layered double hydroxides (LDHs) has been evaluated for the dioxygen oxidation of 2,6-di-tert-butyl phenol and 1-decanethiol. Regardless of the LDH layer charge, the complex intercalates ‘edge on’, i.e. with the plane of the macrocycle perpendicular to the LDH layers. This orientation allows the metal centres at edge sites to be accessible for reaction with reagents from solution. Although intracrystal microporosity sufficient for the interlayer adsorption of organic reagents is absent, a low charge density LDH is more active and, hence, preferred as a support for the immobilisation of the metallomacrocyclic molecules for the purpose of remediating contaminated ground water and industrial effluents.

    9. Page 92
      Abstract

      The molecular sieving properties of alumina pillared clays have not been demonstrated as clearly as with zeolites. The isopropylation of biphenyl has been studied as a test reaction for pillared clay catalysts. Clays of various charge densities, charge localisation and extent of layer stacking, and also an acid washed montmorillonite and two zeolites were investigated. Under liquid phase conditions at 250°C, the reaction was under diffusion control. Thus, the activities and selectivities of the catalysts were mainly dependent on their pore structure. In particular, the mesoporous volume was critical for catalytic activity. When calcined at 350°C, alumina pillared fluorohectorite exhibited greatly enhanced catalytic activity and acidity, whereas calcination at 500°C resulted in a relatively inactive clay with greatly diminished base chemisorption properties. Using27Al, 29Si and 19FMAS-NMR, FTIR spectroscopy and mass spectrometric analysis of volatile by-products, it was shown that between 30°C and 500°C, specific lattice fluorine atoms adjacent to charged sites in the octahedral layers are replaced by hydroxyl groups. The Brensted acidity of hydroxyl groups is enhanced by the near-neighbor fluorine at neutral sites in the octahedral sheet. Between 350°C and 500°C, a second process occurs that causes dehydroxylation of the layers, and this results in a sharp decrease in the acidity and catalytic activity of the pillared clay.

    10. Page 102
      Abstract

      A series of 48 clays was studied to find relationships between materials data and rheological behaviour. The results emphasise the significance of the type of expandable minerals in clays for their ability to form slips of high solid concentration: montmorillonite prevents deflocculation even when present only in small amounts. Expandable layers in mixed-layer minerals are less critical; indeed they just cause increased consumption of deflocculant for the formation of castable slips. Ion exchange data of clays give additional information on their rheological properties.

    11. Page 107
      Abstract

      Extrusion through a range of dies shows that volume adjustment in deforming plastic clay takes place by sliding on a network of boundary surfaces which surround less sheared domains of clay. During extrusion, accommodation to the die geometry proceeds by adjustment between domains and reshaping of domain boundaries, rather than by individual clay particles responding to pressure gradients in a stepless plastic deformation.

      A freezing technique was used for visualising domain structure in a group of structural clay bodies, whose particle size distribution, plasticity and clay mineralogy were characterised. Domain structure helps to explain the deformation phenomena of clay bodies with different characteristics, and aspects of their manufacturing defects and durability in service that are texture dependent.

    12. Page 112
      Abstract

      The acid treatment of clays has long been studied with the intention of modifing their texture and acidity for catalytic applications. Here we have investigated the effect of the kaolin calcination temperature on the final textural properties of the acid dealuminated metakaolins. Calcination temperatures rangingfrom 550 to 925° C were used, and dealumination was carried out with sulfuric acid. Textural properties were studied by nitrogen adsorption and mercury porosimetry. Only pores wider than 200 Å are unaffected by different calcination temperatures. Furthermore, the increase in such temperature follows the increase in microporosity up to 925° C. It is to be supposed that the resulting microporosity is due to the removal of gibbsite layers from the metakaolin structure. The micropores generated by the acid attack are approximately 11 Å wide. Mesopores may be formed by partial destruction of the structure.

    13. Page 116
      Abstract

      Brazil is a major producer of processed kaolin with an annual production of approximately 0.8 million tonnes. The largest mining operations export paper coating kaolin from sedimentary deposits in the Amazon basin. Brazilian deposits are of diverse origins and in addition to sedimentary kaolin, include kaolin derived from pegmatites, granite, volcanic rocks and anorthosites. Amazon sedimentary kaolin comprises euhedral kaolinite with high iron and titania (lattice-held), and low alkali contents. South-eastern pegmatites are a mixture of kaolinite and 7 Å/10 Å-halloysitt and can be very low in iron and titania. North-eastern pegmatites contain predominantly euhedral kaolinite. Kaolins from granites generally have higher iron contents than those from pegmatites and invariably are mixtures of kaolinite and 7 Å halloysite. Volcanic rocks produce a fine siliceous kaolin while anorthosites give kaolin with iron and titania levels similar to those derived from granitic rocks. Kaolins formed from basement rocks are utilised mainly for ceramics and some for paper filling.

    14. Page 122
      Abstract

      Within the framework of research aimed at the characterisation of the raw materials used by the Italian brick industry, 52 clays from Tuscany and Umbria (central Italy) were studied from the chemical, mineralogical and granulometric viewpoints and related to the main properties of heavy-clay products. The corresponding 55 heavy-clay products (masonry bricks, hollow products, roofing tiles and paving bricks) are produced using both carbonate-bearing and carbonate-free clays. Porosity and pore size distribution are clearly correlated to both clay mineral and carbonate content. Efflorescences are discussed with reference to both soluble salts and pore size distribution.

    15. Page 129
      Abstract

      Over 230 kaolin mines are located in the lower valleys of the Chubut River, Chubut Province, Argentina. Kaolin formed by alteration of a volcanic rhyolitic complex of Jurassic age and has been extensively explored and tested for suitability as a filler for use in the paper industry. The <2 μm fraction shows good response to bleaching to give a high brightness filler of medium to low abrasion. Coarser kaolin fractions have higher quartz content and are suitable only for ceramics. Mineralogy is dominantly kaolinite and 7 Å halloysite, with minor quartz. At depth, illite, mixed layer clays and smectite are also present. Isotopic composition of structural water in kaolin supports formation by weathering, although some parasitic exhalative hydrothermal activity in areas of structural control may have increased the depth of kaolinisation.

  7. Page 135
    Abstract

    The extent and charge of the surfaces of clays provides one of their most important features. Hence, it is hardly surprising that many papers presented under this Theme describe properties of clays and modifications of their surfaces which have actual or potential applications in industry or to the improvement of our environment.

    Clays, and above all, bentonites, may be used to adsorb toxic compounds and to remove undesirable substances from solutions and dispersions. A particular advantage is that adsorption capacity and selectivity can be optimised and adjusted to the intended uses by modifying the external surface and the interlayer space.

    The initial interaction of partially hydrolysed solutions ofAl3+ salts with Na-and Ca-montmorillonites (Na-, Ca-mont) has been studied. Mixing of either Na-mont or Ca-mont gels with base-hydrolysed AlCl3 solution resulted in rapid uptake of Al without layer expansion, although >90% of the interlayer cation was exchanged out of the clay into solution. Repeated washing of these gels generated a basal spacing at 1.85 nm, characteristic of that previously reported for Al13-pillared clay gels. However, ion-exchange with very dilute pillaring solutions ([Al]<104M) resulted in direct formation of the expected pillared clay gel. Under some conditions ion-exchange of the Al13 species into the clay gel was reversible; treatment of an Al13 exchanged gel with 0.05 M NaCl solution resulted in collapse of the clay spacing to 1.25 nm. Washing reversed the process, showing that the All3 species remained associated with the clay after Na+ ion-exchange. However, treatment with more concentrated NaCl (2.0 M) resulted in irreversible exchange of the Al species. A mechanism for the reversible pillaring process is proposed.

    A Volciay montmorillonite was treated with solutions of partially hydrolysed TiCl4 in HCl with variable S = [HCl]I[Ti] ratios. Polymeric Ti cations entered the interlamellar space by cation exchange. The size of these polymeric cations depends on the Ti concentrations and on the washing mode of the exchanged clays (with or without dialysis).

    Apr heating at 300°C, two different behaviours were observed. The samples prepared at S≤ 1.32 exhibited clear signs of pillaring but were heterogeneous, with a d001 spacing at 16 Å or above (up to 36 Å), and another at 11 Å, which corresponds to collapsed domains. On the other hand, in the samples prepared at S > 1.32, only collapsed situations (d00l = 10 Å) were found.

    The surface area increased from 89 m2 g1 in the Na montmorillonite to 170–200 m2 g1 in the pillared clays. These values for the pillared samples depend only on the washing mode, whatever the titanium concentration (dialysis increases the surface area). Treatment with HCl alone is enough to increase both the surface area and the total porous volume Vp by 50%, (89 to 144 m2g1 and 0.20 to 0.29 × 106 m3g1 in the reference sample). Vp depends on the titanium concentration. For the smallest pillars, Vp is about 0.36 × 106 m3 g1 as compared to Vp≈0.22 x 106 m3g1 for the highest pillars. The microporous volume (calculated by the t-plot method of De Boer), which is constant for all the pillared clays and more than twice the value for unpillared clays, could be taken as a criterion for successful pillaring.

    Monte Carlo simulations are applied to model the adsorption behavior of iris (1, 10-phenanthroline) metal(II) ([M(phen)3]2+) chelate when it is adsorbed by a clay. A two dimensional sheet of linked [SiO4]4 and [AlO4]5 tetrahedra is used for the model tetrahedral sheet. The thermal distribution of the configurations is obtained using the Metropolis algorithm at 300 K. The binding states are compared when the racemic or enantiomeric chelates are intercalated between the two separated sheets with or without external anions. In the case of adsorption within the cation exchange capacity (CEC), the interlayer distance at the thermal equilibrium is calculated to be 10.5 Å for both the racemic and enantiomeric chelates. On the other hand, in the case of adsorption over the CEC, the interlayer distance is calculated to be 18.5 Å and 20.7 Å for the racemic and enantiomeric chelates in the presence of the external SO42 anions, respectively. The racemic mixture with divalent anions has a mean binding energy 31.1 kJ mol1 lower than the pure enantiomer. The adsorbed racemic mixture forms a bimolecular layer with external anions at the middle of two sheets.

    The interaction of montmorillonite and beidellite with tribridged bi-nuclear iron(III) complex cations of the type [L2Fe2O(RCO2)2]2+ (RCO2=n-CxH2x+1CO2 with x = 1, 4, 9, 11, 13, 15, 17; L = 1,4,7-trimethyl-1,4,7-triazacyclononane, L as a facially coordinating tridentate ligand) has been studied by introducing the complexes into the clay interlayers by cation exchange reactions. Due to the limited stability of the complexes in water, the reactions were carried out in the aprotic solvents acetonitrile and dimethylsulfoxide, which exhibit a great influence on exchange equilibria. The exchange reactions are fully reversible but complex, with respect to both kinetic problems and space requirements. The stability of the complexes in the interlayer region has been proven by vibrational and electronic spectra. In the case of montmorillonite, an additional lattice expansion can be achieved only with small dipolar aprotic solvent molecules. The complexes used in this work can be seen as structural models for hemerythrin and other iron-containing proteins with enzymatic activity.

    A clay-modified electrode composed of [Ni(cyclam)]2+ intercalated into a synthetic saponite supported on a glassy carbon electrode showed electrocatalytic activity at –1.4 V (vs saturated calomel electrode) for the reduction of carbon dioxide to carbon monoxide. The initial current efficiency for carbon monoxide production attained almost 100%. Both a bare glassy carbon and the clay modified electrode produced formic acid when an electrolysed solution contained [Ni(cyclam)]2+ and carbon dioxide.

    Hinokitiol (ß-thujaplicin, C10H12O2) inhibits the growth of many bacteria. However, hinokitiol sublimates easily at a low temperature (53°C), limiting its use as a bactericidal or bacteriostatic agent. In the present paper, we studied the intercalation of hinokitiol in montmorillonite and its extraction from the complex by repeated washing with water and aqueous NaCl solutions. Simple contact of montmorillonite with hinokitiol in ethanol did not form an interlayer complex.

    However, montmorillonite formed the interlayer complex by a two-step intercalation: (1) An interlayer complex of quaternary alkyl ammonium and montmorillonite was prepared. (2) This complex was contacted with hinokitiol dissolved in ethanol. When the hinokitiol-alkyl ammonium-montmorillonite complex was dispersed in water and an aqueous NaCl solution, the complex released hinokitiol gradually. The relationship between the concentration of extracted hinokitiol in solvent and the frequency of washings showed a relatively large extraction at the beginning followed by a gradual extraction. The change in the interlayer spacing of the complex with the frequency of washings corresponded well to that of total hinokitiol extracted from the complex. This would indicate that hinokitiol in the interlayer of montmorillonite was dominantly released, along with slight release ofthat on the external crystalline surface. By releasing hinokitiol gradually in aqueous solutions, the hinokitiol-alkyl ammonium-montmorillonite complex can serve as a bactericidal and bacteriostatic material.

    In the literature, a number of cases have been reported where isotherms associated with binary exchange reactions curried out in the forward and backward directions were markedly different. The causes of this hysteresis phenomenon an still poorly understood. A number of possible mechanisms have been identified, including exchange site heterogeneity on the clay surfaces, differential hydration of cations, intercrystalline swelling hysteresis, and inaccessibility of exchange sites caused by domain or quasi-crystal formation. In and of itself, none of these mechanisms can explain all the available experimental data. However, a recent conceptual model that combines the last two mechanisms appears die, at least qualitatively, to account for the observed effects of factors such as the nature and concentration of the exchanging cations, the clay type, and the extent of dehydration of the clay. In the present article, we describe a set of experimental results for binary exchange reactions involving the cations Na, K, Ca, and Mg on four clays (Wyoming, Camp Berteau, Otay, and Chambers montmorillonites). The experimental results suggest that for a given cation pair and at a fixed total normality in solution, the magnitude of the hysteresis loop increases in the order Wyoming < Camp Berteau < Otay < Chambers. The layer charge of the clays increases in this same order. Light transmission data confirm that the extent of formation or breakdown of quasi-crystals in the clay suspensions depends on the direction of the exchange reaction. These observations are in qualitative agreement with the predictions of the conceptual model.

    A spectroscopic study was carried out of optically active tris(2,2′-bipyridyl) ruthenium(II) dichloride ([Ru(bpy)3]2+) or tris( 1,10-phenanthroline) ruthenium(II) dichloride ([Ru(phen)3]2+) complexes adsorbed on a Laponite clay. Under 5% loading of cation exchange capacity (CEC), the spectral properties of the incorporated Ru(II) complexes exhibit significant differences between the racemic mixture and enantiomer. However, when the loading level was lowered below 0.01% of CEC, they showed identical spectralfeatures in both types of adsorbates. This indicates that intermolecular interaction of probe molecules is promoted within the clay systems. A study of the emission decay profile abo supports this interpretation.

    Aliquots of montmorillonite suspension were mixed with Al(NO3)3 and Fe(NO3)3 at an initial Fe/Al molar ratio (R) of 0, 0.1, 0.3, 0.5, 1.0 (5 mmol of Alg1 clay), and ∞ (5 mmol of Fe g1 of clay); the pH was maintained at 6.0 for 4 h and was adjusted once daily for 4 d. Subsequently all the samples were aged in suspension both at 25°C (100 d) and at 70°C (30 d). Aliquots of each sample collected at different time periods were centrifuge-washed at 15000 g, and the precipitates were examined hy XRD analysis. The hydroxy-Al (R = 0) and the hydroxy-Fe (R=∞) interlayers were not stable either with time or temperature. The mixed hydroxy-Al-Fe components formed stable interlayers in montmorillonite, as indicated by the d-spacing of 17.27 Å (R = 1.0, aged for 30 d at 70°C) which collapsed to around 12.80 Å (control 9.60 Å) on preheating the sample at 550° C for 2 h, indicating that Fe stabilised the hydroxy-Al interlayers in montmorillonite.

    Several mechanisms for the formation of siliceous Al interlayers in expansible layer silicates in acidic soils and sediments have been elucidated and the interlayer adsorption of hydroxy-aluminosilicate (HAS) ions is one of them. Although smectites have been shown to adsorb a substantial amount of HAS ions with a ‘proto-imogolite’ structure, little is known of the interactions between HAS species and vermiculites that have a lower interlayer expansibility. This study was to examine the extent to which HAS ions are adsorbed by vermiculite and the nature of HAS species adsorbed.

    A Ca-saturated, <2 μm fraction of vermiculite (Transvaal, S. Africa) was reacted with an HAS solution, which was prepared from a hydroxy-Alsolution and Si(OH)4 (0.00105 M Si, 0.00204 M Al), for one week at 25° C. The adsorption of Si and Al from the HAS solution was determined. The solid phase reaction products (> 100 Å) were collected by ultrafiltration and examined by XRD and FTIR. The exchangeability of the adsorbed species was also examined.

    Compared with smectites, vermiculite adsorbed a smaller but significant amount of HAS species (0.17 mol Stand 1.01 mol Al kg1 clay) from the HAS solution. The adsorbed species were largely non-exchangeable, since only 7–6% of the adsorbed Si and 6.1% of the adsorbed Al were displaced upon washing with a 0.25 M CaCl2 solution. The low average SUAI molar ratio of the adsorbed species (0.17±:0.03 vs 0.52±0.01 of the original HAS solution) and the low d spacing of the HAS treated vermiculite (14.0 Å) suggested that only those HAS species with a low Si/Al ratio and a small size were adsorbed due to the limited expansibility of vermiculite. This study indicated that the interlayer features of expansible layer silicates determine their capacity to adsorb HAS species and the nature of HAS species adsorbed.

    Muscovite mica was investigated using an electrokinetic technique after ion exchange and low temperature calcination at 300° C. Samples calcined after ion exchange with H+ and K+ showed similar electrokinetic properties to the freshly cleaved sample whereas the Li+ exchanged and calcined sample showed a significant reduction in the magnitude of the zeta potential. This reduction was attributed to an irreversible fixation of Li+ ions to the basal plane of the muscovite mica. Modelling of the electrical double layer properties using a Gouy-Chapman-Stern-Grahame approach indicated that greater than ninety percent of the charge in the lattice had been neutralized by the calcination procedure.

    Intercalation of kaolinites from different localities has shown that the relative yield for the reaction of polar organics and displacement by N-methyl formamide, or for the intercalation of alkali halide, may vary according to the bulk physical properties of the kaolinite. Thus, kaolinites may show intercalation yields ranging from 30% to >95% depending on such factors as particle size and crystallinity. In order to enhance these reactions, a series of grinding processes have been investigated to provide, on average, >90% intercalate yields for two reactions of specific interest: (i) kaolinite intercalation with NMF and (ii) kaolinite intercalation of alkali halides. In each case, these grinding processes provide significant improvements in intercalate yields and for some kaolinites, complete intercalation can be achieved.

    Two interlayer forms of hydrotalcite (HT) containing nitrate (HT-N) or carbonate (HT-C), have been synthesised, and the adsorption of orthophosphate and silicate on the two forms was studied at p. 9. After adsorption the products were examined by X-ray diffraction and infrared analysis (IR). All adsorption isotherms could be fitted to a simple Langmuir equation. The affinity for phosphate is much higher than for silicate. Adsorption maxima are around 1 × 10–3 mole P or Sigl for HT-N and 0.1 × 103 and 0.5 × 103 mole gl, respectively, for phosphate and silicate tuhorbed to HT-C. Adsorption reactions are relatively fast; 90 % of adsorption taking place within the first 20 minutes. Interlayer nitrate, but not carbonate, is exchangeable with phosphate or silicate. Based on thickness of interlayers and IR characteristics of interlayer-phosphate and -silicate, these species do not seem to be present as free’, electrostatically held, ions, but to substitute for triple coordinated hydroxyls (ligand exchange) of the hydroxide layers. Polymerisation of silicate takes place in the interlayer, but not on external particle surfaces. For HT-C, adsorption of phosphate is best explained by bonding at crystal edges, whereas silicate also seems to bond at planar surfaces. Hence pyroaurite-type compounds such as HT appear to be useful for the study of interactions of oxyanions with hydroxide surfaces and for the study of surface-induced polymerisation of silicate.

    Synthetic magadiite, kenyaite and their silicic acids were obtained as spherical and loosely packed aggregates composed of platelike crystals. Particle dispersion for these compounds was investigated by scanning electron microscope (SEM) and electrokinetic characterisation. SEM suggested that the particles of these compounds were dispersed into individual platelets by treatment with KOH, Li OH and NH4OH solutions, but this phenomenon did not occur in Na OH solutions. A kenyaite particle was also found to be more easily delaminated than one of magadiite. The effectiveness of alkaline solutions for dispersion followed the order LiOH > KOH > NH4OH. Measurement of relative acoustophoretic mobility of the particles and solution specific conductivity indicates that the dispersion is due to the increase in electrical double-layer repulsion between particles surrounded by OH anions in solution.

    1. Page 137
      Abstract

      Clays, and above all, bentonites, may be used to adsorb toxic compounds and to remove undesirable substances from solutions and dispersions. A particular advantage is that adsorption capacity and selectivity can be optimised and adjusted to the intended uses by modifying the external surface and the interlayer space.

    2. Page 145
      Abstract

      The initial interaction of partially hydrolysed solutions ofAl3+ salts with Na-and Ca-montmorillonites (Na-, Ca-mont) has been studied. Mixing of either Na-mont or Ca-mont gels with base-hydrolysed AlCl3 solution resulted in rapid uptake of Al without layer expansion, although >90% of the interlayer cation was exchanged out of the clay into solution. Repeated washing of these gels generated a basal spacing at 1.85 nm, characteristic of that previously reported for Al13-pillared clay gels. However, ion-exchange with very dilute pillaring solutions ([Al]<104M) resulted in direct formation of the expected pillared clay gel. Under some conditions ion-exchange of the Al13 species into the clay gel was reversible; treatment of an Al13 exchanged gel with 0.05 M NaCl solution resulted in collapse of the clay spacing to 1.25 nm. Washing reversed the process, showing that the All3 species remained associated with the clay after Na+ ion-exchange. However, treatment with more concentrated NaCl (2.0 M) resulted in irreversible exchange of the Al species. A mechanism for the reversible pillaring process is proposed.

    3. Page 151
      Abstract

      A Volciay montmorillonite was treated with solutions of partially hydrolysed TiCl4 in HCl with variable S = [HCl]I[Ti] ratios. Polymeric Ti cations entered the interlamellar space by cation exchange. The size of these polymeric cations depends on the Ti concentrations and on the washing mode of the exchanged clays (with or without dialysis).

      Apr heating at 300°C, two different behaviours were observed. The samples prepared at S≤ 1.32 exhibited clear signs of pillaring but were heterogeneous, with a d001 spacing at 16 Å or above (up to 36 Å), and another at 11 Å, which corresponds to collapsed domains. On the other hand, in the samples prepared at S > 1.32, only collapsed situations (d00l = 10 Å) were found.

      The surface area increased from 89 m2 g1 in the Na montmorillonite to 170–200 m2 g1 in the pillared clays. These values for the pillared samples depend only on the washing mode, whatever the titanium concentration (dialysis increases the surface area). Treatment with HCl alone is enough to increase both the surface area and the total porous volume Vp by 50%, (89 to 144 m2g1 and 0.20 to 0.29 × 106 m3g1 in the reference sample). Vp depends on the titanium concentration. For the smallest pillars, Vp is about 0.36 × 106 m3 g1 as compared to Vp≈0.22 x 106 m3g1 for the highest pillars. The microporous volume (calculated by the t-plot method of De Boer), which is constant for all the pillared clays and more than twice the value for unpillared clays, could be taken as a criterion for successful pillaring.

    4. Page 157
      Abstract

      Monte Carlo simulations are applied to model the adsorption behavior of iris (1, 10-phenanthroline) metal(II) ([M(phen)3]2+) chelate when it is adsorbed by a clay. A two dimensional sheet of linked [SiO4]4 and [AlO4]5 tetrahedra is used for the model tetrahedral sheet. The thermal distribution of the configurations is obtained using the Metropolis algorithm at 300 K. The binding states are compared when the racemic or enantiomeric chelates are intercalated between the two separated sheets with or without external anions. In the case of adsorption within the cation exchange capacity (CEC), the interlayer distance at the thermal equilibrium is calculated to be 10.5 Å for both the racemic and enantiomeric chelates. On the other hand, in the case of adsorption over the CEC, the interlayer distance is calculated to be 18.5 Å and 20.7 Å for the racemic and enantiomeric chelates in the presence of the external SO42 anions, respectively. The racemic mixture with divalent anions has a mean binding energy 31.1 kJ mol1 lower than the pure enantiomer. The adsorbed racemic mixture forms a bimolecular layer with external anions at the middle of two sheets.

    5. Page 161
      Abstract

      The interaction of montmorillonite and beidellite with tribridged bi-nuclear iron(III) complex cations of the type [L2Fe2O(RCO2)2]2+ (RCO2=n-CxH2x+1CO2 with x = 1, 4, 9, 11, 13, 15, 17; L = 1,4,7-trimethyl-1,4,7-triazacyclononane, L as a facially coordinating tridentate ligand) has been studied by introducing the complexes into the clay interlayers by cation exchange reactions. Due to the limited stability of the complexes in water, the reactions were carried out in the aprotic solvents acetonitrile and dimethylsulfoxide, which exhibit a great influence on exchange equilibria. The exchange reactions are fully reversible but complex, with respect to both kinetic problems and space requirements. The stability of the complexes in the interlayer region has been proven by vibrational and electronic spectra. In the case of montmorillonite, an additional lattice expansion can be achieved only with small dipolar aprotic solvent molecules. The complexes used in this work can be seen as structural models for hemerythrin and other iron-containing proteins with enzymatic activity.

    6. Page 166
      Abstract

      A clay-modified electrode composed of [Ni(cyclam)]2+ intercalated into a synthetic saponite supported on a glassy carbon electrode showed electrocatalytic activity at –1.4 V (vs saturated calomel electrode) for the reduction of carbon dioxide to carbon monoxide. The initial current efficiency for carbon monoxide production attained almost 100%. Both a bare glassy carbon and the clay modified electrode produced formic acid when an electrolysed solution contained [Ni(cyclam)]2+ and carbon dioxide.

    7. Page 169
      Abstract

      Hinokitiol (ß-thujaplicin, C10H12O2) inhibits the growth of many bacteria. However, hinokitiol sublimates easily at a low temperature (53°C), limiting its use as a bactericidal or bacteriostatic agent. In the present paper, we studied the intercalation of hinokitiol in montmorillonite and its extraction from the complex by repeated washing with water and aqueous NaCl solutions. Simple contact of montmorillonite with hinokitiol in ethanol did not form an interlayer complex.

      However, montmorillonite formed the interlayer complex by a two-step intercalation: (1) An interlayer complex of quaternary alkyl ammonium and montmorillonite was prepared. (2) This complex was contacted with hinokitiol dissolved in ethanol. When the hinokitiol-alkyl ammonium-montmorillonite complex was dispersed in water and an aqueous NaCl solution, the complex released hinokitiol gradually. The relationship between the concentration of extracted hinokitiol in solvent and the frequency of washings showed a relatively large extraction at the beginning followed by a gradual extraction. The change in the interlayer spacing of the complex with the frequency of washings corresponded well to that of total hinokitiol extracted from the complex. This would indicate that hinokitiol in the interlayer of montmorillonite was dominantly released, along with slight release ofthat on the external crystalline surface. By releasing hinokitiol gradually in aqueous solutions, the hinokitiol-alkyl ammonium-montmorillonite complex can serve as a bactericidal and bacteriostatic material.

    8. Page 173
      Abstract

      In the literature, a number of cases have been reported where isotherms associated with binary exchange reactions curried out in the forward and backward directions were markedly different. The causes of this hysteresis phenomenon an still poorly understood. A number of possible mechanisms have been identified, including exchange site heterogeneity on the clay surfaces, differential hydration of cations, intercrystalline swelling hysteresis, and inaccessibility of exchange sites caused by domain or quasi-crystal formation. In and of itself, none of these mechanisms can explain all the available experimental data. However, a recent conceptual model that combines the last two mechanisms appears die, at least qualitatively, to account for the observed effects of factors such as the nature and concentration of the exchanging cations, the clay type, and the extent of dehydration of the clay. In the present article, we describe a set of experimental results for binary exchange reactions involving the cations Na, K, Ca, and Mg on four clays (Wyoming, Camp Berteau, Otay, and Chambers montmorillonites). The experimental results suggest that for a given cation pair and at a fixed total normality in solution, the magnitude of the hysteresis loop increases in the order Wyoming < Camp Berteau < Otay < Chambers. The layer charge of the clays increases in this same order. Light transmission data confirm that the extent of formation or breakdown of quasi-crystals in the clay suspensions depends on the direction of the exchange reaction. These observations are in qualitative agreement with the predictions of the conceptual model.

    9. Page 178
      Abstract

      A spectroscopic study was carried out of optically active tris(2,2′-bipyridyl) ruthenium(II) dichloride ([Ru(bpy)3]2+) or tris( 1,10-phenanthroline) ruthenium(II) dichloride ([Ru(phen)3]2+) complexes adsorbed on a Laponite clay. Under 5% loading of cation exchange capacity (CEC), the spectral properties of the incorporated Ru(II) complexes exhibit significant differences between the racemic mixture and enantiomer. However, when the loading level was lowered below 0.01% of CEC, they showed identical spectralfeatures in both types of adsorbates. This indicates that intermolecular interaction of probe molecules is promoted within the clay systems. A study of the emission decay profile abo supports this interpretation.

    10. Page 183
      Abstract

      Aliquots of montmorillonite suspension were mixed with Al(NO3)3 and Fe(NO3)3 at an initial Fe/Al molar ratio (R) of 0, 0.1, 0.3, 0.5, 1.0 (5 mmol of Alg1 clay), and ∞ (5 mmol of Fe g1 of clay); the pH was maintained at 6.0 for 4 h and was adjusted once daily for 4 d. Subsequently all the samples were aged in suspension both at 25°C (100 d) and at 70°C (30 d). Aliquots of each sample collected at different time periods were centrifuge-washed at 15000 g, and the precipitates were examined hy XRD analysis. The hydroxy-Al (R = 0) and the hydroxy-Fe (R=∞) interlayers were not stable either with time or temperature. The mixed hydroxy-Al-Fe components formed stable interlayers in montmorillonite, as indicated by the d-spacing of 17.27 Å (R = 1.0, aged for 30 d at 70°C) which collapsed to around 12.80 Å (control 9.60 Å) on preheating the sample at 550° C for 2 h, indicating that Fe stabilised the hydroxy-Al interlayers in montmorillonite.

    11. Page 187
      Abstract

      Several mechanisms for the formation of siliceous Al interlayers in expansible layer silicates in acidic soils and sediments have been elucidated and the interlayer adsorption of hydroxy-aluminosilicate (HAS) ions is one of them. Although smectites have been shown to adsorb a substantial amount of HAS ions with a ‘proto-imogolite’ structure, little is known of the interactions between HAS species and vermiculites that have a lower interlayer expansibility. This study was to examine the extent to which HAS ions are adsorbed by vermiculite and the nature of HAS species adsorbed.

      A Ca-saturated, <2 μm fraction of vermiculite (Transvaal, S. Africa) was reacted with an HAS solution, which was prepared from a hydroxy-Alsolution and Si(OH)4 (0.00105 M Si, 0.00204 M Al), for one week at 25° C. The adsorption of Si and Al from the HAS solution was determined. The solid phase reaction products (> 100 Å) were collected by ultrafiltration and examined by XRD and FTIR. The exchangeability of the adsorbed species was also examined.

      Compared with smectites, vermiculite adsorbed a smaller but significant amount of HAS species (0.17 mol Stand 1.01 mol Al kg1 clay) from the HAS solution. The adsorbed species were largely non-exchangeable, since only 7–6% of the adsorbed Si and 6.1% of the adsorbed Al were displaced upon washing with a 0.25 M CaCl2 solution. The low average SUAI molar ratio of the adsorbed species (0.17±:0.03 vs 0.52±0.01 of the original HAS solution) and the low d spacing of the HAS treated vermiculite (14.0 Å) suggested that only those HAS species with a low Si/Al ratio and a small size were adsorbed due to the limited expansibility of vermiculite. This study indicated that the interlayer features of expansible layer silicates determine their capacity to adsorb HAS species and the nature of HAS species adsorbed.

    12. Page 192
      Abstract

      Muscovite mica was investigated using an electrokinetic technique after ion exchange and low temperature calcination at 300° C. Samples calcined after ion exchange with H+ and K+ showed similar electrokinetic properties to the freshly cleaved sample whereas the Li+ exchanged and calcined sample showed a significant reduction in the magnitude of the zeta potential. This reduction was attributed to an irreversible fixation of Li+ ions to the basal plane of the muscovite mica. Modelling of the electrical double layer properties using a Gouy-Chapman-Stern-Grahame approach indicated that greater than ninety percent of the charge in the lattice had been neutralized by the calcination procedure.

    13. Page 196
      Abstract

      Intercalation of kaolinites from different localities has shown that the relative yield for the reaction of polar organics and displacement by N-methyl formamide, or for the intercalation of alkali halide, may vary according to the bulk physical properties of the kaolinite. Thus, kaolinites may show intercalation yields ranging from 30% to >95% depending on such factors as particle size and crystallinity. In order to enhance these reactions, a series of grinding processes have been investigated to provide, on average, >90% intercalate yields for two reactions of specific interest: (i) kaolinite intercalation with NMF and (ii) kaolinite intercalation of alkali halides. In each case, these grinding processes provide significant improvements in intercalate yields and for some kaolinites, complete intercalation can be achieved.

    14. Page 201
      Abstract

      Two interlayer forms of hydrotalcite (HT) containing nitrate (HT-N) or carbonate (HT-C), have been synthesised, and the adsorption of orthophosphate and silicate on the two forms was studied at p. 9. After adsorption the products were examined by X-ray diffraction and infrared analysis (IR). All adsorption isotherms could be fitted to a simple Langmuir equation. The affinity for phosphate is much higher than for silicate. Adsorption maxima are around 1 × 10–3 mole P or Sigl for HT-N and 0.1 × 103 and 0.5 × 103 mole gl, respectively, for phosphate and silicate tuhorbed to HT-C. Adsorption reactions are relatively fast; 90 % of adsorption taking place within the first 20 minutes. Interlayer nitrate, but not carbonate, is exchangeable with phosphate or silicate. Based on thickness of interlayers and IR characteristics of interlayer-phosphate and -silicate, these species do not seem to be present as free’, electrostatically held, ions, but to substitute for triple coordinated hydroxyls (ligand exchange) of the hydroxide layers. Polymerisation of silicate takes place in the interlayer, but not on external particle surfaces. For HT-C, adsorption of phosphate is best explained by bonding at crystal edges, whereas silicate also seems to bond at planar surfaces. Hence pyroaurite-type compounds such as HT appear to be useful for the study of interactions of oxyanions with hydroxide surfaces and for the study of surface-induced polymerisation of silicate.

    15. Page 207
      Abstract

      Synthetic magadiite, kenyaite and their silicic acids were obtained as spherical and loosely packed aggregates composed of platelike crystals. Particle dispersion for these compounds was investigated by scanning electron microscope (SEM) and electrokinetic characterisation. SEM suggested that the particles of these compounds were dispersed into individual platelets by treatment with KOH, Li OH and NH4OH solutions, but this phenomenon did not occur in Na OH solutions. A kenyaite particle was also found to be more easily delaminated than one of magadiite. The effectiveness of alkaline solutions for dispersion followed the order LiOH > KOH > NH4OH. Measurement of relative acoustophoretic mobility of the particles and solution specific conductivity indicates that the dispersion is due to the increase in electrical double-layer repulsion between particles surrounded by OH anions in solution.

  8. Page 211
    Abstract

    An understanding of the crystal structures, syntheses and transformations of clay minerals, all of which are discussed under this Theme, provides a rational basis for future applications of clays as well as for descriptions of their origin and place in the natural world.

    A model is proposed based on high-temperature crystal structural data to explain oxidation in chlorite, and high-pressure differential thermal analytical (HP - DTA) data are used to examine the effect of Mg vs Fe substitution on the metastable phase relations of chlorite. A structure refinement of a clinochlore—IIb-4 crystal at 550° Cshowed that the M4 site exhibits large thermal expansion but contains small trivalent cations. For a crystal containing iron, any oxidation at this site is minimised by site expansion. The ordering of trivalent cations into M3 is destabilising due to electrostatic repulsions originating in other M3 sites, M4 sites, and elsewhere in the structure. Thus, the chlorite structure constrains the amount of iron oxidation possible.

    Both clinochlore (Mg) and chamosite (Fe) dehydroxylate over a narrow temperature range to a ‘modified chlorite’ structure at conditions of P(H2O) < P(total). At conditions of P(total) = P(H2O), clinochlore dehydroxylates over a small temperature interval, but chamosite requires a wide temperature interval. For P(total) = P(H2O) and the Mg system, a metastable invariant point at about 855° C and 50 bars exists with four emanating reactions involving a modified chlorite structure, vapour, liquid, and an anhydrous phase believed to be transitional between modified chlorite and the stable product, olivine. The phase relations of chamosite are compared, and the topologies of both systems are discussed.

    An infrared emission cell, which is capable of operation up to 1500°C, is coupled to a Fourier transform infrared spectrometer and used to measure the emission spectra of a series of 1:1 and 2:1 clay minerals with both dioctahedral and trioctahedral structures. The clay minerals studied include kaolinite, pyrophyllite, tremolite, anthophyllite and rectorite. Emission spectra were obtained at a nominal resolution of 4 cm-1 over a temperature range of 100–1200°C at 100 degree intervals. Importantly, the analysis is done in situ and the clay reactions are observed as they are taking place. Excellent correspondence was obtained between the emission and absorption spectra of the clay minerals. Dehydroxylation temperatures were measured corresponding to the loss of the v (OH) hand. Dehydroxylation is temperature dependent and is related to the hydroxyl structures in the clays. The OH groups in the palygorskite clays were found to be lost in stages. The hydroxyl groups in tremolite were retained up to 1100°C which may be attributed to intracrystalline hydroxyls. Thermal transformation of clay minerals may be divided into low temperature transformations (<400°C), intermediate reactions (400–700°C) and high temperature reactions where topotactic reactions produce the high temperature phases. The dioctahedral clays form mullite and silica at high temperatures, as evidenced by the 1160cm-1 peak. Saponite, sepiolite and palygorskite thermally transform to talc-like structures which recrystallise to enstatite and silica above 800°C.

    The objective of this study was to investigate the characteristic change of smectite at the initial stage of illitisation. The experiment was conducted on a montmorillonite type of dioctahedral smectite. The smectite was mixed with KCl solution or synthetic sea water and maintained at 90°C and 180°C, for up to 3years. The treated samples were examined by XRD to determine the interstratified structure of illite/smectite (I/S). Significant illitisation was not identified under these conditions. Only random mixed-layer of I/S having 80–90% smectite was observed in the samples treated with high potassium concentration at 180°C.

    For these samples, the Greene-Kelly test was also conducted to investigate the charge location in expandable layers. The interstratified structure of the tetrahedrally-charged layer/the octahedrally-charged layer (T/O) was determined by comparison with calculated XRD patterns. The proportion of the tetrahedral charge increased in some samples. This result is in agreement with a previous study on natural samples. The structure of interstratification of T/O varied with the potassium concentrations of the solutions. In the samples treated with high potassium concentration, regular type of T/ O mixed-layer was observed. The mixed-layer of this type can be regarded as a precursor of regular I/S mixed-layer.

    Mössbauer spectroscopic characterisation of vermiculites has previously shown that an Fe-containing material with spectral parameters similar to those for ilmenite can occur in the interlayers of the Vermiculite structure. This leads to the suggestion that the Vermiculite and Fe TiO3 structures might be coherent through a common anion framework. To further investigate the possibility of the existence of cations which might act as FeTiO3 precursors in the Vermiculite interlayers, vermiculite was intercalated with Fe2+, Fe3+ and Sn4+ to simulate Fe and Ti ilmenite cations. Their interaction with the tetrahedral Fe3+ present in the vermiculite was determined using 57Fe and ll9Sn Mössbauer spectroscopies.

    The Mg-saturated vermiculite (Llano) used for the experiments contained only tetrahedrally coordinated Fe (IVFe3+) which prevented any tetrabedral-Fe/octahedral-Fe interactions within the structure. The low quadrupole splitting value for the vermiculite IVFe3+ confirmed this lack of electronic interaction. As might be expected, all the intercalated cations were octahedrally coordinated in the interlayers. Compared with the values for Mg-vermiculite, the isomer shift parameters for the vermiculite IVFe3+ component were lower when the vermiculite was intercalated with Fe cations than when intercalated with Mg but similar for Sn. However the recoilless fraction for the IVFe3+ site changed significantly with different cations present. The absolute absorption intensity was 0.38% with Mg2+ present compared with 0.41% with Fe2+ and 0.22% with Sn4+. This indicates that the type of interlayer cation influences the bonding strengths between adjacent sites, but not necessarily the electronic behaviour at the tetrahedral site. Upon intercalation of the Fe2+ into the structure, partial oxidation of this Fe2+ occurred. However, once the Fe2+ cations were intercalated into the vermiculite they were very resistant to oxidation, providing evidence that trasferring electrons to the vermiculite layer does not readily occur.

    The vapour pressures of different states of hydration of vermiculites (particle size fraction 5–20 μm) were determined by following dehydration with simultaneous thermoanalysis (TG/DSC). The pH2O at the phase boundary was defined by reducing the total pressure after determining the relative humidity at atmospheric pressure. At a given pressure, temperatures were increased with a constant heating rate. A critical aspect of this method is that the homogeneity of pH2O close to the surface of the dehydrating vermiculites is affected by the rate of heating. In order to eliminate this problem, critical temperatures of the phase transitions were extrapolated to a zero heating rate from measurements made at heating rates varying from 0.1 to 10 Kmin−1. Considering the inaccuracy of integrating DSCpeaks, the enthalpies of dehydration of the vermiculites were determined from the onset temperatures of dehydration as exhibited by DSC curves by using the Clausius-Clapeyron equation which describes the temperature-pressure relation at equilibrium.

    Stability diagrams of the different states of hydration were determined for vermiculites of different origin (phlogopitic-biotitic) as well as of different cation saturation. The four different states ofhydration found for the Mg-saturated phlogopitic vermiculite have been characterised by Debye-Scherrer X-ray diffractograms.

    Dehydration and rehydration of a phlogopitic vermiculite (particle size fraction 5–20 × 10−6 m) saturated with alkaline earth cations were investigated in the temperature range between 20 °C and 300° C, using a position sensitive X- ray detector and a heating device. At heating or cooling rates of 0.1° C min−1, changes of the state of hydration could be observed in a sequence of temperature steps of 0.2 °C.

    As an example the results obtained by investigating the transition between states of hydration of Sr-vermiculite are emphasised. Two types of transitions could be observed: Those in which the number of water layers is preserved (monomolecular, d002: 12.1 Å ↔ 1 2.3 Å; bimolecular, 15.2 Å ↔ 15.3 Å ↔ 15.4 Å) and the more drastic structural transformation accompanied by loss or gain of a water layer (d002 12.3 Å ↔ 15.2 Å).

    Among those of the first type two cases can be further distinguished in that transition occurs either continuously with the integrality of d001 reflections being sustained (d002:12.1Å ↔ 12.3 Å) or stepwise (d002 15–2Å ↔ 15.3 Å ↔ 15.4 Å). The more drastic type of transition passes through interstratification, which at a 1:1 proportion of the differently hydrated interlayers forms a superstructure with an integral series of d001 reflections.

    As the comparison of variously saturated vermiculites revealed, this general pattern of transformation exists irrespective of the species but varying in critical temperatures according to the hydration energy of the saturating cation.

    Hydrothermal synthesis is an attractive and much used method to produce well-defined and pure phyllosilicates. One of the most important parameters controlling the synthesis is the acido-basic character of the reaction medium.

    An acidic and fluorinated medium was applied to the (SiO2, Al2O3, CF, H2O)-system (C=Li, Na, NH4), for the synthesis of silicates of general chemical formula: Cx (Al2) (Si4-xAlx) O10 [(OH)2_yFy];0 < x < 0.9 ; 0 <y <2.

    It was found that the smaller the ionic radius of the monovalent C+ ions and the higher the acidity of the hydrogel and the HF concentration the smaller the tetrahedral substitution degree x.

    As shown by Scanning Electron Microscopy, small crystal whose shapes probably depend on the size of the interlayer cations were obtained. They were characterised by X-Ray Diffraction,29Si, 27Al and Fig. 19F MAS NMR Spectroscopy and Thermal Analysis.

    Likewise, the (SiO2, Ga2O3, NH4F,H2O)-system leads to (Ga) -layersilicates.

    In particular, the fluorinated layer silicates show controlled chemical compositions, are well crystallised, homogeneous and thermally stable. Moreover, the acidic medium is the only one generally leading to beidellite-type layer silicates in which the degree of tetrahedral substitution is controllable.

    The growth process of sericite crystals in a single sample collected from the hydrothermal sericite deposit at Mitoya, Shimane Prefecture, Japan is discussed based on electron microscopic observation. The sample was divided into 6 particle size clay fractions. The sericite fractions contain lath-shaped (<2 μm particles) and irregular-shaped (>2 μm particles) crystals. Microtopographic observation of crystal surfaces shows the lath-shaped crystals to be characterised by elongated spiral step-patterns, while complicated step patterns with some dislocation centers are observed on the irregular-shaped particles. Based on these results, it was concluded that the coarsening of sericite crystals is interpreted to be controlled by aggregation of smaller particles.

    In the presence of K+ and HPO42− ions, kaolinite, montmorillonite, micas (sericite) and gibbsite are transformed to taranakite. The reaction rate is at maximum in the pH-range 3.2–4.3. At p. 2, phase KAllHpo20 is formed, at p. 5 phase KAl2(OH)(PO4)22H2O. The latter is chemically identical with iron free leukophosphite, but differs in X-ray and structural data. In the absence of silica and at elevated temperatures (100°C) KAl(HPO4)2H2O transforms quantitatively to taranakite, even at p. 2.0. With smectites and micas the nucléation of taranakite occurs more rapidly than with kaolinite, which may be earned by epitactic nucléation due to some structural similarities of these minerals with taranakite. Taranakite {(K3Al5(HPO4)6(PO4)2(H2O)12}(H2O)6 has a trigonal rhombohedral layer lattice (space group R3 c- D63d, a = 8.7025(11)Å c = 95.050(13) Å). The individual 13.8 Å thick layers are separated by planar monolayers of 6H2O molecules per formula unit. The remaining 12H2O molecules belong to (AlO, OH2) -octahedra and are part of the rigid layers. The potassium ions are positioned in the central plane of the layers. They cannot undergo cation exchange reactions. The interlayer water is easily removed, thus forming jrancoanellite K3Al5(HPO4)6(PO4)2(H2O)12. The experimental conditions for dehydration are highly affected by crystal size, crystallinity, heating rate and isomorphous substitutions (62–80°C at ambient conditions or 20°C at pH2O ≤ 0.66 Pa). As long as traces of H2O remain in the interlayer region, rehydration to taranakite is possible. With regard to this aspect the reaction taranakite =francoanellite + nH2O resembles the reaction of hydrated halloysite = halloysite and H2O

    Kaolinite:alkali halide intenalates have been success fully prepared by grinding the salt with kaolinite in the absence of water as well as by reaction of kaolinite with concentrated aqueous solution of alkali halide. For the dry grinding method the rate of intercalation is greatest for salts with the lowest melting point. The basal dimensions of the intercalates increase with increasing size of ion. The layered intercalate structure survives the dehydroxylation of the kaolinite at 500–600° C, at which point the excess salt can be dissolved to give an XRD-amorphous material. For each alkali metal, M, the amorphous product has approximate stoichiometry MAlSiO4 and transforms to a crystalline phase of the same stoichiometry at surprisingly low temperature.

    Kaolinite-.alkaline earth halide intercalates and their derivatives have also been prepared, but only by the concentrated aqueous solution method. These intercalates display little long-range order, but can still be dehydroxylated to yield upon rinsingXRD-amorphous materials with approximate stoichiometry MAl2Si2 O8 (M = Ca, Sr and Ba). Upon heating these materials transform initially to adopt the hexacelsian structure then at higher temperature transform to feldspar- related structures.

    While it is possible to react other metal halides, such as halides of transition metals and magnesium, with kaolinite to achieve at least partial intercalation the subsequent reaction steps are generally not successful. This is usually due to the instability of these salts during the dehydroxylation step resulting in de-intercalation.

    Synthetic pyroaurites with Mg:Fe(III) ratios between 1.54 and 2.56 have been investigated by X-ray diffraction, ultraviolet-visible, infrared, and Mössbauer spectroscopy. All diffractions may be indexed using a hexagonal cell having a0= 0.311 nm and c0 values between 2.310 and 2.230 nm. The value of c0 is linearly correlated with the layer charge (Fe(III)/(Mg+Fe(III)) ratio). No superstructure reflections due to regular ordering of Mg and Fe atoms in the metal hydroxide layer were observed. Infrared spectroscopy did not indicate changes in the structure within the Mg:Fe(III) ratio investigated. Ultraviolet—visible spectroscopy and Mössbauer spectra show changes which are interpreted at mainly caused by the presence of a distribution of sites for Fe(III) in the metal hydroxide layer with slightly different surroundings.

    1. Page 213
      Abstract

      A model is proposed based on high-temperature crystal structural data to explain oxidation in chlorite, and high-pressure differential thermal analytical (HP - DTA) data are used to examine the effect of Mg vs Fe substitution on the metastable phase relations of chlorite. A structure refinement of a clinochlore—IIb-4 crystal at 550° Cshowed that the M4 site exhibits large thermal expansion but contains small trivalent cations. For a crystal containing iron, any oxidation at this site is minimised by site expansion. The ordering of trivalent cations into M3 is destabilising due to electrostatic repulsions originating in other M3 sites, M4 sites, and elsewhere in the structure. Thus, the chlorite structure constrains the amount of iron oxidation possible.

      Both clinochlore (Mg) and chamosite (Fe) dehydroxylate over a narrow temperature range to a ‘modified chlorite’ structure at conditions of P(H2O) < P(total). At conditions of P(total) = P(H2O), clinochlore dehydroxylates over a small temperature interval, but chamosite requires a wide temperature interval. For P(total) = P(H2O) and the Mg system, a metastable invariant point at about 855° C and 50 bars exists with four emanating reactions involving a modified chlorite structure, vapour, liquid, and an anhydrous phase believed to be transitional between modified chlorite and the stable product, olivine. The phase relations of chamosite are compared, and the topologies of both systems are discussed.

    2. Page 219
      Abstract

      An infrared emission cell, which is capable of operation up to 1500°C, is coupled to a Fourier transform infrared spectrometer and used to measure the emission spectra of a series of 1:1 and 2:1 clay minerals with both dioctahedral and trioctahedral structures. The clay minerals studied include kaolinite, pyrophyllite, tremolite, anthophyllite and rectorite. Emission spectra were obtained at a nominal resolution of 4 cm-1 over a temperature range of 100–1200°C at 100 degree intervals. Importantly, the analysis is done in situ and the clay reactions are observed as they are taking place. Excellent correspondence was obtained between the emission and absorption spectra of the clay minerals. Dehydroxylation temperatures were measured corresponding to the loss of the v (OH) hand. Dehydroxylation is temperature dependent and is related to the hydroxyl structures in the clays. The OH groups in the palygorskite clays were found to be lost in stages. The hydroxyl groups in tremolite were retained up to 1100°C which may be attributed to intracrystalline hydroxyls. Thermal transformation of clay minerals may be divided into low temperature transformations (<400°C), intermediate reactions (400–700°C) and high temperature reactions where topotactic reactions produce the high temperature phases. The dioctahedral clays form mullite and silica at high temperatures, as evidenced by the 1160cm-1 peak. Saponite, sepiolite and palygorskite thermally transform to talc-like structures which recrystallise to enstatite and silica above 800°C.

    3. Page 225
      Abstract

      The objective of this study was to investigate the characteristic change of smectite at the initial stage of illitisation. The experiment was conducted on a montmorillonite type of dioctahedral smectite. The smectite was mixed with KCl solution or synthetic sea water and maintained at 90°C and 180°C, for up to 3years. The treated samples were examined by XRD to determine the interstratified structure of illite/smectite (I/S). Significant illitisation was not identified under these conditions. Only random mixed-layer of I/S having 80–90% smectite was observed in the samples treated with high potassium concentration at 180°C.

      For these samples, the Greene-Kelly test was also conducted to investigate the charge location in expandable layers. The interstratified structure of the tetrahedrally-charged layer/the octahedrally-charged layer (T/O) was determined by comparison with calculated XRD patterns. The proportion of the tetrahedral charge increased in some samples. This result is in agreement with a previous study on natural samples. The structure of interstratification of T/O varied with the potassium concentrations of the solutions. In the samples treated with high potassium concentration, regular type of T/ O mixed-layer was observed. The mixed-layer of this type can be regarded as a precursor of regular I/S mixed-layer.

    4. Page 231
      Abstract

      Mössbauer spectroscopic characterisation of vermiculites has previously shown that an Fe-containing material with spectral parameters similar to those for ilmenite can occur in the interlayers of the Vermiculite structure. This leads to the suggestion that the Vermiculite and Fe TiO3 structures might be coherent through a common anion framework. To further investigate the possibility of the existence of cations which might act as FeTiO3 precursors in the Vermiculite interlayers, vermiculite was intercalated with Fe2+, Fe3+ and Sn4+ to simulate Fe and Ti ilmenite cations. Their interaction with the tetrahedral Fe3+ present in the vermiculite was determined using 57Fe and ll9Sn Mössbauer spectroscopies.

      The Mg-saturated vermiculite (Llano) used for the experiments contained only tetrahedrally coordinated Fe (IVFe3+) which prevented any tetrabedral-Fe/octahedral-Fe interactions within the structure. The low quadrupole splitting value for the vermiculite IVFe3+ confirmed this lack of electronic interaction. As might be expected, all the intercalated cations were octahedrally coordinated in the interlayers. Compared with the values for Mg-vermiculite, the isomer shift parameters for the vermiculite IVFe3+ component were lower when the vermiculite was intercalated with Fe cations than when intercalated with Mg but similar for Sn. However the recoilless fraction for the IVFe3+ site changed significantly with different cations present. The absolute absorption intensity was 0.38% with Mg2+ present compared with 0.41% with Fe2+ and 0.22% with Sn4+. This indicates that the type of interlayer cation influences the bonding strengths between adjacent sites, but not necessarily the electronic behaviour at the tetrahedral site. Upon intercalation of the Fe2+ into the structure, partial oxidation of this Fe2+ occurred. However, once the Fe2+ cations were intercalated into the vermiculite they were very resistant to oxidation, providing evidence that trasferring electrons to the vermiculite layer does not readily occur.

    5. Page 235
      Abstract

      The vapour pressures of different states of hydration of vermiculites (particle size fraction 5–20 μm) were determined by following dehydration with simultaneous thermoanalysis (TG/DSC). The pH2O at the phase boundary was defined by reducing the total pressure after determining the relative humidity at atmospheric pressure. At a given pressure, temperatures were increased with a constant heating rate. A critical aspect of this method is that the homogeneity of pH2O close to the surface of the dehydrating vermiculites is affected by the rate of heating. In order to eliminate this problem, critical temperatures of the phase transitions were extrapolated to a zero heating rate from measurements made at heating rates varying from 0.1 to 10 Kmin−1. Considering the inaccuracy of integrating DSCpeaks, the enthalpies of dehydration of the vermiculites were determined from the onset temperatures of dehydration as exhibited by DSC curves by using the Clausius-Clapeyron equation which describes the temperature-pressure relation at equilibrium.

      Stability diagrams of the different states of hydration were determined for vermiculites of different origin (phlogopitic-biotitic) as well as of different cation saturation. The four different states ofhydration found for the Mg-saturated phlogopitic vermiculite have been characterised by Debye-Scherrer X-ray diffractograms.

    6. Page 240
      Abstract

      Dehydration and rehydration of a phlogopitic vermiculite (particle size fraction 5–20 × 10−6 m) saturated with alkaline earth cations were investigated in the temperature range between 20 °C and 300° C, using a position sensitive X- ray detector and a heating device. At heating or cooling rates of 0.1° C min−1, changes of the state of hydration could be observed in a sequence of temperature steps of 0.2 °C.

      As an example the results obtained by investigating the transition between states of hydration of Sr-vermiculite are emphasised. Two types of transitions could be observed: Those in which the number of water layers is preserved (monomolecular, d002: 12.1 Å ↔ 1 2.3 Å; bimolecular, 15.2 Å ↔ 15.3 Å ↔ 15.4 Å) and the more drastic structural transformation accompanied by loss or gain of a water layer (d002 12.3 Å ↔ 15.2 Å).

      Among those of the first type two cases can be further distinguished in that transition occurs either continuously with the integrality of d001 reflections being sustained (d002:12.1Å ↔ 12.3 Å) or stepwise (d002 15–2Å ↔ 15.3 Å ↔ 15.4 Å). The more drastic type of transition passes through interstratification, which at a 1:1 proportion of the differently hydrated interlayers forms a superstructure with an integral series of d001 reflections.

      As the comparison of variously saturated vermiculites revealed, this general pattern of transformation exists irrespective of the species but varying in critical temperatures according to the hydration energy of the saturating cation.

    7. Page 243
      Abstract

      Hydrothermal synthesis is an attractive and much used method to produce well-defined and pure phyllosilicates. One of the most important parameters controlling the synthesis is the acido-basic character of the reaction medium.

      An acidic and fluorinated medium was applied to the (SiO2, Al2O3, CF, H2O)-system (C=Li, Na, NH4), for the synthesis of silicates of general chemical formula: Cx (Al2) (Si4-xAlx) O10 [(OH)2_yFy];0 < x < 0.9 ; 0 <y <2.

      It was found that the smaller the ionic radius of the monovalent C+ ions and the higher the acidity of the hydrogel and the HF concentration the smaller the tetrahedral substitution degree x.

      As shown by Scanning Electron Microscopy, small crystal whose shapes probably depend on the size of the interlayer cations were obtained. They were characterised by X-Ray Diffraction,29Si, 27Al and Fig. 19F MAS NMR Spectroscopy and Thermal Analysis.

      Likewise, the (SiO2, Ga2O3, NH4F,H2O)-system leads to (Ga) -layersilicates.

      In particular, the fluorinated layer silicates show controlled chemical compositions, are well crystallised, homogeneous and thermally stable. Moreover, the acidic medium is the only one generally leading to beidellite-type layer silicates in which the degree of tetrahedral substitution is controllable.

    8. Page 249
      Abstract

      The growth process of sericite crystals in a single sample collected from the hydrothermal sericite deposit at Mitoya, Shimane Prefecture, Japan is discussed based on electron microscopic observation. The sample was divided into 6 particle size clay fractions. The sericite fractions contain lath-shaped (<2 μm particles) and irregular-shaped (>2 μm particles) crystals. Microtopographic observation of crystal surfaces shows the lath-shaped crystals to be characterised by elongated spiral step-patterns, while complicated step patterns with some dislocation centers are observed on the irregular-shaped particles. Based on these results, it was concluded that the coarsening of sericite crystals is interpreted to be controlled by aggregation of smaller particles.

    9. Page 253
      Abstract

      In the presence of K+ and HPO42− ions, kaolinite, montmorillonite, micas (sericite) and gibbsite are transformed to taranakite. The reaction rate is at maximum in the pH-range 3.2–4.3. At p. 2, phase KAllHpo20 is formed, at p. 5 phase KAl2(OH)(PO4)22H2O. The latter is chemically identical with iron free leukophosphite, but differs in X-ray and structural data. In the absence of silica and at elevated temperatures (100°C) KAl(HPO4)2H2O transforms quantitatively to taranakite, even at p. 2.0. With smectites and micas the nucléation of taranakite occurs more rapidly than with kaolinite, which may be earned by epitactic nucléation due to some structural similarities of these minerals with taranakite. Taranakite {(K3Al5(HPO4)6(PO4)2(H2O)12}(H2O)6 has a trigonal rhombohedral layer lattice (space group R3 c- D63d, a = 8.7025(11)Å c = 95.050(13) Å). The individual 13.8 Å thick layers are separated by planar monolayers of 6H2O molecules per formula unit. The remaining 12H2O molecules belong to (AlO, OH2) -octahedra and are part of the rigid layers. The potassium ions are positioned in the central plane of the layers. They cannot undergo cation exchange reactions. The interlayer water is easily removed, thus forming jrancoanellite K3Al5(HPO4)6(PO4)2(H2O)12. The experimental conditions for dehydration are highly affected by crystal size, crystallinity, heating rate and isomorphous substitutions (62–80°C at ambient conditions or 20°C at pH2O ≤ 0.66 Pa). As long as traces of H2O remain in the interlayer region, rehydration to taranakite is possible. With regard to this aspect the reaction taranakite =francoanellite + nH2O resembles the reaction of hydrated halloysite = halloysite and H2O

    10. Page 260
      Abstract

      Kaolinite:alkali halide intenalates have been success fully prepared by grinding the salt with kaolinite in the absence of water as well as by reaction of kaolinite with concentrated aqueous solution of alkali halide. For the dry grinding method the rate of intercalation is greatest for salts with the lowest melting point. The basal dimensions of the intercalates increase with increasing size of ion. The layered intercalate structure survives the dehydroxylation of the kaolinite at 500–600° C, at which point the excess salt can be dissolved to give an XRD-amorphous material. For each alkali metal, M, the amorphous product has approximate stoichiometry MAlSiO4 and transforms to a crystalline phase of the same stoichiometry at surprisingly low temperature.

      Kaolinite-.alkaline earth halide intercalates and their derivatives have also been prepared, but only by the concentrated aqueous solution method. These intercalates display little long-range order, but can still be dehydroxylated to yield upon rinsingXRD-amorphous materials with approximate stoichiometry MAl2Si2 O8 (M = Ca, Sr and Ba). Upon heating these materials transform initially to adopt the hexacelsian structure then at higher temperature transform to feldspar- related structures.

      While it is possible to react other metal halides, such as halides of transition metals and magnesium, with kaolinite to achieve at least partial intercalation the subsequent reaction steps are generally not successful. This is usually due to the instability of these salts during the dehydroxylation step resulting in de-intercalation.

    11. Page 266
      Abstract

      Synthetic pyroaurites with Mg:Fe(III) ratios between 1.54 and 2.56 have been investigated by X-ray diffraction, ultraviolet-visible, infrared, and Mössbauer spectroscopy. All diffractions may be indexed using a hexagonal cell having a0= 0.311 nm and c0 values between 2.310 and 2.230 nm. The value of c0 is linearly correlated with the layer charge (Fe(III)/(Mg+Fe(III)) ratio). No superstructure reflections due to regular ordering of Mg and Fe atoms in the metal hydroxide layer were observed. Infrared spectroscopy did not indicate changes in the structure within the Mg:Fe(III) ratio investigated. Ultraviolet—visible spectroscopy and Mössbauer spectra show changes which are interpreted at mainly caused by the presence of a distribution of sites for Fe(III) in the metal hydroxide layer with slightly different surroundings.

    1. Page 275
      Abstract

      Interest in an unusual fibrous clay was generated by the fact that the air dry specimen resembled a kaolin mineral in its chemistry, thermal behaviour and IR spectrum, but its XRD pattern was not that of a kaolin, and its morphology was long thin laths. It formed a thixotropic gel at 1% concentration. When the clay was resampled and kept tuet it ivas identified as hydrated halloysite, 10 Å; the particles were tubular (−300 Ä diameter), and very long, up to 30 μm in length. The tubes were so uniform in diameter that interference between the opposing tube walls produced a ripple in their X-ray diffraction pattern. The tube walls were very thin (5–6 kaolin unit layers), and when the clay was dried the tubes collapsed to laths. The dried clay gave a series of diffraction lines with the strongest at −8.5 Å. When treated with glycerol this line moved to 8.2 Å and a strong line developed at 6.85 Å. Despite the persistence of the 8 Å line in terms of method of preparation and heating, it is not considered to be an interlayer spacing. The diffraction pattern is greatly modified by interference between the, collapsed, opposing tube walls, so the interlayer spacing of −7.2 A is not observed in the diffraction pattern. The cation exchange capacity of the field dried clay was −10 cmol(s) kg−1, while that of the wet clay was −20 cmol(+) kg−1. The reason for the difference is not understood. As expected from its morphology, the day had a high surface area which depended on method of preparation.

    2. Page 285
      Abstract

      The spatial distribution, chemical characterisation and identification of clay minerals in diagenetically altered sedimentary rocks can be determined with digital images produced in an electron microprobe. Samples are prepared by epoxy impregnation, polishing and carbon coating. The X-ray and backscattered (BSE) images may be obtained at variable magnifications from 35 to >2000x. The X-ray images are typically collected for as many as 16 elements employing a 256 by 256pixel matrix and variable magnifications. The practical upper limit due to the relatively large diameter of the X-ray excitation volume is a magnification that results in a pixel width of 3–6μm. Resolution of the BSE image is much higher and approaches the 0.5 van diameter of the electron beam. Both types of images may be processed by filtering and combinations of Boolean operators to produce maps of mineralogical variability. For example, illite is recognised by the coincidence of pixels with K, Al, and Si with or without Fe and Mg.

      In typical applications to diagenetic studies of clay minerals in sandstones, the technique is especially useful for illustrating the varying composition of clay minerals and their location within the sand body. The results provide much more specific information than XRD measurements obtained by disaggregating the samples. Illite/smectites in pore-lining and pore-filling locations often reveal different K, Al, Fe, and Si ratios. Chlorites formed as alteration rinds are often more Fe-rich than those filling pores. Poor count rate statistics for X-ray spectra obtained at dwell times of a fraction of a second limit the chemical distinctions to major element differences greater than 10 percent of the quantity present. BSE images reveal smaller scale, but difficult to quantify, compositional variability. The methodology is applicable to a wide variety of petrographic and stereological studies.

    3. Page 291
      Abstract

      Fluorine atoms and OH groups have similar radii and electronegativities, so that the second one may be more or less substituted by the first one in layer silicates. Such substitutions have recently been studied by 19 F and 1H broad band NMR High resolution 19 F MAS NMR spectroscopy is a complementary method used in the study of natural and synthetic dioctahedral 2:1 layer silicates.

      There is generally only one signal on each spectrum with numerous spinning side-bands. Partial or complete substitutions ofmagnesium or gallium atoms for aluminium atoms lead to changes of the chemical shift (with opposite signs). 19F chemical shifts and mean electronegativity of the octahedral elements satisfy the linear relationship:

      δ19F = αχp+ b,

      where δ is the chemical shift andthe mean electronegativity of the octahedral elements: a and b are constants. Thus, the chemical shift is principally controlled by electronegativity, which allows the prediction of new chemical shifts, according to the occupation of the octahedral sheet.

      19F MAS NMR spectroscopy is a new and powerful tool which offers the possibility ofexploring octahedral sheet composition offluorinated layer silicates.

    4. Page 295
      Abstract

      Electric dichroism measurements have been used to study the orientation of metal complexes on clays. The chelates investigated are [Ru(phen)3]2+ (phen = 1,10-phenanthroline) and[Ru(dmbpy)3]2+ (dmbpy = 5,5′-dimethyl-2,2′- bipyridine) when these complexes are adsorbed by a colloidally dispersed clay. Linear dichroism is induced in the metal- to-ligand charge transfer (MLCT) transition band by applying an electric fieldpulse to a suspension of a clay and a chelate. The orientation of a bound chelate is determined from the magnitude of the induced dichroism. As a result, [Ru(phen)3]2+ is deduced to be oriented with its three-fold symmetry (C3) axis in the vertical direction with respect to a day surface, while [Ru(dmbpy)3]21 has its C3 axis more than 50 degrees from the vertical direction.

    5. Page 300
      Abstract

      Volatile multicomponent mixtures including combinations of carboxylic acids, aldehydes, amines, sulfides and disulfides have been shown to be attractive to wild canids (dogs and foxes) and have been used to elicit behavioural responses useful for wild dog control. Adsorption of these organics onto a clay presents one method of preparing a suitable slow release formulation for odour generation. This paper reports the use of Diffuse Reflectance Fourier Transform Infrared Spectroscopy (DRIFT) to determine the concentrations of surface adsorbed organics present as mixtures on bentonitic clays.

    6. Page 304
      Abstract

      Adsorption of p-nitrophenol (pNP) from xylene has been used to measure the specific surface area (SSA) of a range of clay minerals, oxides, and subsoils. Non-expanding crystalline clays and oxides gave L-type isotherms with a distinct plateau above an equilibrium concentration (c) >40 mM of pNP. The S&4 values derived from the plateau adsorption, using a molecular area of0.424 nm2, are in excellent agreement with the corresponding BET-N2 areas. For allophanes and smectites, the isotherms showed an inflection at a certain value of c, beyond which adsorption increased linearly with c. The SSA of three allophane samples, derived from pNP adsorption at the beginning of the linear portion (point ‘X), was about half that of the corresponding BET-N2 value. A plausible explanation is that the interspherule surface of allophane aggregates is accessible to N2 but not to the much larger pNP molecules. Adsorption of pNP beyond point X’ may be ascribed to interspherule penetration and, in the case of smectites, to interlayer entry although the latter process may occur earlier. X-ray diffraction analysis showed that the basal spacing of smectite-pNP complexes increased with the amount adsorbed, indicating that intercalation did not occur randomly within a crystal. The extent of intercalation also depended on the nature of the exchangeable cation (Ca2+ ≫ Na+). For subsoils containing no smectite in the clay fraction, there was a reasonably good correlation between the pNP and nitrogen areas. Since pNP intercalates into smectite, the SSA of smectite-rich subsoils is generally larger than the corresponding BET-N2 value. On the other hand, pNP adsorption tends to underestimate the 5S/4 of subsoils rich in allophane and hydroxy-interlayered clays. This is because the interspherule pores of allophane aggregates, and the interstices of the interlayer hydroxide sheet in chloritic minerals, are less accessible to pNP than to nitrogen. However, for non-expanding clays and metal oxides as well as for subsoils dominated by such minerals, adsorption of pNP provides a simple means of estimating their SSA. An added advantage is that the method can be routinely applied in most soil laboratories as the analysis only requires access to a visible spectrophotometer.

    7. Page 311
      Abstract

      Identification of the clay minerals, kaoliniite, montmorillonite, illite, Vermiculite and chlorite was made possible using an idtrasonic technique. This technique easily provided important information about phyllosilicates.

      The basic principle of the method derives from differences in the activation energies of the different clay minerals and their resolution of the ultrasonic waves. It was found that the activatioit energy of smectites increased more than two times after spraying ivith ethylene glycol, indicating that the spacing between the layers has a pronounced effect on ultrasonic waves. Further elaboration of this method could establish a neiv and reliable toolfor phyllosilicate identification.

    8. Page 317
      Abstract

      An attempt has been made to improve the efficiency of iron oxide dissolution by the dithionite reduction route. Λ study of the effect of various parameters was carried out; these included time, temperature, complexants, amounts of dithionite, oxalate, sodium carbonate andpH. Based on these studies, a method has been proposed for the dissolution of iron oxides from soil clays. In this method, the reduction by sodium dithionite is carried out in a solution containing sodium oxalate and sodium carbonate at a pH of 8.05 and a temperature of 10CFC. These conditions produced the maximum dissolution of iron oxides. This DCO (dithionite–carbonate–oxalate) method was tested with standard clay minerals and some soil clays and also compared with dithionite-citrate-bicarbonate (DCB) method. Dissolution of iron was greater by DCO than by DCB treatment, in each of the clay samples tested. XRD patterns of DCO treated clay samples showed definite improvements and no artefact formation or any other deleterious effects. The method, therefore, appears to be efficient and well suitedfor the removal of iron oxides from soil clays.

    9. Page 323
    1. Page 333
      Abstract

      The oxygen- and hydrogen-isotope compositions (expressed as δ 18O and δ D values) of a mineral are related to its temperature of formation and the isotopic composition of the water from which it crystallized. For minerals formed near the Earth ‘s surface during weathering, the water is typically groundwater, or meteoric water (i.e. water in isotopic equilibrium with the atmosphere). As the isotopic composition of meteoric water varies principally as a function of the temperature of rainfall condensation at the time of recharge of groundwaters, both rainfall and unevaporated groundwaters have δ 18O and δ D values related to global mean temperatures. Thus there is a regular relationship between latitude/altitude and isotopic values.

      During weathering, this same variation in isotopic composition as a function of temperature is preserved in newly formed minerals that are precipitated in equilibrium with groundwater. This relationship holds true for a variety of weathering minerals but is most easily demonstrated for the clay minerals, where it has been shown that theEf^O values for kaolinite are retained for up to 240 Ma. Thus gross distinction between cold-climate and warmer weathering is readily made.

      Tectonic movement of weathered cratons either by translation or uplift can also be detected by comparison of the preserved δ 18O values in kaolinite with those calculatedfor the present situation from the isotopic composition of modern rainfall. Conversely, if the rates of motion of such plates are known, the interpolated ages of weathering can be deduced. A broad conclusion of this work is that cold-climate weathering is a common event, but not so well preserved at current high latitudes owing to Quaternary glaciation. The presence of deep weathering profiles in the geological record should not be necessarily taken to indicate past ‘tropical’ conditions or latitude.

      The above methodology is ideally suited to the study of clays from both deep-weathering and reworked-sedimentary environments in the Gondwanaland continentalfragments. For example, oxygen-isotope results from clays in Australia and India demonstrate the expected substantial climate shifts related to continental translations since the Permian. Isotopic results from South America, where continental drift was less during the Cretaceous and Tertiary periods, show only minor changes with time. The existence of the Australian Eastern Highlands since the Cretaceous is also strongly supported by isotopic results on surficial clay minerals.

    2. Page 339
      Abstract

      The clay mineral composition of suspended and deposited dust is compared with that of soils and sediments of arid eastern Australia. Kaolinite–Illite (KJI) ratios are used to describe differences in clay mineral composition. K1I ratios in soils and sediments decrease from north to south. Suspended dust contains no or low amounts of smectite, deposited dust shows appreciable amounts of this clay mineral (between 10 and 40%). Smectite content does not show any variation with latitude. Background dust usually reflects the clay mineralogy of the soils surrounding the sampling station, whereas the clay mineral composition of dust from major dust storm events is determined by spatial and temporal variations in dust sources at the time of the dust storm. K1I ratios and smectite content decrease with distance from source, showing a size-dependent relationship: kaolinite and smectite often form larger aggregates, which travel smaller distances than single clay particles. Entrainment of fluvial sediment also contributes to dust composition variability.

    3. Page 346
      Abstract

      Steeply dipping ferricrete forms prominent cuestas up to one kilometre in length in the Telford Basin of South Australia. Distinctive ferruginous voidal concretions occur within the ferricrete and appear to have formed by weathering of siderite andpyrite-rich layers (hardbars) within interburden and overburden sediments ofTriassic coal measures. Minera- logically, the concretions consist dominantly of goethite with varying amounts oflepidocrocite, hematite, quartz, kao Unite, smectite, micas, rutile, anatase andpossibly greigite. Crystals of gypsum occur in some central voids. Concentric layers of the concretions are enhanced by distinctive colour variations that correspond to changes in iron oxy- hydroxide mineralogy.

      Comparison of unweathered siderite and voidal concretions, using light microscopy (thin sections), reveals a similar framework of silt-sized grains, suggesting concretion formation by isovoluminous weathering of siderite. Siderite is not as dense as goethite, lepidocrocite or hematite thus, during weathering transformations of siderite and the accompanying loss of CO2, less space would be occupied. Consequently, if the original size of the sideritic mass was fixed by the surface transformation of siderite to hematite, further weathering would result in the development of central voids, containing variably iron-depleted clays. The alternating formation oflepidocrocite, goethite and hematite is considered to reflect varying rates of oxidation of ferrous iron depending upon differing rates of oxygen supplies. The intimate association of goethite and lepidocrocite suggests that the minerals have formedfrom the same source of mobilised Fe2+ in the same general environment. However, some fluctuations in the environment must have occurred to favour the dominant formation of one oxide over the other. Gypsum crystals occur in some central voids and were probably derived from siderite (calcium) and pyrite (sulphur). Concretion formation thus involves in situ weathering of siderite, progressively inwardly transforming siderite to hematite, goethite and lepidocrocite, developing central voids variably filled with detrital iron-depleted clays and gypsum.

    4. Page 355
      Abstract

      Structural Fe oxidation occurs readily in micas that have been expanded by a replacement of interlayer K with hydrated cations. This paper examines whether Fe oxidation can occur when the mica layers remain contracted. The 10–20 μm size-fractions of a phlogopite, a magnesian biotite and a siderophyllite, having initial Fe(II) contents of 0.3, 1.9, and 3.2 atoms per unit cell (44 +/− charges) respectively, were studied. The two weathering solutions consisted of a 1 M KOAc (pH 4.7)-0.1 M K2EDTA mixture (‘H2O2-free) and this solution made with 25% H2O2 (‘H2O2 ‘). Treatments were applied in perfluoroalkoxy (PFA) polymer flow-through cells immersed in a water bath maintained at 8(fCfor periods up to 36 days. No oxidation was observed in any of the micas treated with the H2O2-free solution. With the H2O2 solution, however, the biotite showed a 7-fold increase in Fe(III) content during the first week oftreatment after which little further oxidation occurred. The phlogopite showed no oxidation, and only limited oxidation was observed in the siderophyllite. In the H2O2-free solution, and in the H2O2 solution after 9 days of treatment, homogeneous dissolution ofFe, Al, and Mg was observedfor all the micas. During the period of rapid oxidation in the H2O2 solution, however, preferential dissolution of the trivalent ions (Fe and Al) relative to the divalent Mg ion was observed for the biotite sample. Enhanced dissolution rates for the biotite (up 600%) and siderophyllite (up 60%) in the H2O2 -free solution after treatment with the H2O2 solution suggested that extensive exfoliation of the biotite and minor exfoliation of the siderophyllite had occurred. No change in dissolution rate was observed for the phlogopite after these treatments. The unexpected resistance of the siderophyllite to oxidation and dissolution was probably due to its high selectivity for K. This property inhibited expansion of the interlayer and thus prevented ejection of cations to balance the increase in layer charge that would arise from structural Fe oxidation. Because the extent ofexfoliation closely paralleled the trends in structural Fe oxidation seen for the three micas we conclude that extensive oxidation does not occur without an expansion or exfoliation of the mica layers.

    5. Page 362
      Abstract

      Little is known about the geodynamic context of the type halloysite first analysed by Berthier in 1826, but it is clear that this halloysite occurrence is linked to a karstic environment and in relation with sulphides capable of releasing acid weathering products. Several new occurrences of halloysite were recently discovered in Belgium (Ardeme Massif at the carbonate boundary at the bottom of infilled giant karstic sink-holes (100 m to 1 km wide, 100–200 m deep). The occurrences consist of more or less pure halloysite (up to 99%) bodies of different shapes, sometimes of large size. These new occurrences are considered to be similar to those of the type locality at Angleur.

      The development of the cryptokarstic pockets results from the superposition of two main sets of beds on the same peneplained Dinantian limestones: (1) a permeable cover of marine sands deposited during an Upper Eocene/Oli- gocene transgression; (2) the development of a woody continental environment leading to fluviatile and lacustrine organic bearing deposits, during a regression in the Middle Miocene Age. With the regression and the fall of the sea level, the dissolution of the carbonate substratum was initiated below the permeable sandy cover leading to local karstic subsidence. Due to the karstic subsidence, lakes and ponds formed on the paleosurface and organic bearing Neogene sediments were progressively trapped. At the same time, the sands were leached and bleached by acid solutions coming from the organic matter and/or from the weathering ofthe pyritic brown coal. At the bottom ofkarstic pockets, the solutions were neutralised at the limestone contact and (silica) alumina gels precipitated. Halloysite crystallised from these gels at the sand/limestone interface or in the limestone margins of the interface ifit was porous.

    6. Page 367
      Abstract

      Well documented sites of saprolites and acid soils were chosen from the main crystalline terranes in France (Alps, Morvan, Central Massif and Brittany). Genesis and organisation of clay minerals were studied, using qualitative and semiquantitative mineralogical methods, high-resolution transmission electron microscopy and analytical electron microscopy. Saprolites and acid soils present two different weathering systems, which are dependant on the age of the superficialformations. Saprolites are ancient (end of Tertiary) and may contain a large diversity of clay minerals. The clays were formed, at least in the beginning, in a globally draining (sandy) environment, which involves intensive leaching and hydrolysis ofprimary minerals. Kaolinite (or halloysite) forms from mica andfeldspar, and Al-interlayered Vermiculite forms from mica. The formation of smectite in saprolites requires more specific conditions: relatively basic (in weatheredplagioclase) and/or hydromorphic conditions. Soils are recent and typically developed under acidic conditions with high organic matter concentration. Primary phyllosilicates (mica and chlorite) are the principal source of clay minerab. Soil clay formation processes involve mainly inheritance, microdivision and transformation of the micas. Such clays have low exchange capacities with Al saturation.

    7. Page 373
      Abstract

      During an investigation of basement weathering around the margins of the Tin Serinine basin, SEM observations revealed the presence of abundant newly formed quartz grains associated with the altered Kfeldspars of Pan-African granites (porphyritic biotite-bearing pink variety). The quartz appears to grow onto feldspar which is in the process of being dissolved, but no evidence can be found for coatings or alteration products. This implies that the potassic feldspar undergoes congruent dissolution and the newly formed quartz crystallises at the expense of silica in solution by epitaxial growth onto pre-existing seeds of microcline. The stability limits associated with precipitation of quartz and kaolinite indicate that, at low temperature (25° C), congruent dissolution of microcline contributes to the precipitation of quartz alone over a wide range of pH.

    8. Page 378
      Abstract

      This paper presents details of a sepiolite deposit of meerschaum type representative of the well-known deposits found in the vicinity of Eskisehir City, Central Anatolia. In this deposit, sepiolite is typically found as pebbles and lumps accumulated in the base of a conglomerate sequence, together with magnesite fragments and pebbles derivedfrom nearby ultrabasic rocks which contain magnesite. Palygorskite occurs in intergranular space throughout the conglomerate sequence. Sepiolite and palygorskite have been formed by silica-rich ground waters percolating through the conglomerate during early diagenesis. Sepiolite was formed by the in-situ replacement of pre-existing magnesite. Palygorskite was precipitated directly from solution in which the Al concentrations had reached the level nquiredfir the formation of this mineral.

    9. Page 383
      Abstract

      Sepiolites are highly developed in the Neogenetic Basin of Madrid and are mined at Vicdlvaro (Madrid). Seventy-two typical samples have been studied from different quarries where the sepiolite is associated with Mg-smectites, lutitic arkoses, calcretes and silcretes. The microfabric, texture, mineralogy and geochemical data allows us to recognise two distinct sepiolite-bearing levels: a lower level with a mean sepiolite content up to 90% (sepiolites) with laminar massive and intraclastic textures and an upper level with average contents of 50–60% (sepiolitic clays) accompanied by smectite and calcite with granular and nodular textures. The sepiolite has formed by diagenesis of sediments in a lacustrine environment rich in Mg-smectites and where Li is present in high concentrations. The sepiolites form from deep sediments at the centre and edges of lakes while sepiolitic clays form from the sediments of shallow lakes and swampy areas. The sepiolites are formed during late diagenesis by regional groundwaterfluxes related to the arkose aquifer that cames F which is significantly incorporated into sepiolite. A progressive replacement of the Mg-smectite with dolomite, F-sepiolite and finally silica phases occurs during this stage. The sepiolitic clays are formed during the early diagenesis stage in paleosols (vertisol) associated with smectites and calcite. These processes are favoured by the existence of a low rate of sedimentation and a semiarid climate. The sepiolite is associated with increasing intensity of weathering which changes the secondary porosity characteristics. The sepiolite evolves from cavity-lining to channel andfinally fissurai.

    10. Page 393
      Abstract

      Three main beds ofmono-mineralic Santonian argillites are found in the pelitic Ubatuba Formation in the southern portion of the Campos Basin, offshore Brazil. These beds are usually 1 or 2 metres thick and can be distinguishedfrom the surrounding marine shales and siltstones by their low resistivity and high porosity. These physical properties reflect the homogeneous mineralogical composition of the argillites, composed almost exclusively of highly expandable dioctahedral illite-smectite mixed-layer clay minerals.

      The mode of occurrence of the ‘3-Dedos’ argillites (beds with abrupt top and base contacts, lack offissility, local occurrence of spheroidal exfoliation, short thickness and good lateral continuity, repetitive and episodical character), the specific mineralogical composition and the vestiges of primary petrographic texture indicate they originate from volcanic ashes. Thus, these argillites can be classified as bentonites.

    11. Page 399
      Abstract

      Lacustrine sedimentary rocks of the Huka Falls Formation in the Wairakei geothermal system occur at depths between 180 and370 m where the measured drillhole temperatures range from 9(F to 220° C. Cores from 4 wells selected in the Te Mihi sector of the system were analysed. The sediments contain clay minerals of variable composition and structure. Smectites occur below 90° C. The illite/smectite mixed-layer clays develop with interlayer occupancy of the illite component increasing regularly with temperature from 90° to 230° C, and end member illite occurs where temperatures exceed 180° C. An empirical relationship between the illite-smectite mixed-layer clays and their temperature was determined in the temperature range from 90° to 230° C. The crystallinity of the (001) reflections for illite/smectite mixed-layer clays increases with increasing host rock temperature and depth, with most 20 (Kubler Index) values ranging from 1.40 to 0.44. The Kubler Indices of the illites are from 0.35 to 0.44. An empirical relationship between the Kubler Index of the illite/smectite (001) peak (air-dried) and the measured drillhole temperatures (T) from 90 to 230° C is, T(°C) = 249 – 89.3C, where C is the Kubler Index (r = 0.92). The Kcontentper component of O10(OH)2 of illite/smectite increases with increasing drillhole temperature and correlates negatively with their Kubler Indices. Chlorites are all trioctahedral, iron rich and contain about 5–10% smectite layers in their structures. Microprobe results indicate increasing AllV andFe contents but decreasing Si content with increasing depth and temperature in the sediments. Chlorite crystallinity (ChC), as represented by the (002) reflections, varies only slightly with the drillhole temperature and shows consistently high values throughout (averageΔ°2θ = 0.28).

    12. Page 405
      Abstract

      The characteristics of the diagenetic evolution of smectite in source rocks ofthe Songliao Basin and their relation to oil and gas generation, migration and enrichment are described. The diagenetic evolution of smectite in the Basin is principally controlled by burial depth and temperature. The depth and temperature at which smectite converts to illite-smectite mixed-layer clay are 1250–1450 m and 62–70° C respectively. The depth and temperature for conversion of illite-smectite mixed-layer clay to illite are 2500–3000 m and 110–120° C respectively. The conversion of smectite to illite-smectite mixed-layer clay occurs at two sharp boundaries separated by a depth of approximately 450 m. Depositional environment in the Basin has a direct influence on smectite diagenesis: in deep-lacustrine sags rich in organic matter the conditions are favourable for the conversion of smectite to chlorite; in fluvial and deltaic sags smectite converts preferentially to illite. Smectite diagenesis has a strong correlation to organic maturity. The most favourable conditions for oil and gas migration are at the boundaries where there is rapid conversion of smectite to illite-smectite mixed layer clay. The boundaries are therefore strongly correlated with the location ofoil and gas bearing rocks.

    13. Page 411
      Abstract

      The purpose of this compilation paper is to establish the clay mineral distribution in the Tertiary basin of Campo Arañuelo and to compare it with the Madrid basin. The Campo Arañuelo and Madrid basins are located in central Spain and form a major portion of the Tajo Basin, which is one of the three large continental Tertiary basins in the interior ofthe Iberian Peninsula. The Campo Arañuelo basin is a small graben (2000 knf) limited by a granitic range in the north and by a granitic-metamorphic one in the south. During the Tertiary a series of alluvial fans sourcedfrom granitic areas developed to thickness of 600 m. The detrital sedimentary record is quite homogeneous and uniform with a general predominance of montmorillonite, kaolinite and illite in the clay fraction. The montmorillonite content of the sediments increases with distance from the granitic source area while kaolinite and illite contents are highest near to the source area. In the Madrid basin (10 000 km2) moderate to high subsidence rates during the Tertiary accompanied the accumulation ofmore than 3500 m of continental sedimentary units. This basin shows the development from alluvial fans to lacustrine deposits.

    14. Page 417
      Abstract

      The Aveiro basin constitutes the northernmost section of the Portuguese occidental Meso-Cenozoic sedimentary basin. Sedimentological, mineralogical and geochemical studies were carried out on cores taken from three boreholes located in the northern sector of the Aveiro sedimentary basin. Clay minerals and the accompanying non-clay minerals were utilised for lithostratigraphical discrimination and for paleosurface definition. At the base of the lower Cretaceous sandstones a more pelitic layer, reddish and a few meters thick, occurs. This layer is very rich in kaolinite. Towards the top of the layergoethite (associated with hematite) becomes an important component. A much coarser and thick layer underlies the pelitic layer and is composed almost entirely ofquartz both in the sand and in siltfractions and by kaolinite and opal CT in the clay fraction. The available mineralogical and geochemical data allow us to interpret both zones as exposed horizons; one related to ferralitic weathering in wet and warm conditions, the other related to silicification caused by water-table fluctuation.

    15. Page 421
      Abstract

      The clay mineral content of the <2 μm fraction of Silurian to Cretaceous pelitic rocks of the Paraná Basin has been studied by X-ray diffraction. Illites, followed by chlorites, are the most abundant clay minerals in these rocks. Illite- smectite interstratifieds, smectites and kaolinites are subordinated but locally important. The strong detrital features exhibited by these minerals and their general spatial distribution in time indicate that their present occurrence was controlled mainly by climate and source rock composition. Detrital illite and chlorite characterise the pelites and diamictites of the glacially influenced Itararé Group (Permo-Carboniferous). A high content and a great diversity of types of illite-smectite interstratifieds as well as kaolinite marker beds (tonsteins) are found in the Rio Bonito Formation (Permian) that was deposited under humid climatic conditions. Smectites occur mainly under the arid and semi-arid conditions prevailing during deposition of the Upper Permian and Mesozoic sediments. Pure smectite beds (bentonites), which are found intercalated with the Permian rocks, represent volcanic contributions to the basin. Corrensite, associated with oolitic limestone marker beds, was formed by early diagenesis under high-magnesium marine conditions specially in the Teresina Formation (Permian).

    16. Page 429
      Abstract

      Research on the clay mineral composition, distribution and provenance of selected physils from Carboniferous and Permian sequences from the sedimentary basins of Paganzo, Calingasta-Uspallata and the area ofthe Islas Malvinas in Argentina; from the Tarija Basin in Bolivia andfrom the Uruguayan Basin has been carried out. The main Carboniferous clay mineral associations comprise

      Paganzo Basin: (1) kaolinite-illite-chlorite; (2) (l4c-14sm)-illite-chlorite; (3) illite-chlorite, with (14c-l4v) or (I4c-l4sm);

      Calingasta-Uspallata Basin: chlorite-illite;

      Uruguayan Basin: kaolinite-illitef 10-l4sm);

      Tarija basin: (l4c-l4sm)-illite-chlorite.

      The Permian clay mineral associations of the Paganzo, Tarija and Uruguayan basins are characterised by the systematic presence of smectite, which is accompanied by chlorite-smectite in the Tarija Basin, illite-(10-l4sm) in the Uruguayan Basin and illite-(l4c-l4v) in the Islas Malvinas area. Sedimentation occurred in a wide range of depositional environments which are often structurally controlled. Most of the clay minerals are detrital but illite crystallinity also indicates diagenetic or metamorphic mineral forming processes. Smectite is considered as derived from accumulated tephra fall deposits in subaqueous environments in Permian deposits of Uruguay and the Tarija basin, proving volcanic activity. Zeolites and non-marine ostracoda suggest saline-alkaline lake conditions in the Upper Permian of Uruguay. Clay mineral associations in the Uruguayan Basin suggest no connection with the Paraná Basin (Brazil), at least until the Upper Permian.

    17. Page 436
      Abstract

      Stratigraphical studies carried out on the Tunas Formation (Pillahuincó Group), in the Abra del Despeñadero area show whitish to orange smectitic horizons. X-ray diffraction as well as petrological and SEM analyses on samples from a rock section made it possible to define these smectitic levels, mainly montmorillonitic in composition, as the product of argillogenesis off vitreous materials, pyroclastic in origin, that accumulated as tephra fall deposits in aqueous environments, in association with minor amounts ofillite and Vermiculite. The grade of alteration of these pyroclastic beds increases upwards. They are recognisable as intercalations in the normal siliciclastic sedimentary rocks which are characterised essentially by quartz andplagioclase. Associated are other clayey, non pyroclastic, non smectitic deposits, inter- bedded with sandy fining-upwards cycles and channelised bodies, suggesting a deltaicpaleoenvironment with distributary mouth bars and avulsion processes in probable interdistributary bays. Similar events mentioned by other authors for equivalent units in South Africa, indicate that important tephra falls, during the Permian, could have originated from the activity of a magmatic arc located to the south-west of the sample area and related to the Patagonian plate approaching close to the southern margin of the Gondwana Continent.

    1. Page 443
      Abstract

      Recent studies of proto-imogolite, imogolite, allophanes with Si:Al molar ratios 1:2 and 1:1, proto-halloysite and hydrous feldspathoid and poorly crystalline aluminosilicates on weathered primary minerals were reviewed to contribute to understanding of poorly crystalline aluminosilicate clay minerals in soils and weathering products. Up-to-date information on nomenclature, natural occurrence, crystalline state, chemical structure and conditions for formation are provided.

    2. Page 449
      Abstract

      Optical microscopy on thin sections and transmission electron microscopy on microparts of clay fabrics in thin sections, isolated by microdrilling, were used to study the occurrence of 2:1 phyllosilicates resulting from hypogene pyroxene alteration in andesitic pyroclasts in the humid tropics. After supergene rock weathering, 2:1 clay minerals were liberated mainly in the coarse grain size fractions of soils. Routine X-ray diffraction (XRD) techniques performed on separated grain size fractions of a Costa Rican Andisol clearly demonstrated a microdivision of sand-sized clay pseudomorphs in the C horizon into clay-sized particles at the soil surface. In contrast, in Oxisols of Guadeloupe (French Antilles), iron compounds liberated by ferrallitic weathering conditions, immediately impregnated the 2:1 clay fabrics, thus preserving them and rendering them undetectable by routine XRD analysis. Since optical and (sub)microscopical techniques suggested the presence of 2:1 phyllosilicates in the sand fraction of the C horizon of the Oxisol, samples were submitted to intensive ultrasonic dispersion, chemical treatments and adapted analysis conditions. Finally we were able to identify 2:1 (interstratified) clay minerals. Near the soil surface, physical breakdown of iron-impregnated clay fabrics leads to the appearance of small fragments in the fine silt and clay fractions. Consequently, inheritance of hypogene-altered pyroxene may be the origin of the frequently reported occurrence of minor amounts of 2:1 phyllosilicates in highly weathered volcanic soils. The technique used appeared to be a complementary tool in routine X-ray diffraction analyses and well adapted for unravelling the impact of in situ mineral weathering and clay neoformation on micrometre and nanometre scales.

    3. Page 457
      Abstract

      Reported occurrences of 1:1/2:1 mixed-layer clays in soils lead to the identification of two major groups: kaolinite-smectite (K-Sm) and halloysite (10 Å)-smectite (H10-Sm). These minerals occur in distinct soil environments. Soil H10-Sm forms in silica/base-rich volcanic ash soils at the expense of pyroclastprimary minerals through syngenesis (co-neoformation’) which is strongly micro-environment dependent. Soil K-Sm forms in silica/base-depleted environments at the expense of smectite: the interstratified mineral is a pedogenic transient phase in the conversion of smectite into kaolinite. Both syngenetic and desilication pathways are illustrated by two case studies.

    4. Page 462
      Abstract

      Accurate solubility constants (log K°) values are essentialfor determining the relative stabilities of minerals in soils and sediments. The minerals considered here are kaolinite, halloysite, imogolite, and the allophane with the Al/Si ratio of 2. Gibbsite, quartz and amorphous silica are used as reference minerals.

      Most log K° values (at 25° CJ were initially calculatedfrom free energies of formation (A Gp of mineral and aqueous species published in the most recent compilations of the US Geological Survey (in 1979), the US National Bureau of Standards (in 1982) and also from the major soil mineral compilation of Lindsay (1979). These compilations do not include imogolite and allophane, which are discussed separately. The derived log K° values for each of the minerals kaolinite, halloysite, gibbsite, quartz and amorphous silica showed considerable variation between compilations, which substantially affected corresponding stability fields in a log(H4SiO40) vs log (Al3+) + 3pH solution activity diagram.

      A critical evaluation was therefore made of available solubility measurements for these minerals to derive more reliable log K° values. The following log K° values were selected: quartz (-4.00); amorphous silica (-2.72); gibbsite, untreated (8.04); gibbsite, acid-treated (7.76); kaolinite, well-crystalline (6.53); kaolinite, poorly-crystalline (7.47). The kaolinite values were derived after adjustment for the contribution to total soluble silicon of recently described aluminium-silicate complexes. A logK0 value of 12.08 for imogolite was estimated from an imogolitegibbsite equilibrium measurement. Because appropriate solubility data is not currently available for halloysite, a logK0 value of 8.95 was calculated from AGy° values of species in the dissolution reaction. A logK0 value of 24.2 was calculated for allophane (Al/Si = 2) assuming that it has the same stability as imogolite.

      A solution activity diagram was constructed using the selected log K° values. Within the normal range of soluble silicon activities in soil solutions, well-crystalline kaolinite is the most stable mineral of those considered here, except at very low silicon activities where gibbsite is the most stable. The metastable minerals imogolite (or allophane) and halloysite have intersecting stability fields with imogolite (or allophane) more stable than halloysite until quite high soil solution silicon activities are reached. The precise extent of these fields remains uncertain ivhile there is debate on the gibbsite log K0 value which affects the derivation of log K° values for imogolite, allophane, and halloysite.

    5. Page 469
      Abstract

      The dispersive potential of a sodic soil is determined by the evaluation of repulsive forces responsible for the spontaneous dispersion of clay particles from aggregates. Dispersive potential (Pdis) measures the balanced effects of adsorbed sodium and electrolyte concentration in a soil and hence, can be used as a single index of soil sodicity. In the present study, Pms toas highly related to the % spontaneously dispersed clay only when the soils (n = 57) were grouped according to the dominant clay mineralogy. The hydration energy of a cation on a clay surface is related to the layer charge. Therefore, when the cation exchange capacity of the soils measured by the conventional ammonium chloride method (CECam) was used in a multiple regression including the data for all soils on and the % of dispersed clay, a significant relationship was obtained (R2 = 57.9%). However, when CEC estimated by percolating methylene blue solutions through columns (CECmb) of soil aggregates was used in the multiple regression the correlation was improved and highly significant (R2 = 74.9%). CEC by the methylene blue method appears to be related to the sites available for coulombic interactions with water molecules which are important in clay dispersion. CECmb and Pdis relationships could be used to develop a universal criterion for soil sodicity.

    6. Page 473
      Abstract

      A range of soil clay fractions of different mineralogical composition were separated from soils using pretreatments limited to Na+ saturation and dispersion. Samples were deferrated by dithionite-citrate treatment and the extracts analysed to identify the key mineral components dissolved. These components ranged from pure iron (oxy)hydroxides, hereafter referred to as iron oxides, to silicate clays. Specific surface areas of iron oxides dissolved ranged from around 100 m2 g−1 for samples with XRD-identifiable iron oxide minerals, to 200–800 m2 g−1 for samples containing XRD- identifiable amorphous iron oxides. These surface areas are in accordance with published data. Computed surface areas of material lost on deferration of some iron-substituted smectite samples absent of free iron oxides, were about 800 m2 g−1, which is consistent with the dissolution of portions of 2:1 silicate layers. One sample of high BET-N2 surface area (224 m2 g−1) consisting of randomly interstratified illite/smectite material, and characterized as having a very high affinity for adsorbed zinc, lost one third of its BET area on deferration without appreciable weight loss (-3%) or dissolution office iron oxides. The dissolved silicate material had a computed area of 2500 m2 g−1 which is interpreted as being due to dissolution of a large number of minor projections with silicate composition at particle edges. On the basis of the data obtained it is proposed that these projections are localized extensions of smectite octahedral layers. The physical nature and the chemical implications of these proposed clay edge projections are discussed.

    7. Page 481
      Abstract

      In this classification oflateritic covers four major types are distinguished: ferricretes, latosols, conakrytes and bauxites.

      In ferricretes, hematite is associated with kaolinite, forming mottles, nodules and metanodules. When, at the top of profiles, goethite and sometimes gibbsite develop at the expense of hematite and kaolinite, protopisolitic and pisolitic dismantling facies are formed. Ferricretes, in which hematite and kaolinite form concretions, are widespread and are the most common iron accumulations.

      Latosols are soft lateritic covers with a microglaehular structure. Red latosols, like ferricretes, are essentially formed by an association of hematite and kaolinite, but with larger proportions of goethite and with the presence of gibbsite.

      Lateritic bauxites are concentrations of aluminium with which iron is very often associated. Four major types oflateritic bauxites: protobauxites, orthobauxites, metabauxites and cryptobauxites are defined as a function of the nature of iron and aluminium minerals as well as their relative distributions in profiles.

      Protobauxites are lateritic soils where gibbsite and goethite form together under very humid climates. Orthobauxites are allites or alferrites, rich in gibbsite and red in colour, which do not exhibit a concretionary structure. Iron may be concentrated in hard caps called conakrytes and located close to the top of the bauxitic profiles. Conakrytes are reticular and non nodularferrites or ferrallites in which hematite and goethite dominate and where gibbsite could be present in small proportions. The presence of kaolinite at the bottom of the profiles is not necessary. Metabauxites are boehmitic and show a concretionary or pisolitic structure; iron is dissociatedfrom aluminium and is frequently concentrated as hematite in a kaolinitic ferricrete located at the bottom of the bauxitic profile. Kaolinite always appears at the bottom of metabauxite profiles and less frequently at the base of orthobauxites. In cryptobauxites, kaolinite is abundant at the top and at the bottom of the profiles so that the gibbsitic layer is embedded between two kaolinitic horizons.

      This petrological and geochemical classification of latérites is based on reactions of hydration-dehydration and ofsilication-desilication regulated by temperature, water activity and chemical composition of the parent material. Lateritic bauxites, ferricretes and latosols are witnesses of the succession of paleoclimates throughout the last 150 million years, since the Atlantic opening.

    8. Page 487
      Abstract

      A study of forms of free silica has been made on ‘tepetate’ horizons of the Mexico Valley. These indurated horizons have a fragipan consistency and they are observed in soil profiles derivedfrom several pyroclastic deposits. We have made successive observations using optical microscopy, and SEM coupled with EDS in situ micro-analysis on thin sections, or on smear slides in order to better distinguish the fine particles of the matrix. The focus of this study is the matrix, microlaminated clay infillings and bright yellow cutans in voids or fissures. HRTM observations coupled with element microanalysis have been made on ultra thin sections of infillings and cutans. The clay mineral analysis has been made by XRD on the <2μm fraction and fine clay; anda chemical analysis of<2μm. A selective dissolution of Si and Al elements by NaOH has been done on the fine earth. We then made observations using TEM. and micro-analysis of isolated particles. Forms of free silica have been observed at every scale ranging from milli to nanometer. There occur phytoliths, nanometric opal and perhaps micrometric cristobalite. Phytoliths are present in the matrix and infillings. Opal-A has been seen only by HRTM in fillings and cutans, but its location remains to be precisely determined as does that of cristobalite.

    9. Page 494
      Abstract

      In order to estimate the weathering rate, different methods (geochemical, mineralogical, element flux measurements) have been used. In addition, an experiment was carried out in order to understand weathering under non-equilibrium conditions. The research area lies in the cool and perhumid summit region of the Black Forest. The Bärhalde granite has undergone 10 000 years of soil formation since the last glaciation, which formed the parent material by physical weathering of the solid rock. The average weathering rate is about 2 kmolc ba−1 y1 equivalent to a 400–500 kg loss of feldspar. The plagioclase, until now, has weathered more rapidly than orthoclase, but a tendency towards a higher participation of K-bearing minerals is now visible. So far, all three methods furnished results for overall weathering rates in the same order of magnitude, but each bit gave different information. Actual weathering rates, up to now, can only be estimated satisfactorily by element flux measurements on an ecosystem level. For the other methods, study periods of longer than 20 years are needed for a proper assessment.

    10. Page 499
      Abstract

      The clay minerals formed during the weathering of basalt in an upland environment in New Zealand were investigated. Earlier studies had indicated that the weathering products ranged from amorphous constituents through to gibbsite and kaolin indicative of a strongly weathering environment. This soil therefore appeared anomalous in weathering sequence studies.

      Soils were sampled on rounded ridge crests where downslope drift and atypical weathering effects were not present, and their mineralogy studied. The mineralogy of the sand fractions indicate that the soils were formed largely from basalt, although some loess was also present. The clay fractions were dominated by smectite, with smaller amounts of kaolin, largely present as a kaolin/smectite intergrade. The weathering of basalt in this environment therefore follows the expected course.

      Allophane and gibbsite are common constituents of podzolised soils formed under Nothofagus forest in upland regions, and it is probable that the original samples were from sites on the edge of the basalt outcrop and contain material influenced by podzolising processes under forest remnants which have since disappeared.

    11. Page 504
      Abstract

      A red soil has developed in the north eastern Llano Uplift, central Texas, on Precambrian graphite- and biotite-bearing Packsaddle Schist. Regional mapping shows variations in development of the red soil relative to distance from a retreating escarpment of Cretaceous sedimentary cover. The westernmost exposure of red soil is well developed, with a thick argillic Bt horizon. To the east, nearer the escarpment, the red soil is thinner and more orange, and is present only on topographic highs. XRD analysis shows the clays in the C horizon to be macroscopic and clay-size vermiculite, tubular halloysite, and illite. Very little vermiculite remains in the Bt horizon, where the clay is dominated by kaolinite. Transmission electron microscopy shows a general transformation of tubular halloysite to platy kaolinite from the base of the C horizon to the top of the Bt horizon.

      Rb–Sr isotopic analyses were performed on an ultra-fine clay fraction (<0.05 μm) from samples from the Bt horizon approximately 6, 4, and 3 km away from the eastern escarpment. Before analysis, ion exchange reagents were used to remove loosely bound Rb and Sr (that would have behaved as open chemical systems over geologic time) while leaving the more tightly bound parent and daughter intact. After treatment by diverse ion-exchange reagents, aliquots of clay contained correlated values of 87Sr/86Sr and 87Rb/86Sr, data that are the basis of a Rb–Sr isochron. From the west to the east, the three localities provide apparent Rb–Sr ages of 52.5, 31, and 27 Ma, which are interpreted to reflect the erosional retreat of the escarpment and initial exposure of basement metamorphic rock to weathering. Three lines of evidence indicate that the Clear Creek soil is indeed a relict soil: (a) the differential degree of development of the soil depends on the distance from the eastern escarpment, (b) the soil is currently being eroded, and (c) the pattern of Rb–Sr ages.

    12. Page 510
      Abstract

      Clay mineralogy of four soil profiles at different altitudes on Meihua Mountain was studied. Kaolinite, mixed-layer vermiculitic minerals, gibbsite, mica and different forms of iron oxide were found as the main clay minerals in all the soils, but in differing in quantities. Kaolinite decreased, whereas mixed-layer vermiculitic minerals and gibbsite increased with elevation. Because of an increase in gibbsite, the ratio SiO2/Al2O3, or SiO2/R2O3 cannot be used as indices of weathering intensity of mountainous soils.

    13. Page 515
      Abstract

      Mica particles in a root rhizosphere weather faster than similar particles in a soil matrix. Conversely, potassium (K) fertilisation retards or reverses the weathering process. This paper reports on the combined effect of long-term rice- wheat cropping and fertiliser treatment on mica weathering and K availability in a Typic Camborthid soil (Hafizabad series). A 20 g sample of soil from A p, AB, and Bw horizons was treated to remove carbonate (with 1M NaOAc buffered to p. 5.0), organic matter (with 30% H2O2), and Fe-oxide (with citrate-bicarbonate-dithionite (CBD) buffered at p. 7.3) and fractionated into sand (2 - 0.05 μm), silt (50-2μm), coarse clay (2-0.2 μm) and fine clay (<0.2 μm). Mica in the silt, coarse clay and fine clay was determined by digestion in HF. Vermiculite and smectite in the coarse clay and fine clay were determined by the CalMg and K/NH4 CEC method. As a pool of plant K, K in soil and sand and silt fractions was determined by the boiling HNO3 method. Also NH4. OAc-extractable K in the soil was determined by shaking 5 g soil for 30 min in 50 ml 1 M, p. 7, NH4OAc solution, followed by filtration. Potassium was assayed by atomic absorption spectroscopy.

      The clay fractions from K-plus treatments (NK, PK, NPK) contained a larger quantity of mica than the K-less treatments (N, NP). The ‘control’ treatment contained higher amounts of mica than other treatments. This was attributed to limited plant growth. Vermiculite in the clay fraction showed a trend opposite to that of mica distribution. Addition of K not only retarded mica weathering but also transformed fine clay vermiculite to a mica-type structure. Treatment effect on smectite was visible only in the fine clay of the Ap horizon. The K-less treatments resulted in reduced boiling HNO3- extractable K from the sand fractions. Also, the sand of Ap horizons contained the lowest boiling HNO3-extractable K Similarly K-less treatments contained less extractable K with boiling HNO3 than the K-plus treatments for the silt fraction. Ammonium acetate extractable K increased for the treatment where K was applied

  9. Page 540