This paper addresses the application of thermal analytical methods (DTA/DSC/TG/ MS) in soil investigations, especially the interrelation of exothermic effects and analytical or preparative parameters. For the superimposed decomposition/ oxidation reactions of siderites, FeCO3, the heating rate is shown to have a decisive influence because low heating rates tend to suppress the endothermic effect. In mixtures of small amounts of organic matter (glucose and starch) with talc, the phyllosilicate acts as a catalyst by influencing the intensity of combustion of the organic matter.

The physical properties of three montmorillonites before and after heating were studied by reflectance colorimetry, X-ray diffraction (XRD) and infrared (IR) and Mössbauer spectroscopies. Colorimetric data, though not directly relatable to sample mineralogy, provide cheap and rapid information on the temperatures of sample breakdown. X-ray diffraction provides information on sample mineralogy and the transformations that develop during firing and allows the major phases formed in the course of firing to be identified, whereas IR data provide information on water and hydroxyl loss and the final breakdown of the montmorillonite structure. Mössbauer spectroscopy gives information on Fe2+ oxidation (where applicable) and reveals subtle variations in Fe site coordination and Fe mineralogy, including the formation of Fe oxides – that can be quantified in proportions as low as 0.1% – following structural breakdown of the montmorillonites.

This paper presents various aspects, revealed by Mössbauer spectroscopy, of structural and magnetic properties of Al-substituted small-particle soil-related oxides. For goethite we focus on the relations between the hyperfine fields on the one hand, and crystallinity and Al content on the other. It is argued that these relations only provide a rough estimate of the Al content in natural samples. The ferrimagnetic-like behaviour reflected in the external-field Mössbauer spectra (4.2 K, 60 kOe) of certain Al goethites is presented. The spectra obtained for lepidocrocites are not spectacular, but confirmed that up to ∼10 at.% Al can be incorporated in the structure. Three differently-made series of hematites are considered. The Morin transition and spin structures in hematite are very sensitive to crystallinity and Al content, and probably to the presence of structural OH–. Integral low-energy electron Mössbauer spectroscopy on non-substituted samples indicates that the Morin-transition temperature in the surface layers (2 to 5 nm) is not significantly shifted from the bulk value. Measurements in extremely high magnetic fields (140 kOe) have shown that a spin-flip transition is induced in highly-substituted samples which exhibit no Morin transition in zero field. The use of external fields is crucial for the characterization and precise determination of the hyperfine parameters and site occupancies for maghemites, and for phase analyses of magnetic soils.

Visible to near-infrared (NIR) reflectance spectra and mid-IR transmittance spectra are presented here for a collection of dioctahedral smectites. Analysis of the structural OH vibrations is performed by comparing the NIR combination and overtone bands with fundamental stretching and bending absorption features in the mid-IR region. Second derivatives are used to determine the actual band centres, which are often shifted slightly by a spectral continuum in the reflectance or transmittance spectra. New bands have been identified near 4170 and 4000 cm–1 in the NIR spectra of nontronite with tetrahedral substitution. A related band is observed near 4100 cm–1 for montmorillonites with substantial tetrahedral and/or octahedral substitution. These bands are correlated with the mid-IR bands near 680 cm–1 for nontronite and near 630 cm–1 for montmorillonite. Comparison of the OH overtone and combination bands with the fundamental stretching and bending vibrations gives consistent results.

Analysis of the near-infrared (NIR) spectral bands of phyllosilicates, together with the mid-infrared bands, enables testing and confirmation of band assignments for the structural OH vibrations. Spectral analyses of selected smectites and serpentine-kaolin minerals are presented here. The results of this study indicate that dioctahedral smectites may contain both in-plane and out-of-plane OH-bending vibrations, as suggested by Farmer (1974). In-plane bands occur near 800 ­ 915 cm–1, while the weaker out-of-plane vibrations occur near 600 ­ 700 cm–1 and are enhanced in dioctahedral smectites when the structure is disrupted by substitutions. Analysis of the OH-stretching vibrations and their NIR overtone bands is also presented for both smectites and serpentine-kaolin minerals. These overtones are more straightforward for serpentines than for kaolinite, and a strong overtone associated with the kaolinite OH-stretching band at 3620 cm–1 is found that supports its assignment as an OH-stretching band.

X-ray powder diffraction patterns were simulated for nano-sized hematite, goethite and lepidocrocite by three-dimensional integration in reciprocal space. The cell-edge lengths were refined together with the size parameters X and Xe of the Thompson-Cox-Hastings function, which for orthorhombic structures was extended by a biaxial broadening parameter Xo. Variations of the structure factors across broad peaks resulted in apparent peak shifts and concomitant shifts in celledge lengths, which were significantly correlated with the size parameters for hematite and partially correlated for goethite and lepidocrocite. Regression equations are given for correcting cell-edge lengths obtained from Rietveld fits for size-induced shifts.

The crystal-chemical mechanisms by which transition metals are associated with goethite are fundamental to our understanding of the solubility and bioavailability of micronutrients and heavy metals in soils, and in the formation of laterite ore deposits. Transition metals such as Cr, Mn and Ni may sorb onto goethite by forming surface precipitates, surface complexes or by replacing Fe3+ in the goethite structure. In the work reported here, we investigated the local coordination environment of Cr, Mn and Ni in synthetic goethite using EXAFS spectroscopy. We demonstrate the isomorphous substitution for Fe3+ by Cr3+ (up to 8 mol.%), Mn3+ (up to 15 mol.%) and Ni2+ (up to 5 mol.%). We find, however, that the next-nearest-neighbour coordination environment changes with composition. The perturbations are likely to be responsible for limiting the accommodation of Cr3+, Mn3+ or Ni2+ in the FeOOH structure.

Some soils from the western Amazon region contain KCl-extractable Al contents 5 to 10 times greater than is typical for highly weathered soils containing predominantly kaolinite and gibbsite. We studied a soil sequence from the Brazilian western Amazon consisting of two Typic Udifluvents on the levee of the Javari River, one Aeric Endoaquent in the backswamp, and two Typic Hapludults on an adjacent terrace. We used wet chemical and X-ray diffraction (XRD) analysis to characterize several size fractions of the 0 to 0.2 m layer of the soils. The exchangeable Al content was very high in the Aquent and Udults (up to 180 mmolc kg–1), but the ‘total’ Fe content was low in all samples (<60 g kg–1). Smectite, vermiculite, hydroxy-interlayered smectite and kaolinite dominate the fine silt and clay fractions of all soils. The Fluvents contain illite in all size fractions and chlorite in the coarse clay and fine silt fractions. The Aquent and Udults have no chlorite, and small amounts of illite occur only in the coarse clay and fine silt fractions. Lepidocrocite was identified in the Aquent. Chlorite, which occurs in the sand, fine silt, and coarse clay fractions of the Fluvents, and pyrophyllite, which occurs in the fine silt fractions of all soils and in the coarse clay of the two Ultisols, appears to be inherited from the parent sediments. The hydroxy-interlayered 2:1 phyllosilicates that form as a result of weathering are the cause of the very high exchangeable Al contents.

Three soil profiles classed as brown forest soils on schistose parent material, in close proximity but under different land uses, have been studied. The profile under 50 year-old Sitka spruce trees is the most acidic, the second profile under rough grassland is less acidic, and the third profile in an arable field which has been fertilized is the least acidic. The clay fractions (ς μm) of all three profiles are very similar. However, computer simulations of the X-ray diffraction patterns of clay phases using a modified version of NEWMOD revealed two possible weathering trends: (1) an increasing proportion of vermiculite in interstratified mica-vermiculite in the upper horizons of the arable and forested soils; (2) formation of high-charge corrensite by weathering of chlorite in all three profiles but least pronounced in the arable soil. The differences in clay mineralogy amongst the profiles are minor, but these two different weathering trends may be due to the effects of different land use.

Soil kaolins from Indonesia and Western Australia and a range of reference kaolins were studied using Mössbauer spectroscopy, electron paramagnetic resonance (EPR) spectroscopy and SQUID magnetometry. Mössbauer spectra indicate that the Fe within the kaolins is in the highspin Fe3+ oxidation state and that a large fraction of the Fe is present as dispersed atoms residing within the octahedral sites of the kaolinite crystal structure. The EPR spectra are typical for soil kaolins except for the absence of radiation-induced defects for the Indonesian kaolins. The Fe(I) spectra are dominant with a strong symmetric peak at g= 4.3, the presence of Fe(II) spectra is shown by a shoulder on this peak at g= 4.9 and a small phase up peak at g= 9.2.

Low-temperature (5 K) magnetization (M) measurements over large field (H) sweeps (±70 kOe) yielded M(H) curves which are fitted well with Brillouin functions indicating the paramagnetic nature of the kaolins at temperatures down to 5 K. A very small remanent magnetization was detectable in the kaolins. Remanent magnetization to saturation magnetization ratios ranged from 10–4 to 10–3 for the Indonesian kaolins and were all ∼10–3 for the Western Australian kaolins, indicating that at high fields the vast majority of the magnetization of the kaolins is due to paramagnetic ions.

Zero-field-cooled and field-cooled magnetization measurements in small fields (500 Oe) indicate that the Indonesian kaolins are generally free from magnetically-blocked material down to a temperature of 5 K. The magnetic susceptibility of the Indonesian kaolins shows Curie Law behaviour indicating paramagnetic behaviour over all temperatures down to 5 K. Measurements on the Western Australian kaolins indicated the presence of some magnetic material that is magnetically blocked at temperatures below ∼200 ­ 250 K. As a consequence, the magnetic susceptibility showed large deviations from Curie Law behaviour.

Three terrace soils comprising a hydrosequence were examined to determine how the spatial distribution of lepidocrocite was related to depth and duration of saturation. Vertical relief was 1.0 m with well drained, moderately well drained, and somewhat poorly drained pedons spaced ∼60 m apart. All soils contained brittle, slowly-permeable subsoil horizons and were acidic with <35% base saturation throughout the upper sola. The well drained soil (Fragic Hapludult) had no morphological indicators of wetness within a depth of 180 cm, and water was perched above a brittle horizon at 82 cm for a total of only 41 days during the 3.4 year observation period. Nevertheless, trace amounts of lepidocrocite were detected in the subsoil. The moderately well drained soil (Typic Fragiudult) was saturated at a depth of 180 cm for 6% of the time, and water was perched on top of a fragipan at 74 cm for 13% of the time. Lepidocrocite was most abundant in this pedon and reached maximum concentrations below the fragipan in the capillary fringe of the regional water table (150–183 cm). The somewhat poorly drained member of the hydrosequence (Aeric Fragiaquult) was saturated at a depth of 180 cm for 96% of the observation period and also contained perched water above a fragipan for >90% of the time. Lepidocrocite occurred throughout this pedon but was most concentrated in fragipan horizons (86–135 cm) between the perched and regional zones of saturation. These horizons were saturated from 22 to 48% of the observation period. The results of this study suggest that lepidocrocite formation was favoured in horizons that were saturated for 5–50% of the time when soil temperatures exceeded 5°C.

The formation of clay minerals was investigated in Spodosols developed in the subalpine belt, with similar exposure, climate and age, but deriving from different parent materials. All the soils were classified as Haplic Podzols and showed the characteristic eluviation and illuviation features of Fe, Al and organic carbon. However, varying parent material lithology led to different clay mineral assemblages in the soil. Smectite could be found in the E horizons of soils developed from granodiorite and tonalite materials. Its formation was strongly dependent on the presence of chlorite in the parent material. If nearly no other 2:1 mineral components, such as chlorite, are present in the lower soil horizons, then a residual micaceous mineral becomes the dominant clay mineral. The latter derives from a mica-vermiculite interstratified mineral.

Vivianite [Fe3(PO4)2·8H2O] is easily oxidized in the presence of air. In this work, we studied the oxidation and incongruent dissolution of vivianite in a calcareous medium containing an anion-exchange resin (AER) that acted as a sink for phosphate. Freshly prepared vivianite suspensions with calcite sand and an AER membrane were oxidized and stirred by bubbling air or CO2-free O2. Experiments were finished when oxidation rate and P removal by the AER became slow, which was at 53 days (air system) or 28 days (CO2free O2 system). At these times, the respective values of the Fe(II)/total Fe ratio were 0.29 and 0.12, and the respective values of the atomic P/Fe ratio were 0.15 and 0.11. The final product of the oxidation was poorly crystalline lepidocrocite in the form of thin (1 ­ 5 nm), irregular lamellae that were soluble in acid oxalate. The unit-cell edge lengths of this lepidocrocite were a= 0.3117, b= 1.2572, and c= 0.3870, vs. a= 0.3071, b= 1.2520 and c= 0.3873 nm for the reference lepidocrocite. The lepidocrocite lamellae contained occluded P (P/Fe atomic ratio = 0.03 ­ 0.04). The results of this and a previous study made us hypothesize that this occluded P is structural and occupies tetrahedral sites adjacent to the empty octahedral sites in the sheets extending on ﹛010﹜.

Goethite, lepidocrocite, magnetite and akaganeite were synthesized in 0.8 M, 2 M and 5 M NaCl solutions at various temperatures (25, 40, 60°C) under slightly acidic to slightly alkaline pH with or without Si additions. Elevated temperatures prevent complete oxidation of initial Fe2+ solutions and magnetite and siderite precipitate, accompanied by goethite and lepidocrocite. At higher salinity, O2 solubility is reduced and its distribution is limited, leading to coprecipitation of lepidocrocite, akaganeite and goethite.

Lepidocrocite morphology changes from plates at pH 5.5 through rods at pH 7 to multi-domainic crystals at pH 8.2, due to enhanced crystal growth along the c axis. Salinity and temperature have opposite effects on lepidocrocite crystallinity.

Goethite crystals are multi-domainic and twinning appears only at elevated temperatures. Increases in temperature and salinity improve goethite crystallinity as observed by IR spectra. Addition of Si up to Si/Fe = 0.1 retards crystal growth and Si-OH-stretching bands appear. At Si/Fe = 1 most of the precipitate is short range ordered.

Platy and rod-shaped lepidocrocite from the Thetis and Atlantis II Deeps, were probably formed under the slightly acidic conditions of the hydrothermal brines. The Si concentration was greater in Atlantis II Deep than in Thetis Deep, leading to larger lepidocrocite and goethite crystals in the latter.

Multi-domainic goethite could have precipitated throughout. Pure phase goethite might have precipitated in the less concentrated brine, whereas mixtures of goethite and lepidocrocite might have precipitated in the more concentrated brine, depending mainly on oxidation rate and oxygen mobility within the brine.