Many soils developed from volcanic rocks in southern Brazil exhibit spontaneous magnetization caused by the presence of fine-grained maghemite (γ-Fe2O3), but few attempts were made to quantify or characterize this important soil component. To that end, clays were separated from freely drained soils derived from acid (≥63% SiO2), intermediate (54–62% SiO2), and basic (≤53% SiO2) igneous rocks produced by the Paraná flood volcanism. The sample set included soils with a wide range of pedogenic development on different landscape positions. The Fe oxide mineralogy of these samples was examined by using a combination of selective dissolution, magnetic susceptibility, and X-ray diffraction (XRD) techniques. Hematite and maghemite were the primary Fe oxides in mature soils (Oxisols, Ultisols, and Alfisols) developed from basic rocks; whereas goethite was dominant in all other soils, especially those formed from acid-intermediate rocks. The association of maghemite with basic rock materials suggests that it was primarily formed by oxidation of lithogenic magnetite. A strong, positive correlation (R2 = 0.89) was obtained between mass specific magnetic susceptibility (χ) of the clay fractions and maghemite contents estimated by XRD. Either method could be used for quantitative analyses, but χ was more sensitive than XRD at low maghemite concentrations (<2 wt. %). The clay-sized maghem-ites were superparamagnetic with an estimated value for the mass specific magnetic susceptibility (χlf) value of 91,000 × 10−8 m3 kg−1 and frequency dependent variations of 10–15%. The maghemites also had low unit cell constants, which, if attributed entirely to replacement of Fe by Al, would correlate with Al substitutions in the range of 5–16 mole %. Selective dissolution of the soil maghemites was achieved by treatment of Fe oxide concentrates with 1.8 M H2SO4 at 75°C for 2 h.

Hydrotalcites of high aluminum content have been synthesized from aluminate liquors of varying composition and activated magnesia obtained by calcination of hydroxide or hydroxycarbonate precursors. Lattice parameter measurements and chemical analyses of 21 synthetic hydrotalcites show that the aluminum substitution $$\left( {\rm{X = \frac{{Al}}{{Al + Mg}}}} \right)$$ for most of the products is about 0.35, which is at the maximum experimentally-observed limit of solid solubility. Pillared hydrotalcites were also prepared by molybdate, Chromate, and silicate anion replacement. A maximum distance of 10.4 Å between the brucite-like layers was observed for the Mo7O246− intercalated material.

Submicron-sized (~3–60 nm) powders of Al-substituted magnetite were synthesized in the laboratory by precipitation methods by mixing appropriate molar volumes FeCl2, FeCl3 and AlCl3 solutions and precipitating with 20% NH4OH. Precipitates were dialyzed for 48 hr to remove excess salts and then freeze-dried. The nominal Al mole fractions [Als = Al/(Al + Fe)] in the initial precipitate ranged from 0.001 to 0.42. Portions of the resulting powders were heated sequentially in air at 400° and 500°C. Powders were examined using X-ray diffraction (XRD), transmission electron microscopy (TEM), and visible and near-IR reflectance spectroscopy. Solubilities were determined in ammonium oxalate (pH = 3) and dithionite-citrate-bicarbonate (DCB) solutions. As determined by XRD, the mineralogy of precipitated powder samples was predominantly magnetite. Powders having Als > 0.20 contained minor goethite and a poorly crystalline iron oxide phase (ferrihydrite?), and powders having Als > 0.25 also contained gibbsite. The color of the magnetites was black throughout the range of Al-substitution. Powders heated to 400°C were reddish brown; Munsell colors ranged from 5R 2/2 to 10R 3/4 for Als from 0.1–41.5%, respectively. By XRD, these powders were maghemite, but hematite was also detected by Mössbauer spectroscopy. XRD and Mössbauer data indicate powders heated to 500°C are hematite; their Munsell colors are not noticeably different from the corresponding 400°C samples. Mean crystallite dimensions (MCDs) of the magnetite powders increase with the Al mole fraction from ~10 nm for Als = 0.001% to a maximum value of 35 nm for Als = 0.15 and decrease slightly with further increasing Al substitution. Heating magnetite powders to 400°C did not change the MCDs significantly. Heating to 500°C resulted in hematites having MCDs larger than those for corresponding precursor magnetites for Als < 0.10. The opposite is true for hematites derived from magnetites having Als > 0.10. For hematite powders with Als > 0.05, MCD decreased with increasing Al-substitution. Solubilities of powders in oxalate solutions were independent of Al content and decreased in the order unheated samples (mostly magnetite) >400°C-heated samples (maghemite + hematite) >500°C-heated samples (hematite). All powders dissolved completely in DCB. The low crystallinity of the magnetite powders and the presence of ferrous iron are responsible for their relatively high solubility in oxalate solutions.

Aluminum substitution is a common phenomenon in environmental iron oxides and oxyhydroxides, affecting the color, magnetic character, surface features, etc. Several methods for preparing Al-substituted iron oxyhydroxides can be found in the literature, resulting in samples with particular properties. In the present study, the synthesis of aluminum-substituted goethites, AlxFe1-xOOH with 0 ⩽ x ⩽ 0.15, by homogeneous precipitation and the transformation to aluminum-substituted hematites, (AlxFe1-x)O3, are presented. The goethite samples were produced at 90°C from solutions of urea and iron and aluminum nitrates in the presence of ammonium sulfate (GU series). Although attempts were made to incorporate up to 33 mole% of Al into the goethite, only ~15 mole% was found to be within the structure, due to the final pH, ~7, of the synthesis. Another feature of these goethites was a lateral alignment of the tabular particles. By heating batches of the GU samples at 400°C and 800°C, two series of Al-hematites were obtained, denoted here as the HX400 and HX800 samples, respectively. X-ray diffraction, thermal analysis, Karl-Fischer titration, transmission electron microscopy, and Mössbauer spectroscopy were used to characterize the samples. The X-ray patterns showed the samples to be pure iron phases, with particle sizes of ~10 nm for the GU and HX400 samples, and of ~70 nm for the HX800 samples. An inversion in the intensities of the (104) and (110) diffraction peaks of hematite was observed to be dependent on the aluminum substitution and was explained by small particle sizes, shape anisotropy, and the presence of nanopores. The cell parameters of both GU and HX samples showed a small decrease with increasing aluminum substitution up to x ≈ 0.15. The amount of adsorbed sulfate, presumably as an aluminum hydroxy sulfate gel, increased with aluminum substitution in all GU and HX samples, reaching a maximum of ~6.5 wt.% for the highest substitution. Heating at 100°C did not remove all of the adsorbed water, and significantly higher temperatures were required to achieve complete removal. Mossbauer spectra at 295 K and 80 K are typical for small-particle goethite and hematite, and revealed that Al-for-Fe substitution in all samples seems to be limited to ~15 mol.%.