The compositions of humic acids (HAs) from various Histosols in North America and Europe, of similarly treated plant‐extracted materials (PEMs), of coal‐extracted humic acids, and of International Humic Substances Society (IHSS) Florida peat were quantified by solid‐state 13C nuclear magnetic resonance (NMR). In order to obtain quantitative intensities, the peak areas in direct‐polarization 13‐kHz magic‐angle spinning (DPMAS) 13C NMR spectra were corrected for incomplete relaxation by factors measured in cross‐polarization spin‐lattice relaxation time (CP/T1) experiments with total sideband suppression (TOSS). The elemental compositions (%C, %H, %O + N) of a peat sample, 8 HAs and 2 PEMs were estimated from the NMR results and compared with chemical analyses, as well as solution NMR for two of the HAs. The results are in good agreement, which shows that DPMAS corrected by CP/T1–TOSS permits quantitative characterization of HAs and PEMs. The compositions of the PEMs deviate significantly from those of the Histosol HAs. The compositions in terms of nine types of chemical groups were computed. The investigated HAs consist of more than 60% of aromatic and CO carbons (including both carbonyl and carboxyl groups). Previous cross‐polarization magic‐angle spinning (CPMAS) NMR experiments have significantly underestimated the ratio of sp2– to sp3–carbons; in particular, the true COO carbon fraction is a factor of two larger than estimated by CPMAS NMR. In spite of their wide range of geographical origins, the compositions of the Histosol HAs appear to be relatively uniform, suggesting that the search for a general model of Histosol HA structure is worthwhile. Eight models proposed in the literature do not reproduce the experimentally determined compositions, but a few models show promising partial agreement.

Analytical and thermodynamic data, EPR, FTIR, solution 1 H and solid-state 13 C cross polarization magic angle spinning NMR and solid-state extended X-ray absorption fine structure (EXAFS) and X-ray absorption nearedge structure (XANES) spectra have been recorded for purified humic acids (HAs) isolated from a German peat (GHA), an Irish peat (IHA), an unpolluted New Hampshire bog soil (NHA) and their tightly bound copper(), iron() and manganese() forms. Brief water washing of partly or fully metal-loaded HAs leaves 'tightly' bound metal in the isolated freeze-dried solids. Most of this metal is removed by washing with 0.1  HCl, indicating acidic HA functional groups as principal metal binding sites. The number of nearest-neighbour atoms coordinated to tightly bound Cu II (four), Fe III (six, probably with distorted geometry) and Mn II (six, undistorted) in solid GHA, IHA and NHA were determined by XANES and EXAFS spectroscopy with reference standards. Isotherms measured at 20.0 ЊC and pH 2.4-3.2 with [M] total = 0.18-25.8 m for tight, reversible Cu 2ϩ (aq), Fe 3ϩ (aq), and Mn 2ϩ (aq) binding by solid IHA and NHA fit the Langmuir model and give the pH-independent stoichiometric site capacities ν i and equilibrium constants K i for metal binding at specific HA sites i = A, B and C. Tight binding sites A, B and C of IHA are occupied by Cu II , sites A and B by Fe III and site A by Mn II , while only identical metal binding site A in NHA is tight enough to resist metal removal by brief water washing. A new helical HA molecular model based on the empirical formula C 36 H 30 N 2 O 15 ؒxH 2 O visualizes metal binding and the likely roles of HAs in biomineralization. Site A is suggested to be carboxylate, mixed ligands probably constitute site B, and site C is tentatively assigned as the interior of the HA helix. Binding free energies and EPR evidence suggest that Cu 2ϩ (aq), Fe 3ϩ (aq) and Mn 2ϩ (aq) rapidly transfer between specific HA binding sites. This affects rates of metal release and transfer to minerals.

Adsorption of the naturally occurring pyrimidine nucleobases and their corresponding nucleosides and nucleotides on compost-derived humic acid (HA) has been studied in aqueous solution. All these nine solutes interact with HA. Adsorption isotherms were measured at 19 solute concentrations from 12-600 µM and 6 temperatures in the range 10.0-40.0°C. Analysis with the Langmuir model indicates that adsorption occurs in three detectable steps A, B, and C, depending on the solute and the experimental conditions. Most of the data fit the Langmuir model, which generates a site capacity V i and an equilibrium constant K i for each detectable adsorption step. Site capacity and solute selectivity data indicate that the HA sites responsible for adsorption steps A, B, and C are chemically different and that pyrimidine solutes are adsorbed through their nucleobase units. Adsorption of pyrimidines on HA is exothermic, thermoneutral, or endothermic, depending on the solute and the adsorption step. Linear correlation of adsorption enthalpies and entropies points to similar surface interactions.