Speciation of aqueous mononuclear Al(III)-hydroxo and other Al(III) complexes at concentrations of geochemical relevance by aluminum-27 nuclear magnetic resonance spectroscopy

Aluminum-27 (27Al) nuclear magnetic resonance (NMR) spectroscopy was used to characterize Al(III)-hydroxo complexes, in aqueous solutions with total Al(III) concentrations of 1.0–10 μM, using a custom-built NMR probe, coil, and sample bottle with low background aluminum impurities. Published 27Al NMR spectroscopy studies have traditionally used total Al(III) concentrations that are generally outside the range of geochemical interest (total [ Al(III) l ≥ 1000 μM). In this study, lower Al(III) concentrations (≤ 10 μM) were used to more closely approximate natural conditions, while allowing the measurement of mononuclear Al(III) species by 27Al NMR spectroscopy. The sensitivity of the 27Al NMR spectroscopy system, as measured by the signal-to-noise ratio (S/N), is S/N= 5 for 1.0 μM Al(III) at pH 2.00 and S/N= 3 for 10 μM Al(III) at pH 5.20. This level of sensitivity is within the range of geochemically relevant Al(III) concentrations found in slightly acidic natural waters. Quantitative models are developed which link observations of NMR chemical shifts and linewidth ratios to the calculated equilibrium speciation of mononuclear Al(III) for 10 μM Al(III) solutions at pH values 2.00 to 5.20 (prepared by titrating acidic AI(III) solutions with pyridine). Linear-regression best fits of the models to the NMR data are used to determine the intrinsic chemical shifts and linewidths of individual mononuclear Al(III) species. The intrinsic chemical shift of each Al(III) species “i”, δi (ppm), is (1) δAl3+ ≡ 0 for Al3+ (defined by convention), (2) δAl(OH)2+ = 3.5 (SE= 1.3, N = 10) for Al(OH)2+, (3) δAl(OH)2+ ≅ 3.7 (SE= 1.4, N = 10) for Al(OH)2+ , and (4) δAl(OH)4− = 79.9 (SE= 0.03, N = 4) for Al(OH)4−; where positive chemical shifts are “downfield,” SE = standard error, and N =number of samples. A convention is delineated in which the linewidth of the Al(III) species/peak of interest is normalized with respect to that of a reference species (Al3+) under the same conditions. Such linewidth ratios are independent of investigation-specific variables such as solution viscosity and temperature. Due to the large sample volume (18.8 mL) used here to achieve increased sensitivity, there is some line broadening caused by magnetic field inhomogeneities; however, this line broadening is constant and reproducible both during and between experimental runs, and it was corrected for in the determination of linewidths of individual Al(III) species. For an absolute linewidth of LWAl3+