Mid-infrared spectral interpretation of soils: Is it practical or accurate?

There has been an increasing interest in the use of mid-infrared spectroscopy for the quantitative analysis of soils. Understanding the spectral bands can be beneficial both for understanding the basis of the quantitative analysis and soil composition and structure in general. Unfortunately, at least from the standpoint of the organic composition of soils, the spectra of most soils are dominated by the spectra of inorganic fractions such as clays, silica, carbonates, etc. One method used to accentuate the signature of the organic fraction has been to ash soil samples and then use spectral subtraction, in theory leaving only the spectral signature of the organic fraction. The objective of this work was to investigate exactly how accurate/practical this procedure is and is derived from recent comments from reviewers. Results examining silt, clays, silica, calcium carbonate, soils and materials such as cellulose and lignin show that, without considerable prior knowledge of a soil sample, spectral interpretation in most regions of the mid-infrared is highly interpretive/subjective at best. While the region for CH absorptions (3000-2800 cm(-1)) is free of interferences, except for carbonates which can be removed by acidification, and the region between 1750-1600 cm(-1) can be interpreted despite the presence of strong silica bands, interpretation of the remaining regions of the mid-infrared are deemed very subjective. This is largely, but not entirely, due to the fact that many types of clay undergo extreme spectral changes from ashing and materials with seemingly similar formulas such as kaolinite (Al2Si2O5(OH)(4)) and pyrophyllite (Al2Si4O10(OH)(2)) change differently when ashed. Thus unless one knows the exact clay/mineral composition of the soil in question and the effects of ashing on said clay/mineral, accurately interpreting the ash subtracted spectra is nearly impossible. Also, even subtracting silica was found to be difficult due to changing specular effects even with KBr diluted samples. From a practical standpoint, one must conclude that the routine use of ash subtraction is not practical for soils as too much needs to be known to judge the results. Published by Elsevier B.V.