Minimization of carbon addition during derivatization of monosaccharides for compound‐specific δ 13 C analysis in environmental research

Little is known about the delta13C composition of monosaccharides representing the largest carbon reservoir in the biosphere. The main reason for this might be that monosaccharides have to be derivatized prior to gas chromatography/combustion/isotope ratio mass spectrometry (GC/C/IRMS) analyses and that a large isotopic correction is necessary for the carbon that has to be added to the original molecule during derivatization, resulting in large uncertainty of the calculated delta13C values of individual monosaccharides. The amount of added derivatization carbon is twice (alditol acetates) or even three times (trimethylsilyl (TMS) derivatives) as high as the amount of the original monosaccharide carbon. In addition, isotope fractionation occurs during acetylation. Therefore, the objectives of this study were (i) to minimize carbon addition during derivatization for GC/C/IRMS measurements of monosaccharides in soil and sediment samples and (ii) to quantify improvements in accuracy and precision of the final results. Minimization of carbon addition was accomplished by derivatization with methylboronic acid (MBA) and TMS thereafter (MBA method). Monosaccharides derivatized with the MBA method instead of TMS reduced the number of added carbon atoms from 2.2-2.7 to 0.3-0.8 per sugar carbon atom. Although the precision of GC/C/IRMS measurements with both methods is comparable (about 0.3 per thousand), delta13C values of an internal standard indicated that the newly developed MBA method is about 2 per thousand more accurate than the TMS method. delta13C comparison between soil samples that differed only slightly in their bulk carbon isotope signature showed that the MBA method is better in proving these small differences on a significant level. Total precision of the whole MBA method including all analytical and calculation steps is better by a factor of almost three than the TMS method.