Compound-specific stable carbon isotope ratios of phenols and nitrophenols derivatized with N,O-bis(trimethylsilyl)trifluoroacetamide

•For GCC–IRMS analysis, an approach avoiding impact of NO2 on δ13C was demonstrated.•Carbon isotope fractionations during derivatizing reactions here were negligible.•Except some labile compounds, the overall bias of the method here was −0.21‰.•Even for the labile compounds, measurement biases ranged +1.2‰ to −1.4‰.•Real sample analysis demonstrates usefulness of the method for fractionation study. We developed an analytical method for measuring compound-specific stable carbon isotope ratios (δ13C) of phenols and nitrophenols in filter samples of particulate organic matter. The method was tested on 13 phenols derivatized with N,O-bis(trimethylsilyl)trifluoroacetamide (BSTFA), together with four nonphenolic compounds. The data obtained by our method required two specific corrections for the determination of valid δ13C values: (1) for nitro compounds, the routine correction with use of m/z 46 for the contribution of 12C17O16O molecules) to m/z 45 was modified due to impact of NO2 on the m/z 46 trace, and (2) for the derivatized phenols, measured δ13C values were corrected for the shift in δ13C due to the addition of carbon atoms from the BSTFA moiety. Analysis of standard-spiked filters showed that overall there was a small compound-dependent bias in the δ13C values: the average bias±the standard error of the mean of −0.21±0.1‰ for the standard compounds tested, except 3-methylcatechol, methylhydroquinone, 4-methyl-2-nitrophenol, and 2,6-dimethyl-4-nitrophenol, whereas the average biases±the standard errors of the mean for those were +1.2±0.3‰, +1.2±0.2‰, −1.2±0.2‰, and −1.4±0.5‰, respectively, when the injected mass of a derivatized compound exceeded 15ngC. In situations where such small biases and uncertainties are acceptable, the method described here could be used to obtain valuable information about δ13C values. We also analyzed a real filter sample to demonstrate the practical applicability of the method.