Based on the technical methods of GB/T 42430-2023 and GA/T 204-2019, this study established an analytical method for headspace injection double-column dual-detector (hydrogen flame ion detector) gas chromatography for the simultaneous analysis of at least 12 volatile compounds, including ethanol, in human blood using two different equipment platforms and chromatographic columns. A 100 μL blood or urine sample and a 0.04 g/L tert-butanol working solution prepared as an internal standard are introduced into the headspace sample bottle and then sealed, mixed, and placed on the headspace sampler rack. Using different equipment platforms and columns, methodological parameters such as the limit of detection (LOD), limit of quantification (LOQ), precision, and accuracy of the method were systematically evaluated. The chromatographic separation of acetone, alcohols and benzenes using the established method was satisfactory. The linear ranges, linear correlation coefficients (r), and LODs of acetone and six alcohols, including ethanol, were 0.10-3.00 g/L, >0.997, and 0.05 g/L, respectively. The LOQs were 0.10 g/L for all other compounds, excluding n-propanol (0.005 g/L). Additionally, the linear ranges, r values, LODs, and LOQs of benzene and four benzene derivatives were 0.05-50 mg/L, >0.995, 0.02 mg/L, and 0.05 mg/L, respectively (Column J&W DB-BAC1 UI and Column Rtx-BAC-PLUS 2). The average recoveries of compounds on J&W DB-BAC1 UI and Rtx-BAC-PLUS 2 columns ranged from 92.2% to 111.6%, and the relative standard deviations (RSDs, n=6) ranged from 0.4% to 7.4%. The LOD, LOQ, precision, accuracy, and linearity of the established method met the requirements of relevant standards, and no significant differences arose between the methodological parameters of the two platforms. CNAS-GL006 (2019) and JJF 1059.1-2012 were used as guides to evaluate the uncertainty of ethanol on two different sets of equipment platforms and chromatographic columns. The ethanol uncertainty was mainly derived from the calibration curve; however, the confidence probability was 95% (k=2). According to the analysis of the verification samples and real samples, the established method is suitable for the high-precision quantitative analysis of acetone and six alcohols and five benzene derivatives in human blood and other body fluids. It can be used in practical scenarios such as judicial identification and the detection of poisons.
Read full abstract