Quantification of lipid biomarkers in sedimentary contexts: Comparing different calibration methods

Abstract

The analysis of lipid biomarkers has gained increasing importance within environmental and archaeological fields because biomarkers are representative of particular plant and animal sources. Accurate quantification of such compounds is essential to obtaining significant information. In this paper, accuracy in the quantification of n-alkanes (as model compounds) in sedimentary contexts was studied using three different approaches: proportions of chromatographic peak areas with regards to an internal standard, instrumental calibration, and matrix-matched calibration. Alkanes were extracted by ultrasonic assisted solid-liquid extraction and analyzed by gas chromatography coupled to mass spectrometry. Instrumental calibration curves were obtained with determination coefficients (R2) higher than 0.9900. The matrix effect was studied (in the range between −100% and 140%) and matrix-matched calibration curves with R2 above 0.9904 were obtained. Matrix matched calibration provided the best results (percentages in the range 10–130%) when compared with instrumental calibration (1–150%) and no calibration (2–274%). Four sediment samples from an archaeological Neanderthal site were analyzed with the three quantification methods and Student’s t-test demonstrated that the results obtained were significantly different: concentrations were non-uniformly under- or overestimated and the highest peak and the alkane profile varied with the different methods. According to these data, matrix-matched calibration should be the preferred method for quantification since then results are more accurate with only a small amount of extra work. This method requires that control sediment samples are collected at the site under consideration.