Possibilities and pitfalls in analyzing (upgraded) pyrolysis oil by size exclusion chromatography (SEC)

• Published in: Journal of analytical and applied pyrolysis Vol. 91; no. 1; pp. 76 - 88
• Main Authors: Hoekstra, Elly, Kersten, Sascha R.A., Tudos, A., Meier, Dietrich, Hogendoorn, Kees J.A.
• Format: Journal Article
• Language: English
• Published: Elsevier B.V 2011
• Subjects: Bio oil; Biomass; Fast pyrolysis; fouling; gel chromatography; Gel permeation chromatography (GPC); molecular weight; Molecular weight distribution; oils; polystyrenes; prediction; pyrolysis; quantitative analysis; Size exclusion chromatography (SEC); Upgraded pyrolysis oil; water content; UVD; Molecular weight distribution; Fast pyrolysis; Size exclusion chromatography (SEC); Biomass; RID; Bio oil; SEC; HDO; THF; Upgraded pyrolysis oil; PAH; Gel permeation chromatography (GPC); GPC; PO
• ISSN: 0165-2370; 1873-250X
• DOI: 10.1016/j.jaap.2011.01.006

The applicability of size exclusion chromatography (SEC) to analyze (upgraded) pyrolysis oil samples has been studied using model compounds, pyrolysis oils and hydrodeoxygenated pyrolysis oils. The assumptions needed for the conversion of the chromatogram to the M w -distribution were validated. It was shown that the conversion of elution volume to molecular weight (based on polystyrene calibration curves) can introduce substantial errors in the prediction of the molecular weight. The conversion of RID response to W(log M) (as plotted on the y-axis of the M w -distribution) is based on the assumption of a compound independent RID response factor and linear response to concentration. While the latter was shown to be true within the concentration range studied, the former was not true: the RID response factor depends on the type of (upgraded) pyrolysis oil. It was shown that within a single pyrolysis oil sample, the RID response for the low molecular weight fraction was a factor 3 lower than the high molecular weight fraction. Furthermore long term column fouling can influence SEC results that cannot be corrected with regular polystyrene recalibrations. Based on the results we recommend SEC not to be used as a quantitative method for characterization (upgraded) pyrolysis oil samples, but as a tool to compare (upgraded) pyrolysis oil samples, preferably prepared using incremental operating conditions and expected to have similar molecular composition. This work has further shown that (i) the ∫UVD dv/∫RID dv ratio can be used as an indication of the sum of the relative aromaticity and conjugated double bond content for (upgraded) pyrolysis oil, and (ii) the negative peak area appearing in the low molecular weight part of the chromatogram can be used to estimate the water content of (upgraded) oil samples.