Challenges Related to Analysis of Anions in Degraded Samples from Pilot and Lab Experiments

Abstract

Analyses of degraded samples from CO2 capture experiments and pilots are complicated because of the diversity of the degradation compounds and the high amine and CO2 contents in the samples. Little work has been conducted focusing on the analytical techniques and factors influencing qualitative and quantitative results. Uncertainties in the results have to be considered. Matrix effects can mask the degradation compounds and thereby influence quantification. Separation of the analytes can be a problem and any identification must be supported by reasonable formation mechanisms for the system. In this study a large range of samples from lab experiments and pilots have been analysed using two different IC columns: IonPac AS 15 and AS11-HC. Factors influencing qualitative and quantitative analyses were investigated, for example matrix interactions, baseline separation, differences between columns and sample pre-treatment. Additionally, quantitative uncertainties were estimated such as reproducibility, comparisons between columns and effect of dilution. The two columns used give different separation of the anions. Acetate and glycolate are only separated on IonPac AS15, while IonPac AS11-HC showed better separation of sulphate and oxalate. Formate and HEOX were not baseline separated on any of these columns. The amine matrix influenced retention time for early eluting peaks and made baseline separation more difficult. Extra peaks caused by the matrix were seen for KGly and in general an amino acid matrix, such as KSAR and KGly, created signal noise, thereby impairing quantification. Sample pre-treatment with an OnGuard II H cartridge was tested and found to be of limited use. It should be used with care not to falsify the analyses. The analytical uncertainty was influenced by missing baseline separation between some analytes, low concentrations of analytes and wide peaks due to low eluent concentration. © 2016 The Authors.