Gibbs energy additivity approaches to QSRR in generating gas chromatographic retention time for identification of fatty acid methyl ester

Siriluck Pojjanapornpun, Kornkanok Aryusuk, Supathra Lilitchan, Kanit Krisnangkura

The Gibbs energy additivity method was used to correlate the retention time (t R) of common fatty acid methyl esters (FAMEs) to their chemical structures. The t R of 20 standard FAMEs eluted from three capillary columns of different polarities (ZB-WAXplus, BPX70, and SLB-IL111) under both isothermal gas chromatography and temperature-programmed gas chromatography (TPGC) conditions were accurately predicted. Also, the predicted t R of FAMEs prepared from flowering pak choi seed oil obtained by multistep TPGC with the BPX70 column were within 1.0% of the experimental t R. The predicted t R or mathematical t R (t R(math)) values could possibly be used as references in identification of common FAMEs. Hence, FAMEs prepared from horse mussel and fish oil capsules were chromatographed on the BPX70 and ZB-WAXplus columns in single-step and multistep TPGC. Identification was done by comparison of t R with the t R of standard FAMEs and with t R(math). Both showed correct identifications. The proposed model has six numeric constants. Five of six could be directly transferred to other columns of the same stationary phase. The first numeric constant (a), which contained the column phase ratio, could also be transferred with the adjustment of the column phase ratio to the actual phase ratio of the transferred column. Additionally, the numeric constants could be transferred across laboratories, with similar correction of the first numeric constant. The TPGC t R predicted with the transferred column constants were in good agreement with the reported experimental t R of FAMEs. Moreover, hexane was used in place of the conventional t M marker in the calculation. Hence, the experimental methods were much simplified and practically feasible. The proposed method for using t R(math) as the references would provide an alternative to the uses of real FAMEs as the references. It is simple and rapid and with good accuracy compared with the use of experimental t R as references.