This paper focuses on the application of a two-level full factorial design to optimize the key derivatization step before the GC-FID and GC-MS analysis of pentacyclic triterpenes. The derivatization reaction was screened for influential factors and statistically significant parameters with a p value less than 0.05. A multi-response optimization based on a desirability function was then applied, while simultaneously considering overall detection enhancement of compounds. Results showed that derivatization using N,O-bis(trimethylsilyl)trifluoroacetamide (BSTFA) and trimethylchlorosilane (TMCS) in pyridine (22:13:65v/v/v) for 2h at 30°C was the most efficient method of derivatizing all the hydroxyl and carboxylic acid groups contained in the triterpene structures. The validity of the method was demonstrated using GC-MS analyzes of a mixture containing eleven standards (β-amyrin, α-amyrin, lupeol, erythrodiol, uvaol, betulin, oleanolic acid, betulinic acid, ursolic acid, maslinic acid and corosolic acid). These compounds are representative of different classes of terpene compounds bearing different functional groups such as alcohols, diols, and carboxylic acids. The derivatization procedure was then tested on four plant extracts: apple pomace, salvia sclarea (dried leaves and flowers), sea buckthorn (Hyppophae rhammnoides L.) berries, and B. serrata resin. The identification of triterpenes was based on the comparison of their retention time and mass spectra to those of standards. The presence of compounds already identified in the literature was confirmed and new ones such as maslinic and corosolic acids were identified in apples, sea buckthorn and salvia sclarea.
Keywords: BSTFA/TMCS/pyridine; Central composite design; Derivatization optimization; GC–MS; Pentacyclic triterpenes; Plant matrices.
Copyright © 2015 Elsevier B.V. All rights reserved.