Development of a GC-FID method for quantitative characterisation of polyglycerols in polyglycerol polyricinoleate (PGPR) present in a lipid matrix

Abstract

PGPR is an emulsifier (E476) widely used in the food industry. In this study, a gas chromatography–flame ionisation detection (GC-FID) method was developed for the quantitative characterisation of the polyglycerol composition of PGPR. The method was validated to analyse quantitatively the polyglycerol species in neat PGPR products and in PGPR samples present in a lipid matrix. This method consists of saponification, acidification and petroleum ether extraction to remove interfering fatty acids, neutralisation, silylation and finally GC-FID analysis. Phenyl β-D-glucopyranoside was used as internal standard as sorbitol proved unsuitable due to its susceptibility to interference from Na/K chloride during silylation. The response factors of glycerol and diglycerol towards phenyl β-D-glucopyranoside were determined using pure standards, while response factors of polyglycerols with a degree of polymerisation of at least 3 could be reliably estimated according to an effective carbon number (ECN) approach. The validity of the method applied to PGPR samples was further supported on the basis of a mass balance considering the experimentally determined polyglycerol and fatty acid content. Moreover, recoveries of di-, tri-, tetra- and pentaglycerol were more than 95% for various PGPR samples added to two different lipid matrices at 2 wt% and 5 wt% concentrations. Furthermore, the method proved to be very repeatable (with relative standard deviation values below 2.2%). On the other hand, the inevitable presence of glycerol in the lipid samples caused fouling of the detector and column overloading, requiring frequent cleaning of the detector and trimming off part of the column.

Keywords:

• PGPR
• polyglycerol
• phenyl β-D-glucopyranoside
• GC-FID

Acknowledgements

We acknowledge the Chinese scholarship council (CSC), as well as the Ugent Special Research Fund (BOF) for funding the doctoral research of Chunxia Su (CSC 201808420300 and BOF22/CDV/025) at Ghent University. We would also like to thank An Maes and Nathalie De Muer for technical assistance.

Disclosure statement

No potential conflict of interest was reported by the author(s).

.

Additional information

Funding

This work was supported by the China Scholarship Council and the Ugent Special Research Fund (BOF) for funding the doctoral research of Chunxia Su (CSC 201808420300 and BOF22/CDV/025) at Ghent University.