On-line assessment of oil quality during deep frying using an electronic nose and proton transfer reaction mass spectrometry

Abstract We describe a novel method for the quality assessment of oil utilized for deep frying. The method is based on the analysis of frying fumes using a custom electronic nose. The quality score could be obtained after less than 3 min of analysis and without interrupting the frying process or sampling the oil directly. The obtained results were correlated with the peroxide value using a multivariate linear regression model. The most relevant variables were selected based on heat maps and on the analysis of variance. The coefficients of determination for palm oil and rapeseed oil were 0.920 and 0.935, respectively. Real-time monitoring of selected VOCs in the frying fumes was performed using proton transfer reaction mass spectrometry. It was shown that 1-heptanol is a possible indicator of rapeseed oil's quality. The proposed method might be used to supplement the currently available techniques and could find implementation in the non-invasive on-line assessment of frying oils' quality during deep frying of foodstuffs.