Analytical Methods Rapid measuring and modelling flavour quality changes of oxidised chicken fat by electronic nose profiles through the partial least squares regression analysis Shiqing Song a , Lin Yuan b , Xiaoming Zhang b,⇑ , Khizar Hayat c , Huangnv Chen b , Fang Liu b , Zuobing Xiao a , Yunwei Niu a a Department of Biology and Food Engineering, Shanghai Institute of Technology, 200235 Shanghai, PR China b State Key Laboratory of Food Science and Technology, School of Food Science and Technology, Jiangnan University, Wuxi 214122, Jiangsu, PR China c Department of Chemistry, COMSATS Institute of Information Technology, Abbottabad 22060, Pakistan article info Article history: Received 11 December 2011 Received in revised form 11 April 2013 Accepted 2 July 2013 Available online 10 July 2013 Keywords: Electronic nose Partial least squares regression analysis Chemical parameters Chicken fat Controlled oxidation Gas chromatography–mass spectrometry abstract The objective of this study was to investigate whether an electronic nose, comprising 18 metal oxide semiconductor gas sensors, could be used for measuring and modelling flavour quality changes of refined chicken fat during controlled oxidation. Partial least squares regression (PLSR) was applied to determine the predictive relationships between the chemical parameters, GC–MS data, free fatty acid profiles and electronic nose responses for controlled oxidation of refined chicken fat. The results showed that perox- ide value (PV) and acid value (AV) were significantly well predicted by the electronic nose responses, whereas p-anisidine value (p-AV) was found to be fairly well predicted especially for deeply oxidised chicken fat. Thus, this study gave evidence of the electronic nose system to be a promising device for future at- or on-line implementation in oxidation control of chicken fat for producing meat flavourings. Ó 2013 Elsevier Ltd. All rights reserved. 1. Introduction A characteristic chicken odour is one of the most important parameters to determine the quality of chicken analogue products such as processed meat flavourings based on hydrolyzed vegetable protein (HVP) through the Maillard reaction. Recently, researchers (Kerscher, 2000; Kerscher & Grosch, 1998) confirmed that the spe- cies-specific aroma in cooked meats like chicken and beef is mainly due to concentration and composition differences in lipid-derived flavour substances. Besides the compounds formed from lipid oxi- dation directly contributing to species-aroma properties, they can also interact with products of the Maillard reaction to give further flavour compounds. Unfortunately, lipid oxidation is responsible for rancidity in fats, oils and lipid-containing foods when degrada- tion product concentrations accumulate above the critical level. Controlled oxidation of lipids to a certain degree has become one of the key technologies for meat flavourings production. Excessive oxidation will lead to an undesirable flavour and if oxidation is inadequate, the special characteristic meat flavour may not devel- op. Therefore, in order to produce desirable characteristic chicken flavour, it is necessary to understand how to achieve and monitor moderate oxidation of chicken fat without causing rancidity. Formation of volatile flavour compounds from lipids is associ- ated with free radical-initiated oxidation triggered by heat in our experiments. Traditionally, the classical chemical methods moni- toring lipid oxidation are the determination of peroxide value (PV), p-anisidine value (p-AV) and acid value (AV). PV is the mea- surement of hydroperoxides, which are the primary oxidation products of lipid oxidation; p-AV and AV are the measures of car- bonyl compounds and hydrolytic rancidity, respectively (Farhoosh & Pazhouhanmehr, 2009; Naz, Sheikh, Siddiqi, & Sayeed, 2004; Rahman, Al-Belushi, Guizani, Al-Saidi, & Soussi, 2009). Conventional gas chromatography-mass spectrometry (GC–MS) is another basic technique used to assess volatile compounds. However, just like chemical measurement, it is generally too time-consuming, complex and labour-intensive for routine quality applications. One promising technology in at- or on-line measure- ment of flavour relevant quality criteria are the electronic nose sys- tems (E-nose). E-nose instruments contain an array of chemical gas sensors that respond differently in the presence of volatiles is de- signed to mimic the human olfactory system. Its use has been pro- posed to direct the needs for routine quality testing in the food industry. In recent years, a number of articles have been published 0308-8146/$ - see front matter Ó 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.foodchem.2013.07.009 ⇑ Corresponding author. Tel.: +86 510 85919106; fax: +86 510 85884496. E-mail address: xmzhang@jiangnan.edu.cn (X. Zhang). Food Chemistry 141 (2013) 4278–4288 Contents lists available at SciVerse ScienceDirect Food Chemistry journal homepage: www.elsevier.com/locate/foodchem