A Feasibility Study of the Rapid Evaluation of Oil Oxidation Using Synchronous Fluorescence Spectroscopy
- PDF / 1,854,059 Bytes
- 7 Pages / 595.276 x 790.866 pts Page_size
- 71 Downloads / 219 Views
A Feasibility Study of the Rapid Evaluation of Oil Oxidation Using Synchronous Fluorescence Spectroscopy Haiyang Gu 1
&
Yanhui Sun 1 & Shulan Liu 1 & Shuangfang Li 1 & Weiwei Zhang 1
Received: 23 April 2018 / Accepted: 22 June 2018 # Springer Science+Business Media, LLC, part of Springer Nature 2018
Abstract Synchronous fluorescence spectroscopy coupled with nonlinear regression analysis was developed and used to discriminate the degree of oil oxidation based on the acid and peroxide values. It is easy to discriminate the degree of oil oxidation after treatments with different heat treatment times and styles for both synchronous fluorescence spectroscopy and low-field nuclear magnetic resonance (NMR). The low-field NMR results proved that oil oxidation probably proceeds via more than one pathway for different treatment styles. To decompose the three-dimensional matrix into two-dimensional data, a parallel factor analysis (PARAFAC) algorithm was used to select an optimized Δλ of 120 nm. Then, an artificial neural network (ANN) was used to build a regression model for both synchronous fluorescence and low-field NMR to evaluate the degree of oil oxidation. Compared with other regression results, synchronous fluorescence coupled with the ANN model achieved the best results, with 1.00 in the training set and 1.00 in the acid value test set at room temperature. The overall results suggest that synchronous fluorescence spectroscopy coupled with the ANN regression algorithm is useful for rapidly evaluating oil quality and perhaps foods with high oil contents. Keywords Synchronous fluorescence spectroscopy . Oil oxidation . Parallel factor analysis . Artificial neural network
Introduction Oxidation stability is one of the basic quality attributes for oil storage, processing, marketing, and consumption. It plays an important role in indicating the degree of oil quality for both industry and consumers. Traditionally, oxidation evaluation has been based on sensory authentication by a professionally trained person, including evaluating the appearance of color attributes and testing the odor and taste sensory attributes (Aparicio et al. 2015). This provides a convenient and immediate assessment of the oil quality and can be performed without any chemical reagents and instruments. This method requires a well-trained and experienced assessor to obtain reliable results (Ihsan et al. 2016). Furthermore, even for different professional assessors, it is difficult to obtain repeatable and quantitative results when using human testers. To minimize these shortcomings, instrument-based methods are considered
* Haiyang Gu [email protected] 1
School of Bio & Food Engineering, Chuzhou University, Chuzhou 239000, China
objective and independent measurement methods that can meet the evaluation requirements for the oil industry. A series of instrumental methods have been developed to evaluate the state of oil oxidation, such as determining the acid (Cao et al. 2014) and peroxide values (Anwar et al. 2014) related to the oxidation stat
Data Loading...
