Interdigitated electrode (IDE) sensor for frying oil degradation assessment during heating

The repeated usage of frying oil had been proven hazardous to human health due to degradation process by thermal reactions that contribute to the formation of polar compounds in the oil, which in turn changes the quality of the oil. The oxidized products and lipid polymers formed in repeatedly-used frying oil would also pose health risks. Many technical approaches are being used to evaluate the degradation of frying oil. These processing include column chromatography to determine the free fatty acids (FFAs), image analyses to measure the total polar compounds (TPC), and Fourier transform infrared spectroscopy (FTIR) to measure the peroxide value (PV). However, these methods are commonly time consuming, laborious, need substantial amounts of solvents, and require skilled operators. Consequently the development of a simple and fast method is essential to assess the degradation of frying oil. In this study, interdigitated electrodes (IDE) sensor was designed to appraise the frying oil degradation at several heating time intervals by measuring the changes on its electrical characteristics. The probe was designed using IDE structure and the photolithography was performed to fabricate pattern for metal and insulation layers of the sensor. Frying oil samples a set of 30 each containing 130 ml of palm oil were heated at 180o C up to 30 hours. For each one hour increment, one sample was taken out of the oven and cooled at room temperature before further analyses. Then the impedance and capacitance measurement were carried out using the IDE sensor connected to a inductance (L), capacitance (C), and resistance (R) meter (4263B, Agilent Technologies, Japan) using Kelvin clip leads (16089E,Agilent, Japan) over five frequencies (100 Hz, 1000 Hz, 10 KHz, 20 KHz, and 100 KHz). The percentage of TPC of each heated sample was measured using a frying oil tester (Testo 270, InstruMartlnc, Germany) whereas the viscosity was tested utilizing a viscometer (Sv-10 Vibro Viscometer, A&D Company Limited, Japan). The experiments were repeated three times and the average of them were taken. The results were analyzed to find the correlation between electrical properties of oil (impedance and capacitance)with oil degradation parameters (TPC and viscosity). The results showed that a significant correlation was found between oil impedance and capacitance with TPC and viscosity. For example, at 100 Hz the correlation between impedance and TPC has coefficient of determination (R2) of 0.92 while the correlation between impedance and viscosity has R2 of 0.85. When the regression equations used to predict TPC and viscosity with impedance measurements were validated using the validation data set, the lowest RMSE of 4.03% and 3.48% respectively was found while the highest RMSE was 4.61%. In addition, the principle component analysis (PCA) model to predict TPC and viscosity using impedance measurements yielded a significant performance with R2of 0.88 and 0.86, respectively. Similarly, the capacitance measurements have significant correlation with TPC and viscosity. For example, at 100 KHz, which is the best correlation given,the correlation between capacitance and TPC has coefficient (R2) of 0.90 while the correlation between capacitance and viscosity has R2 of 0.87. When the regression equations used to predict TPC and viscosity with capacitance measurements were validated using the validation data set, the lowest RMSE of 3.88% and 3.64% respectively was found, where the highest RMSE was 5.04%. In addition, the PCA model to predict TPC and viscosity using capacitance measurements yielded a significant performance with R2 of 0.86 and 0.84, respectively. This study found that the designed IDE sensor has good potential for a simple and inexpensive way of monitoring frying oil degradation. The result from this study could provide the foundation for building a portable sensor to be used in assessing the degradation properties of frying oil.

Preview PDF FK 2015 2RR.pdf Download (1MB) | Preview

Actions (login required)

View Item