Jiang Xingyi, Beverly Robert, Bingi Aravind Kumar, Rao Qinchun, Kidd Jeremiah, Swajian Karen, Cluster Jane, Jackson Lauren
US Food and Drug Administration, Human Foods Program, Office of Applied Microbiology & Technology, Division of Food Processing Science & Technology, 6502 South Archer Road, Bedford Park, IL 60501, USA.
Department of Health, Nutrition, and Food Sciences, Florida State University, 120 Convocation Way, Tallahassee, FL 32306, USA.
J Food Prot. 2025 Sep 10;88(11):100618. doi: 10.1016/j.jfp.2025.100618.
Food manufacturers commonly use filtration with or without filter aids to extend the shelf life of frying oil by removing undesirable substances that form during frying. However, limited research has evaluated the effectiveness of these methods in removing proteins that transfer into frying oil from fried foods. The objective of this study was to assess the efficiency of different filtration methods for removing gluten residues from frying oil. For passive filtration, oil spiked with isolated gluten at a level of 1,000 µg gluten/g oil and oil used for frying 10 batches of breaded shrimp (75 g/batch) were filtered under gravity through metal sieves (pore sizes: 25 µm-2 mm) and cellulose filter papers (pore sizes: 11-25 µm). For active filtration condition selection, gluten-spiked oil (5,000 µg gluten/g oil) was heated (180 °C for 3 min), cooled to 105 °C, and then treated with filter aids - diatomaceous earth (Celite 535) and perlite (Harborlite 900) - at concentrations ranging from 0.05 to 1%, with mixing times of 5-30 min at 1,000 rpm. Subsequently, the oil was filtered through metal sieves (pore sizes: 0.15, 0.6, and 2 mm). Response surface methodology was employed to identify optimal conditions based on their ability to minimize residual gluten. The optimal condition (0.5% filter aid concentration, 30 min mixing time, and 0.15 mm pore size) was then applied to a total of six filter aids. Under these conditions, passive filtration resulted in over 80% reduction in gluten content of oil while the filtration efficiency of active filtration methods was greater than 99.7%. Removal efficiency varied among the six filter aids, with magnesium silicate and diatomaceous earth-based filter aids demonstrating the best performance. Overall, this study highlighted both passive and active filtration as a promising strategy to significantly minimize gluten cross-contact risks associated with reused frying oil, providing practical guidance for gluten management during frying operations.
