Food Control最新文献

{"title":"The relationship between cinnamon's diverse commercial names and food safety risks","authors":"Zhoujian He ,&nbsp;Meng Ye ,&nbsp;Jie Huan ,&nbsp;Huaxue Wu ,&nbsp;Xiaomei Luo ,&nbsp;Liqiong Jiang ,&nbsp;Xiao Gong ,&nbsp;Li Zheng","doi":"10.1016/j.foodcont.2026.111991","DOIUrl":"10.1016/j.foodcont.2026.111991","url":null,"abstract":"<div><div>As a globally utilized spice, cinnamon finds extensive applications across food, pharmaceutical, and cosmetic industries. The term “Cinnamon” collectively refers to dried inner bark products from multiple <em>Cinnamomum</em> species, with diverse commercial nomenclature arising from varied plant parts and processing methods. However, the potential food safety risks associated with this nomenclature complexity remain understudied. This review systematically examines the botanical origins of commercial cinnamon products and analyzes their associated food safety hazards. Ceylon cinnamon (<em>Cinnamomum verum</em>) represents the premium commercial variety, while cassia cinnamon familycomprising six <em>Cinnamomum</em> species dominates global market share. Chemical variability among products may induce allergic reactions. Lead (Pb) emerges as the predominant heavy metal contaminant, with bark-derived products showing highest contamination levels of heavy metals. Economically motivated adulteration constitutes the primary source of food fraud. We propose a comprehensive safety framework involving standardized labeling protocols, enhanced authentication technologies, and improved traceability systems to mitigate cinnamon-related food safety concerns.</div></div>","PeriodicalId":319,"journal":{"name":"Food Control","volume":"184 ","pages":"Article 111991"},"PeriodicalIF":6.3,"publicationDate":"2026-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146024937","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Amoxicillin in broiler chickens: An analysis of marker residue persistence in edible tissues and food safety implications","authors":"Belén Pinto ,&nbsp;Paula Cortés ,&nbsp;Fernanda Suazo ,&nbsp;Ekaterina Pokrant ,&nbsp;Maria José Navarrete ,&nbsp;Andrés Flores ,&nbsp;Aldo Maddaleno ,&nbsp;Constanza Vergara ,&nbsp;Claudio Nuñez ,&nbsp;Javiera Cornejo","doi":"10.1016/j.foodcont.2026.112006","DOIUrl":"10.1016/j.foodcont.2026.112006","url":null,"abstract":"<div><div>Amoxicillin may pose adverse health effects if residues remain in food products. Therefore, Codex Committee on Residues of Veterinary Drugs in Food has prioritized establishing maximum residue limits (MRLs) for β-lactam in chickens. This study aimed to develop and validate an analytical methodology for amoxicillin detection in broiler chicken, in line with updated international standards, and to assess its persistence in edible tissues. Forty broilers were divided into a treated group (n = 30; 20 mg kg⁻¹ orally for 7 days) and a control group (n = 10). Samples collected on days 1, 2, 5, 9, and 18 post-treatment were analysed by liquid chromatography-tandem mass spectrometry. Methods were validated according to Regulation EU/2021/808, VICH GL49 and GL2, and met all key performance criteria (specificity, limit of detection and quantification, linearity, precision, recovery, CCα, matrix effect, stability, and ruggedness). Amoxicillin residues were undetectable in muscle, liver, and kidney, indicating rapid elimination. However, residues persisted in skin-plus-fat (28.63 ± 14.16 μg kg⁻¹ on day 1–5.39 ± 3.33 μg kg⁻¹ on day 3). Withdrawal periods of 5-days for skin-plus-fat based on European Union's MRLs (50 μg kg⁻¹), and 11-days based on method's limit of quantification were established. These findings confirm the rapid depletion of amoxicillin in edible tissues with major persistence in skin-plus-fat. The results highlight the importance of monitoring amoxicillin in edible tissues of broiler chickens. The study provides scientific data that can be included in the risk analysis developed by JECFA for the evaluation of the establishment of MRLs for amoxicillin in poultry by Codex Alimentarius.</div></div>","PeriodicalId":319,"journal":{"name":"Food Control","volume":"184 ","pages":"Article 112006"},"PeriodicalIF":6.3,"publicationDate":"2026-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146170642","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Fusing cross-source spectral data and component-driven analysis for maize grain quality detection","authors":"Qixing Tang ,&nbsp;Dong Wei ,&nbsp;Longbao Liu ,&nbsp;Juan Liao ,&nbsp;Leizi Jiao ,&nbsp;Yachao Liu ,&nbsp;Ke Wang ,&nbsp;Runyu Wang ,&nbsp;Daming Dong","doi":"10.1016/j.foodcont.2026.112004","DOIUrl":"10.1016/j.foodcont.2026.112004","url":null,"abstract":"<div><div>This study addresses the limitations of traditional chemical methods and the poor generalizability of single near-infrared spectroscopy models by proposing a component-driven approach for universal maize quality assessment. The core of our methodology is a novel workflow in which predicted component contents are used as features for quality classification. First, spectral discrepancies between public and laboratory datasets were mitigated through waveband alignment and selection (1180–1460 nm). Subsequently, after optimizing pre-processing algorithms, we employed models trained on public data to predict the contents of key components (e.g., moisture, fat) in laboratory samples. These predicted component values, rather than the raw spectral data, were then used as inputs to machine learning models for intelligent quality classification. Experimental results showed that the Random Forest (RF) model, utilizing the predicted components and Savitzky-Golay + Baseline Correction (SG + BC) pre-processing, achieved optimal performance with a test set accuracy of 91.88 %. This research provides a novel strategy for non-destructive quality evaluation by fusing cross-source spectral data and multi-component synergy, which significantly enhances the model's generalization capability and practical applicability.</div></div>","PeriodicalId":319,"journal":{"name":"Food Control","volume":"184 ","pages":"Article 112004"},"PeriodicalIF":6.3,"publicationDate":"2026-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146024944","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Evaluating the impact of natural anti-microbials on bacterial communities during storage of vacuum packed beef burgers using culture-dependent and culture-independent next generation sequencing approaches","authors":"Angelos Papadochristopoulos ,&nbsp;Joseph P. Kerry ,&nbsp;Narelle Fegan ,&nbsp;Catherine M. Burgess ,&nbsp;Geraldine Duffy","doi":"10.1016/j.foodcont.2026.112007","DOIUrl":"10.1016/j.foodcont.2026.112007","url":null,"abstract":"<div><div>In this study, the impact of three natural agents (shrimp-derived chitosan, mushroom-derived chitosan, cranberry extract) on the bacterial community composition during storage of vacuum packed chilled beef burgers was assessed using culture-dependent and culture-independent approaches. The three agents were added individually to beef burgers at a concentration of 2.5 %, along with a control with no agent added. Burgers were vacuum packed and stored at 3 ± 1 °C for 16 days. After 16 days of chilled storage, cultural analysis showed that beef burgers with added cranberry extract, had significantly (<em>P</em> &lt; 0.05) lower counts than control samples, for total viable bacteria (1.82 log₁₀ CFU g⁻¹), lactic acid bacteria (1.53 log₁₀ CFU g⁻¹) and Enterobacteriaceae (2.5 log₁₀ CFU g⁻¹) while <em>B. thermosphacta</em> was below the level of detection after 8 days of chilled storage (<em>P</em> &lt; 0.05). However, chitosan from both mushroom and shrimp were ineffective against all spoilage bacteria tested. Total viable counts were higher by 1.15 and 0.44 log₁₀ CFU g⁻¹, respectively, than the control at the end of storage (<em>P</em> &gt; 0.05), with lactic acid bacteria being the dominant species. Total bacterial community analysis using 16S rRNA sequencing showed that <em>Carnobacterium</em> spp. were dominant in burgers with added chitosan from both sources and <em>Lactobacillus</em> spp. were dominant in burgers with cranberry extract. In control samples, <em>Lactobacillus</em> spp., <em>Lactococcus</em> spp. and <em>Carnobacterium</em> spp. were approximately equally abundant. The culture-independent approach provided useful additional insights into the impact of natural agents on the vacuum packed beef burger microbiome.</div></div>","PeriodicalId":319,"journal":{"name":"Food Control","volume":"184 ","pages":"Article 112007"},"PeriodicalIF":6.3,"publicationDate":"2026-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146075235","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"N, S-co-doped carbon quantum dots as a versatile platform for monitoring hydrogen sulfide and nitrite based on Fe2+/Fe3+ redox cycle: Applications in food safety and bioimaging","authors":"Yaping Zhong,&nbsp;Kayi Shao,&nbsp;Ruixia Zhang,&nbsp;Linlong Zhang,&nbsp;Dong Wang","doi":"10.1016/j.foodcont.2026.112012","DOIUrl":"10.1016/j.foodcont.2026.112012","url":null,"abstract":"<div><div>Food safety is a key public concern, making the detection of harmful substances in food increasingly urgent. Herein, a multifunctional fluorescence sensing platform was designed utilizing glutathione-derived N, S-co-doped carbon quantum dots (N, S-CDs), capable of sensitively monitoring nitrite (NO₂⁻) and hydrogen sulfide (H₂S) in both food and living biological cells. The detection relies on Fe³⁺-induced fluorescence quenching of N, S-CDs through inner filter effect and static quenching, combined with the redox reactions of NO₂⁻ with Fe²⁺ and H₂S with Fe³⁺. The linear ranges and limits of detection (LOD) were as follows: NO₂⁻, 1–1000 μM and 0.164 μM; H₂S, 25–425 μM and 0.114 μM. Thanks to its ideal biocompatibility, N, S-CDs enabled visualization of NO₂⁻ and H₂S in living cells. Meanwhile, N, S-CDs paper-based fluorescent sensor was constructed to detect NO₂⁻ in pickled meat and H₂S in spoiled meat, demonstrating great potential for biological tracking and portable sensor development.</div></div>","PeriodicalId":319,"journal":{"name":"Food Control","volume":"184 ","pages":"Article 112012"},"PeriodicalIF":6.3,"publicationDate":"2026-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146075236","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Conditioned raw cut steak in modified atmosphere packaging: Prediction, identification, and quality correlation analysis of specific spoilage organisms during storage","authors":"Yumeng Sui,&nbsp;Ziyi Zhang,&nbsp;Biying Zhang,&nbsp;Qiang Wang,&nbsp;Baohua Kong,&nbsp;Zejian Xu,&nbsp;Qian Chen","doi":"10.1016/j.foodcont.2026.112008","DOIUrl":"10.1016/j.foodcont.2026.112008","url":null,"abstract":"<div><div>In this study, specific spoilage organisms (SSOs) were identified from conditioned raw cut steak (CRC-steak) under modified atmosphere packaging (MAP) with 40 % O₂ and 60 % CO₂ through microbial analysis and multi-index correlation analysis. During storage at 4 °C, bacterial diversity analysis indicated that five bacterial species, including <em>Acinetobacter baumannii</em>, <em>Pseudomonas fragi</em>, and <em>Brochothrix thermosphacta</em>, constituted the core dominant microbiota. Additionally, the total volatile basic nitrogen (TVB-N) value and cooking loss of the CRC-steak significantly increased, while shear force significantly decreased (<em>P</em> &lt; 0.05). In addition, a total of 111 volatile compounds and 8 biogenic amines were identified. <em>B. thermosphacta</em>, <em>P. fragi</em>, and <em>A. baumannii</em> exhibited significant positive correlations with the production of key spoilage-related volatile compounds (e.g., 3-methyl-3-butanol, 1-octen-3-ol, and acetoin) as well as cadaverine and tyramine, which could be defined as SSOs. Inoculating these three strains into sterilized CRC-steak significantly enhanced the total viable count and accelerated the increase in TVB-N and values, thereby demonstrating their substantial capacity to induce spoilage. This research not only establishes a scientific basis for controlling the growth of SSOs in MAP steaks but also provides a theoretical basis for developing targeted antibacterial technologies and enhancing the quality of MAP meat products.</div></div>","PeriodicalId":319,"journal":{"name":"Food Control","volume":"184 ","pages":"Article 112008"},"PeriodicalIF":6.3,"publicationDate":"2026-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146075238","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Presence of Listeria monocytogenes in hydroponic controlled environment agriculture (CEA): Contamination transfer in indoor production","authors":"María Ayala-San Nicolás,&nbsp;Pilar Truchado,&nbsp;Ana Allende,&nbsp;Natalia Hernández,&nbsp;Silvia Andújar,&nbsp;Juan Antonio Tudela,&nbsp;María Isabel Gil","doi":"10.1016/j.foodcont.2026.112023","DOIUrl":"10.1016/j.foodcont.2026.112023","url":null,"abstract":"<div><div>Controlled environment agriculture (CEA) presents unique food safety challenges, requiring a clear understanding of pathogen occurrence, survival, and persistence. This study assessed the prevalence of <em>Listeria monocytogenes</em> in a commercial hydroponic CEA facility using the nutrient film technique (NFT) for pak choi cultivation. Environmental monitoring was conducted at three time points (EM1, EM2 and EM3) over 8 months, with 156 samples collected, focusing on food contact surfaces and non-food contact surfaces, including the crop and water. <em>L. monocytogenes</em> was detected in EM1 on the boot covers used for walking through preharvest, harvest, and postharvest areas, demonstrating the effectiveness of this sampling tool. Another positive detection was found in the drainage nutrient solution in EM1, indicating a potential contamination risk to the crop. However, the cleaning and sanitation strategy implemented after EM1 was effective, as no <em>L. monocytogenes</em> was detected in the subsequent monitoring samplings (EM2 and EM3) despite an increased number of samples. A simulated contamination transfer of <em>L. innocua</em> in a CEA facility, mimicking contamination from a puddle and transfer via worker boots and equipment wheels, showed initial viability (T0); however, after 24 h, viable cells were recovered from only a few floor steps. Our key findings indicate that CEA floor surfaces can serve as vectors of microbial contamination and that recirculated nutrient solutions in hydroponic systems represent potential risks of pathogen spread. This study underscores the importance of comprehensive environmental monitoring and targeted cleaning and sanitizing practices to minimize contamination risks in hydroponic CEA systems.</div></div>","PeriodicalId":319,"journal":{"name":"Food Control","volume":"184 ","pages":"Article 112023"},"PeriodicalIF":6.3,"publicationDate":"2026-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146170691","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A lightweight network for non-destructive and precise prediction of loquat bruising volume based on RGB images","authors":"Changsu Xu,&nbsp;Liuyan Feng,&nbsp;Xiangfei Huang,&nbsp;Lang Chen,&nbsp;Minghui Zhu,&nbsp;Yunwu Li","doi":"10.1016/j.foodcont.2026.112044","DOIUrl":"10.1016/j.foodcont.2026.112044","url":null,"abstract":"<div><div>Bruising is one of the most critical factors contributing to economic losses in horticultural crops, as it can occur during harvesting, transportation, and processing. Recent studies have applied deep learning to bruise detection, but most focus only on surface analysis, while traditional optical imaging can quantify bruise volume but is too costly for large-scale use. To address this issue, this paper proposes Loquat Bruise-You Only Look Once (LB-YOLO), an improved YOLOv9-based model incorporating Expanded Window-Multi-Head Self-Attention (EW-MHSA) and Depthwise Separable Convolution (DW-Conv) to enhance efficiency and precision. LB-YOLO detects bruised regions on loquat surfaces and estimates bruise volume ratio by leveraging a correlation model between surface bruise area ratio and bruise volume ratio, developed through Computed Tomography (CT) analysis. The model processes 870 × 580-pixel RGB images and outputs both bruise surface area ratio and estimated bruise volume ratio. The experimental results show that, compared with other networks based on yolov9 and commonly used semantic segmentation models, LB-YOLO has better comprehensive performance. Its surface bruising segmentation accuracy reaches 98.9%, and the average detection speed is 18.1 ms per image. Meanwhile, the average prediction accuracy rate of the volume ratio of bruising was achieved at 94.31%. Edge-device deployment experiments further confirm its reliable performance in real-world environments. This study provides crucial insights for the efficient, non-destructive quantification of bruising in horticultural crops, which may become a significant research direction in the field of precision fruit grading in the future.</div></div>","PeriodicalId":319,"journal":{"name":"Food Control","volume":"184 ","pages":"Article 112044"},"PeriodicalIF":6.3,"publicationDate":"2026-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146170651","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Correlation study between sesame seeds roasting conditions and the physicochemical characteristics, hazardous substances levels, and volatile substances profile of hot-water-substitution extracted flavor sesame (Sesamum indicum L.) oil","authors":"Li Zhou ,&nbsp;Qingyang Zhang ,&nbsp;Jin Zhang ,&nbsp;Jinglun Zhou ,&nbsp;Xinya Huang ,&nbsp;Zhengting Zhu ,&nbsp;Siya Zhao ,&nbsp;Meiyu Zheng ,&nbsp;Yong Yang ,&nbsp;Shu Wang ,&nbsp;Bin Li ,&nbsp;Sihong Zhang ,&nbsp;Dongping He ,&nbsp;Zhigang Hu ,&nbsp;Fenfen Lei","doi":"10.1016/j.foodcont.2026.111976","DOIUrl":"10.1016/j.foodcont.2026.111976","url":null,"abstract":"<div><div>Sesame oil, esteemed for its robust aroma and high content of unsaturated fatty acids and micronutrients, depends predominantly on seed roasting to develop its characteristic flavor. However, this high-temperature process is associated with the risk of generating hazardous compounds, including benzo [a]pyrene (BaP), 3-monochloropropane-1,2-diol esters (3-MCPDE), and glycidyl esters (GEs), which have raised increasing regulatory concerns worldwide. In this study, an integrated approach was employed, encompassing sample morphology, physicochemical properties, levels of hazardous substances, and flavor profiles, to systematically investigate the influence of roasting conditions (180–260 °C for 10–40 min) on the quality, safety, and flavor of hot-water-substitution-extracted sesame oil. Results revealed that roasting induced significant microstructural degradation in the sesame seeds, darkened the oil color, and lowered the acid value, while the peroxide value was increased. Importantly, the concentrations of BaP, 3-MCPDE, and GEs exhibited a pronounced rise when roasting temperature exceeded 220 °C, establishing this threshold as a critical control point for process safety. The relative contents of polyunsaturated fatty acids, such as linoleic and linolenic acid, were slightly reduced under intensified roasting. In parallel, a concomitant increase in key volatile compounds was confirmed by electronic-nose analysis, with short-chain alkanes, aldehydes, and alcohols identified as the principal volatile substances. It was concluded that roasting conditions must be meticulously optimized to balance flavor development against the formation of processing contaminants and the degradation of nutritional components. These findings provide a crucial theoretical foundation for refining industrial sesame-oil production and for establishing science-based criteria in market quality supervision and regulatory standards.</div></div>","PeriodicalId":319,"journal":{"name":"Food Control","volume":"184 ","pages":"Article 111976"},"PeriodicalIF":6.3,"publicationDate":"2026-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146024940","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Efficient removal of aflatoxin B1 from soybean kernels by a coupled ultrasound–ozone process: A mechanistic study","authors":"Xiangyan Zeng,&nbsp;Shuke Liu,&nbsp;Haitao Yu,&nbsp;Laxiang Yang","doi":"10.1016/j.foodcont.2026.111988","DOIUrl":"10.1016/j.foodcont.2026.111988","url":null,"abstract":"<div><div>Aflatoxin B1 (AFB1) contamination in soybean kernels poses a persistent food safety challenge. This study evaluated the coupled ultrasound–ozone (US/O₃) process for AFB1 removal from soybean kernels. After 10 min of treatment, the US/O₃ process removed 96.2 % of AFB1, far exceeding ozone alone (54.5 %), ultrasound alone (36.9 %), and water washing (18.1 %). Mass balance analysis showed that only 10.4 % of the initial AFB1 remained in the kernels and soaking water after US/O₃ treatment, indicating that the process mainly destroyed AFB1 rather than merely transferring it into the liquid phase. A positive, time-dependent synergy factor quantified the cooperative effect between ultrasound and ozone. Electron spin resonance (ESR) spin-trapping combined with radical scavenging experiments identified hydroxyl radicals (HO^•) as the dominant oxidants. Identified transformation products suggested an initial attack on the terminal furan ring, followed by extensive hydroxylation, ring opening, dealkylation, and fragmentation to low-molecular-weight phenolic products. ECOSAR predictions showed higher LC₅₀ values for most products than for AFB1, indicating reduced acute toxicity. Fourier transform infrared (FTIR) spectra, color parameters, and lipid oxidation indices indicated negligible changes in soybean composition and quality. Therefore, the US/O₃ process offers an efficient, quality-preserving strategy for mitigating AFB1 contamination in soybean kernels.</div></div>","PeriodicalId":319,"journal":{"name":"Food Control","volume":"184 ","pages":"Article 111988"},"PeriodicalIF":6.3,"publicationDate":"2026-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146024941","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}