Food Analytical Methods最新文献

Gallic acid (GA) is widely utilized in food, pharmaceutical, and chemical industries. However, its excessive consumption may lead to adverse effects such as gastrointestinal irritation and neurological complications. In this work, a cost-effective room-temperature fabrication approach is presented for the synthesis of magnetic iron oxide (Fe₃O₄) nanoparticles with graphene nanoplatelets (GNPs). The fabricated GNP@Fe₃O₄ hybrid nanocomposite was successfully characterized using various analytical techniques such as Fourier Transform Infra-Red (FTIR), X-ray diffraction (XRD), scanning electron microscopy (SEM), and energy dispersive X-ray (EDX) to confirm its crystalline nature, phase purity, functional group information, surface morphology, and elemental composition. The graphene nanoplatelet and magnetic iron oxide-modified screen-printed carbon electrode (GNP@Fe₃O₄/SPCE) was used for the ultrasensitive electrochemical monitoring of GA in food samples. The fabricated GNP@Fe₃O₄/SPCE sensor demonstrated an excellent electrochemical response toward GA, achieving a low limit detection of limit (LOD) of 0.0009 μM over a wide linear concentration range of 0.08–80 μM. Moreover, low relative standard deviation values from selectivity, stability, reproducibility, and repeatability studies confirm the sensor's high reliability and analytical performance for GA detection. The fabricated GNP@Fe₃O₄/SPCE sensor also exhibited excellent sensing performance, enabling accurate determination of GA in different synthetic and naturally occurring food samples with recovery values ranging from 96.8% to 102.6%. Therefore, the GNP@Fe₃O₄ nanocomposite modified electrode provides a promising platform for the development of next-generation electrochemical sensors for the determination of GA in real food samples.

Graphical Abstract

The national food safety standards for physical and chemical testing methods serve as a fundamental tool for food safety risk assessment and management in China. Since the enactment of the Food Safety Law in 2009, these standards have undergone rigorous consolidation and follow-up evaluation. This study systematically analyzed the development, validation, and refinement processes of these standards over the past decade. The findings demonstrate that the standardization model has yielded a scientifically rigorous and highly operable standard system, supported by a feedback-driven continuous improvement mechanism. This systematic approach, with its emphasis on analytical performance validation and practical operability, offers a valuable reference for establishing or refining similar testing standards in other regions, particularly in terms of method standardization and routine laboratory application. Furthermore, prospective insights into the future development and emerging trends of these testing method standards are discussed, offering a theoretical reference for their continuous optimization and technological advancement. 

The significant bioactivities of essential oils (EOs) containing antimicrobial, antioxidant, and therapeutic properties underpin their utility from food and agriculture to healthcare; nevertheless, conventional extraction of thermally labile EOs often compromises yield and compositional integrity. Herein, a microwave-mediated pretreatment coupled with hydrodistillation extraction (MME) tactic meticulously optimized via response surface methodology (RSM) employing a Box-Behnken design (BBD) is proposed for the efficient recovery of a high-Zingiberene EO from ginger rhizome. Under the optimal extraction conditions with 16 g/L of sodium chloride concentration (A), 720 W of microwave power (B), 24 mL/g of liquid–solid ratio (C), 92 s of microwave irradiation time (D), and 4 h of distillation time, this modified MME protocol achieves a substantial yield of 2.25%, representing the 52% enhancement over 1.48% of conventional hydrodistillation extraction (HDE). The developed quadratic model demonstrates remarkable robustness, as evidenced by a high coefficient of determination (R² = 0.9873), a statistically insignificant lack-of-fit (p = 0.3653 > 0.05), and a highly significant model term (p < 0.0001), thereby unequivocally validating its predictive capacity and reliability. Comprehensive gas chromatography-mass spectrometry (GC–MS) analysis reveals a markedly superior chemical profile of 59 constituents in the MME EOs. Notably, the concentration (25.64%) of pivotal bioactive α-Zingiberene is significantly higher than that in the HDE EOs (22.67%), directly contributing to potent biological efficacy. Moreover, such MME EOs have demonstrated potent antioxidant activity, as evidenced by lower half-inhibitory concentration (IC₅₀) values for both 2,2-diphenyl-1-picrylhydrazyl (DPPH) (3.92 mg/mL vs. 9.95 mg/mL for HDE EOs) and 2,2′-azino-bis(3-ethylbenzothiazoline-6-sulfonic acid) (ABTS) (2.04 mg/mL vs. 3.22 mg/mL for HDE EOs) assays. Additionally, the MME EOs exhibit the remarkably broad-spectrum antimicrobial activity with minimal inhibitory concentration (MIC) values of 8.50 mg/mL against Staphylococcus aureus and 0.53 mg/mL against Bacillus subtilis. These findings unequivocally position MME as a superior and non-thermal route for producing the high-value ginger EOs owning the significant potential for applications in pharmaceuticals and functional nutraceuticals.

Graphical Abstract

Virgin olive oil exhibits optical properties that are influenced by its overall physicochemical composition, which in turn is affected by geographical origin and environmental growing conditions. In this study, a multiparameter optical framework is proposed for the nondestructive characterization of virgin olive oils collected from three geographically distinct regions of Jordan. The framework integrates UV–visible absorption spectroscopy, refractive index dispersion, Faraday rotation, and nonlinear optical z-scan measurements to provide complementary information on the linear, magneto-optical, and photothermal behavior of the oils. Refractive index measurements demonstrated normal dispersion that was accurately described by the Cauchy equation, while UV–visible spectroscopy revealed differences in pigment-related absorption among the investigated samples. Faraday rotation measurements showed measurable variations in the Verdet constant, indicating differences in magneto-optical response, and nonlinear z-scan analysis demonstrated distinct thermally induced refractive index changes that were successfully modeled using the thermal lens model. Collectively, these independent optical measurements produced a multidimensional optical fingerprint capable of differentiating the investigated olive oils according to geographical origin. The proposed methodology is rapid, is nondestructive, and requires minimal sample preparation, making it a promising complementary analytical approach for virgin olive oil characterization and future studies of geographical origin, quality assessment, and authenticity.

Aflatoxins are highly toxic secondary metabolites produced by Aspergillus species. These toxins pose significant safety concerns in herbal products due to carcinogenic hepatotoxic immunosuppressive and mutagenic effects. Their presences in herbal products poses a significant safety concern. This review specifically addresses method development, validation stategies, comparative analytical performance, and emerging detection technologies for aflotoxin determaintion in herbal matrices, while alos integrating recent advances in biosensor based detection, green sample prepartion approaches, and regulatory validation framworks relevant to hearbal products. Sample preparation techniques such as QuEChERS, solid-phase extraction (SPE) and immunoaffinity column (IAC) remain established methods for extracting these toxins from complex herbal and food matrices with recovery rates of 80–110%. TheQuEChERS method is the fastest (10 –30 min) and requires the lowest solvent volume. Among the currently available analytical techniques, LC–MS/MS provides the highest analytical sensitivity and selectivity for aflatoxin determination, with limits of detection (LODs) as low as 0.0005–0.005 µg/kg, and enables simultanious multi-mycotoxin analysis. HPLC-FLD and UHPLC-FLD offers comparatively higer LODs (0.01–0.05 µg/kg and 0.005–0.02 µg/kg, respectively). HPLC-FLD additiaonally requires prior sample derivatization and does not support simultanious multi-analyte analysis. ELISA and biosensor based assays permit rapid screening, with LODs in the range 0.02–0.1 µg/kg. A comparative evalution of sensitivity, accuracy, analysis time, and applicability indicates that LC–MS/MS is the most reliable and robust method for confirmatory analysis. UHPLC-FLD and HPLC-FLD remain suitable options for routine quality control because they provide acceptable sensitivity, reproducibility, and lower operating costs than LC–MS/MS. Accordingly, a tiered analytical approach that combines quick screening with high-sensitivity confirmatory methods represent the optimal approach for ensuring herbal product safety in compliance with regulatory framework established by the various regulatory authorities.

Graphical Abstract

The rising demand for high-quality edible oils, together with the increasing worldwide focus on green technologies and environmentally benign solvents, has intensified the search for sustainable extraction strategies that can maintain both oil quality and recovery. In this study, a novel green extraction approach integrating ultrasound-assisted extraction (UAE) with ethyl acetate was developed for the first time for peanut oil extraction. The present approach differs from most earlier studies by using ultrasound as the principal extraction technique, in conjunction with ethyl acetate as a greener alternative to conventional solvents, followed by comprehensive assessment of the peanut oil. The impacts of extraction temperature (30–60 °C), extraction time (30–90 min), and solvent-to-solid ratio (3–9 mL/g) on oil recovery and key quality attributes including antioxidant activity (DPPH and ABTS), total phenolic content (TPC), total flavonoid content (TFC), free fatty acid content (FFA), and peroxide value (PV) were systematically investigated using a Box–Behnken design (BBD). Optimization was applied to simultaneously maximize oil recovery and TPC, ensuring high extraction efficiency alongside effective retention of bioactive compounds. Under the optimized conditions of 58.46 °C, 30 min, and a 9:1 solvent-to-solid ratio, a peanut oil recovery of 93.48% and a TPC of 714 mg GAE/kg peanut oil were achieved. Extraction temperature and solvent-to-solid ratio were identified as the most powerful factors, exerting a statistically significant positive effect on oil recovery (p < 0.0001). Conversely, increasing temperature had a significant negative impact on DPPH and ABTS as well as TPC of peanut oil, highlighting the critical need for careful process optimization. Overall, replacing hexane with ethyl acetate and integrating UEA demonstrated the feasibility of achieving high-quality oil with high recovery through a green extraction strategy, offering a promising and valuable guidance for the edible oil industry.

Graphical Abstract

Tea sensory profiling using semi-trained consumer-based panels facilitates rapid quality evaluation and identification of sensory drivers. However, few studies have integrated consumer-centric sensory evaluation with biochemical composition to elucidate their underlying interrelation. In this study, thirty tea blends were formulated from four major varieties of Assam tea by mixing them in various proportions. The major biochemical constituents, sensory score, and sensory attributes influencing sample liking across distinct panel segments were estimated using the Semi-Trained Check-All-That-Apply (ST-CATA) method. Reliability assessment using Interclass Correlation Coefficient (ICC) demonstrated excellent reliability of using aggregated panel scores. Correlation analysis showed that soluble protein, soluble sugar, protein to polyphenol ratio, oxalic acid, and ascorbic acid content have significant positive correlation with overall sensory score, while polyphenol, (+)-catechin, and pH exhibited significant negative associations. PLSR explained 61.65% of sample sensory variance. Multiple Factor Analysis (MFA) described the similarity across CATA, biochemical, and sensory score datasets in explaining samples” quality configuration. Correspondence Analysis (CA) and preference mapping showed that “sweet,” “nutty,” “floral,” “smooth,” “brown,” “roasty,” “umami,” and “fresh” flavor were preferred, while “grassy,” “bitterness,” “astringency,” and “dark brown” flavor were not preferable attributes. Our results demonstrated that integrating the ST-CATA method with biochemical analysis provides a rapid framework for identifying key sensory drivers and biochemical constituents for targeted formulation of tea blends.

Early gender identification of ducklings is critical for standardized production and quality control in the meat duck supply chain. Traditional manual methods are inefficient and unsuitable for large-scale automated production. This study utilized near-infrared spectroscopy (NIRS) to capture biochemical differences between male and female ducklings, aiming to develop a stable, lightweight nondestructive detection model for embedded edge devices. Spectral data were collected from 600 duckling samples. To achieve optimal data dimensionality reduction, a multistrategy wavelength selection framework integrating a deep autoencoder, ShuffleNet blocks, and a network gradient attribution backward selection strategy was established. Using a baseline 1D-CNN as a uniform evaluation framework for fivefold cross-validation, the 29 key characteristic wavelengths reversely screened by the 1D-VGG16 architecture were determined as the optimal feature input subset. Subsequently, based on these 29 fixed wavelengths, a dedicated lightweight classification network was constructed using 1D depthwise separable convolution and inverted residual structures to address edge computational constraints. Three 1D attention mechanisms including simple parameter-free attention (SimAM), coordinate attention (CA), and efficient multiscale attention (EMA) were horizontally evaluated. Consequently, the SimAM-lightweight 1D-CNN model was proposed, yielding the optimum comprehensive performance. Experimental results revealed that the lightweight model, with only 61,186 trainable parameters, achieved a mean test accuracy of 99.17% ± 0.75% under fivefold cross-validation and an average inference latency of 21.21 ms. Finally, the trained model was deployed on a Jetson Nano–based intelligent prototype for hardware–software integration and system verification, providing a technical foundation for industrial sorting applications.

This study establishes a green, optimized ultrasound-assisted extraction (UAE) process for valorizing orange peel (Citrus sinensis) waste into a crude flavonoid rich syrup and evaluates its application in functional yogurt production. Probe-type UAE (38 kHz, 100 W) using a hydroethanolic solvent (80:20 EtOH:H₂O) and an intermediate particle size (600 µm–1 mm) achieved a significantly superior mass recovery of 54.4%, outperforming conventional Soxhlet extraction (maximum 30.8%). HPLC characterization confirmed that the crude syrup is rich in flavanones, with crystallized hesperidin reaching 94.4% purity, alongside bitter compounds such as neohesperidin. Fortification of yogurt with the crude syrup at 10% (v/v) was sensorially optimal, providing a distinct but balanced citrus aroma and bitterness, while higher levels (15–20%) led to strong bitterness and consumer rejection. Physicochemical analysis revealed a dilution effect (dry matter decreasing from 9.93% to 8.89%; protein from 4.99% to 3.26%) and a paradoxical increase in both titratable acidity (0.897% to 1.225% lactic acid) and pH (4.18 to 4.22), attributed to the extract's buffering capacity. Importantly, syneresis increased dose-dependently due to polyphenol-casein interactions, while all formulations remained microbiologically safe (no pathogens detected). This work provides a scalable framework for transforming citrus by-products into functional food ingredients, highlighting the critical trade-off between bioactive enrichment and sensory/textural quality.

Chlorate is a strong oxidizing agent with known toxicological effects, but its presence in food has only recently emerged as a safety concern. Agar, widely used in food, pharmaceuticals, and microbiological media, has not been systematically examined for chlorate contamination, and no standardized analytical method exists. This study developed and validated an ion chromatography (IC) method for accurate determination of chlorate in agar and assessed soaking as a mitigation strategy. Chlorate was well separated from coexisting compounds, showing excellent linearity (R² = 0.9992) over 0.078–5.0 mg/L. Optimal extraction was achieved using 65% acetonitrile (1:6 solid–liquid ratio, three repetitions). The method showed low detection and quantification limits (5.3 and 23.8 µg/L), high recoveries (97.0–105.9%), and good precision (RSD 0.57–2.54%). Commercial agar contained 1.77–2.30 mg/kg chlorate, exceeding recommended limits. Soaking for 20–30 min reduced chlorate through diffusion-controlled leaching, and three 30 min cycles achieved 62.5% removal, lowering residues to 0.63 mg/kg. The validated IC method and optimized soaking provide analytical reliability and practical control, underscoring chlorate contamination in agar as an emerging food safety issue.