食品质量安全检测的新型味觉传感器机理研究

Food quality and safety are of close relationship with human health and life security. A novel method will be developed for detection on taste information of food by taking metalloporphrin as colorimetric sensing material. Research will be focused on the sensing mechanism of this new detection method. The trail of reaction procedure between metalloporphrin and liquid phase materials being detected will be followed using density functional theory at B3LYP/LANL2DZ level, the influences on the reaction resulted by different cation atom and meso-substituent will be investigated to reveal the inherent sensing mechanism of reaction. And therefore the result will provide a theoretical guidance for selection of sensing material and a scientific foundation for design of sensor array. Some advanced data analysis means, say, independent component analysis, genetic algorithm, and support vector machine, will be brought into the establishment of nonlinear response models with better simplicity, precision and robustness. Thus the novel method for detection of taste information of food will have been successfully developed. Fish and its products will be selected as experiment material, and main risk factors to the quality and safety of fish and its products, such as microbial contamination, heavy metals residues and fish drugs residues will be taken as inspection index, experiments will be taken to prove the performance of the novel detection method. It is expected to provide a new choice for rapid evaluation of food safety and to be advantageous to the improvement of the theoretical and technical level of detection technology on food quality and safety, and therefor will play an important role in meeting the challenge in food quality and safety in our country.

食品质量安全问题关系到每个人的健康与生命安全。本项目拟开发用于食品质量安全检测的味觉信息传感新方法，重点研究以卟啉配合物作为色敏传感材料的新型味觉传感器机理。采用密度泛函理论，在B3LYP/LANL2DZ水平上追踪金属卟啉与液相物质的相互作用过程，探究金属卟啉不同金属原子和中位取代基的差异对于其与液相物质反应的影响，正确揭示其反应机理。建立理论指导下的传感器筛选和传感器阵列构建的科学方法。在非线性模型的建立中引入独立分量分析、遗传算法及支持向量机等先进手段，提高模型的简约性、精度和鲁棒性。由此建立食品质量安全检测的味觉传感新方法。选取鱼类食品为基本试验对象，以影响鱼类食用安全性的主要因素：微生物污染程度、重金属残留及渔药残留作为检测指标，进行试验验证。以期为食品食用安全性的快速评价提供新途径，提高食品质量安全检测的理论水平和技术水平，为应对我国食品质量安全面临的巨大挑战起到积极作用。

食品质量安全问题关系到每个人的健康与生命安全，提高食品检测技术水平和检测能力是提高食品质量安全水平的有效途径之一。项目重点研究开发基于味觉信息可视化传感的食品质量安全检测新方法。采用密度泛函理论对色敏传感材料金属卟啉及其衍生物的分子体系模型进行理论推导，基于Becke三参数杂化泛函(B3LYP)法在LANL2DZ基组水平上研究金属卟啉及其衍生物分子体系模型相关的分子结构、原子电荷、分子体系能量变化等理化性质，追踪其与液相物质的相互作用过程，揭示其反应机理。以结合能、偶极矩、前线分子轨道等的理论分析计算结果指导味觉色敏传感器的筛选和传感器阵列的构建。引入多种先进数学方法建立非线性识别模型以提高模型性能。由此建立食品质量安全检测的味觉传感新方法。选取鱼类食品为基本试验对象，针对影响鱼类食用安全性的主要因素分别建立了对鱼类微生物污染程度、重金属残留及渔药残留的多种非线性识别模型。以其中鱼体内渔药残留检测模型为例，对鲫鱼样本体内恩诺沙星残留进行实测时，所构建的支持向量机模型对训练集和测试集的预测值和实测值之间的相关系数分别为0.865和0.778，预测均方根误差分别为16.61 μg/kg和18.61 μg/kg。所构建的遗传算法人工神经网络预测模型，预测值和实测值之间的相关系数对于训练集和测试集分别为0.933和0.830，预测均方根误差分别为8.5μg/kg和12.98μg/kg。项目采用所建立的食品质量安全检测的新型味觉信息色敏传感系统和检测装置，以牛肉、果蔬、花生、茶叶和传统酿造食品等多类食品为对象进行试验验证，结果显现了模型的高鲁棒性和方法的泛适性。表明项目所建立的方法可广泛用于对各种食品农产品有关味觉方面的质量安全指标的检测和评定，可望为实现对食品农产品质量和安全信息的一体化检测和监控提供新途径。