Abstract：
This work investigated the feasibility of applying headspace solid phase microextraction-gas chromatography-mass spectrometry (HS-SPME-GC/MS) combining olfactory visualization for flavor characterization of black garlic. Volatile organic compounds (VOCs) analysis was performed to select important differential VOCs during black garlic processing. A multi-channels nanocomposite CSA assembled with two porous metal-organic frameworks was then developed to characterize flavor profiles changes during black garlic processing, and garlic samples during processing could be divided into five clusters, consistent with VOCs analysis. Artificial neural network (ANN) model outperformed other pattern recognition methods in discriminating processing stages. Furthermore, SVR model for odor sensory scores with the correlation coefficient for prediction set of 0.8919 exhibited a better performance than PLS model, indicating a preferable prediction ability for odor quality. This work demonstrated that the nanocomposite CSA combining appropriate chemometrics can offer an effective tool for objectively and rapidly characterizing flavor quality of black garlic or other food matrixes.
摘要：
这项工作研究了应用顶空固相微萃取-气相色谱-质谱（HS-SPME-GC/MS）结合嗅觉可视化表征黑蒜风味的可行性。进行挥发性有机化合物（VOCs）分析以选择黑蒜加工过程中重要的差异VOCs。然后开发了与两个多孔金属有机框架组装的多通道纳米复合材料CSA，以表征黑蒜加工过程中的风味变化，大蒜样品在加工过程中可以分为五组，与VOCs分析一致。人工神经网络（ANN）模型在区分处理阶段优于其他模式识别方法。此外，气味感官评分的SVR模型的预测相关系数为0.8919，表现出比PLS模型更好的性能。表明对气味质量有较好的预测能力。这项工作表明，结合适当的化学计量学的纳米复合材料CSA可以为客观，快速地表征黑蒜或其他食品基质的风味质量提供有效的工具。
