A colorimetric sensing system for visualization of aroma-contributing volatile organic compounds in food and beverage products

• Main Author: Liu, Tianyi
• Other Authors: Chen Xiaodong
• Format: Thesis-Doctor of Philosophy
• Language: English
• Published: Nanyang Technological University 2025
• Subjects: Engineering
• Online Access: https://hdl.handle.net/10356/182445
• Institution: Nanyang Technological University

Odor is an important property for food and beverage products. Pleasant odor is referred to as aroma. In the food industry, aroma is frequently used to assess the quality of products. Aroma analysis can be performed by humans or scientific instruments. Aroma is analyzed by humans in a synthetic way, as all electrochemical signals produced by activated olfactory receptors are combined and interpreted as a whole by human brains. The results generated by human sensory evaluators are believed to be a better prediction of consumers’ reaction. However, sensory analysis suffers from drawbacks like subjectiveness and long training time. In the laboratory, the aroma of food and beverage products is assessed by analyzing the volatile organic compounds (VOCs) present in these products. Gas chromatography-mass spectroscopy (GC-MS) is the most established laboratory-based method for VOC analysis. Unlike humans, GC-based methods are analytical, as each odor is broken down into constituent molecules. Despite its wide use, GC-based methods are facing limitations, such as sophistication, low efficiency, high cost, and lack of portability. Over decades, alternative solutions have been proposed by researchers, with electronic noses (E-noses) being the most studied. E-noses are gas sensors which utilize the electrical signals generated by the interaction between sensing elements and target VOCs. However, such interactions are weak and less specific, and hence, could compromise the sensor’s selectivity. On the contrary, colorimetric sensors relying on specific chemical reactions could provide more useful information on the VOC profile of a sample. Hence, this project aimed to develop a colorimetric sensing system for food and beverage aroma analysis. The key component of the sensing system was a multi-unit paper-based colorimetric sensor; the colorimetric indicators were carefully selected to chemically react with common VOCs in food and beverage products. The sensing system also included a sensing device and a comprehensive methodology. The results generated by the system were in the forms of color VOC fingerprints; color changes were quantified as color differences and Euclidean distances, and further analyzed by principal component analysis (PCA) and partial least squares (PLS) regression analysis. The sensing system’s performance was evaluated with both chemical standards and actual food samples. Systematic studies on three types of food and beverage products were conducted: fragrant sesame oil, roasted coffee beans and fruits. In addition to simple classification, authentication and freshness monitoring, the sensing system was applied to visualize the differences in aroma caused by different processing techniques. Experimentally, it was found that the sensor could respond to common VOCs in food and beverage products, including acids, alcohols, aldehydes, ketones and amines; the indicators could undergo naked-eye observable color changes in 30 minutes. The system could be operated at room temperature, and the samples could be analyzed as received, eliminating the need for tedious sample preparation. The colorimetric sensor combined the synthetic and analytical properties of human olfaction and GC-based methods. It could summarize aroma information as a single pattern, and at the same time, information such as types of VOCs could be inferred from the pattern based on the chemical properties of the indicators. The proposed colorimetric sensing system had demonstrated the potential as a rapid, simple and cost-effective alternative aroma analysis tool in the food industry.