Disposable paper-based sensor array for detection of orange juice adulteration

• Published in: Journal of food measurement & characterization Vol. 18; no. 7; pp. 5779 - 5790
• Main Authors: Mondal, Debasmita, Equbal, Md. Zeeshan, Jain, Akshat, Mishra, Vaibhav
• Format: Journal Article
• Language: English
• Published: New York Springer US 01.07.2024; Springer Nature B.V
• Subjects: Addition polymerization; adulterated products; Algorithms; beverage quality; Chemistry; Chemistry and Materials Science; Chemistry/Food Science; chlorides; Cluster analysis; Clustering; Conducting polymers; Dilution; Discriminant analysis; electronic tongue; Electronic tongues; Embedding; Engineering; Filter paper; Food Science; Frequency analysis; Frequency dependence; Frequency ranges; Frequency response; Fruit juices; Fruits; Impedance; Juices; orange juice; Original Paper; polyaniline; Polyanilines; Polymer coatings; Polymers; Polypyrroles; principal component analysis; Principal components analysis; pyrroles; Sensor arrays; Sensors; silver; Statistical analysis; Substrates; Sucrose; Vector quantization; Filter paper; Statistical analysis; Impedance spectroscopy; Conducting polymer; Orange juice adulteration
• ISSN: 2193-4126; 2193-4134
• DOI: 10.1007/s11694-024-02608-5

An array of filter paper based disposable sensors was developed for identification of orange juice adulteration. Small strips of Whatman filter papers (5 mm × 6 mm) were coated with conducting polymers i.e., polyaniline and polypyrrole (both in chloride doped and de-doped form) to develop the sensors. Silver electrodes were deposited on the polymer-coated sensor substrate using conducting silver paste. The sensors were exposed to various adulterations of orange juice with water dilution and sucrose addition. Impedance measurements from the sensor were monitored over a frequency range of 0.01–10 kHz using a Frequency Response Analyzer. The adulteration of fresh orange juice using water dilution could be detected in the range of 1–30% by volume (in steps of ~ 5–10%) with minimum detection of 1% whereas sucrose (30% w/w) addition was successfully identified from 10 to 60% by volume (in steps of ~ 10%) with minimum detection of 10%. Owing to the non-specific response of the sensors, statistical analysis using Principal Component Analysis, Linear Discriminant Analysis, t-Distributed Stochastic Neighbor Embedding, and k-Means clustering algorithm were utilized to identify the trend in the impedance response shift for various levels of adulterations in orange juice. Among the different classification methods, LDA provided the best results in differentiating between different levels of adulterations followed by k-Means clustering algorithm. The developed sensor is a proof-of-concept for development of disposable electronic tongue with potential applications in beverage quality monitoring.