Discrimination of green tea using an Epigallocatechin-3-gallate (EGCG) sensitive molecular imprinted polymer (MIP) based electrode

Debangana Das (1, *), Trisita Nandy Chatterjee (1), Runu Banerjee Roy (1), Bipan Tudu (1), Santanu Sabhapondit (2), Ajanto Kumar Hazarika (2), Panchanan Pramanik (3), Rajib Bandyopadhyay (1, 4)

1. Department of Instrumentation and Electronics Engineering, Jadavpur University, Kolkata 700098, India.
2. Tocklai Tea Research Institute, Jorhat, Assam, India.
3. Institute of Applied Science and Humanities, GLA University, Mathura, India.
4. Laboratory of Artificial Sensory Systems, ITMO University, Saint Petersburg, Russia. * Corresponding author. Tel/Fax: +91 33 23352587/57254

Carbon - Science and Technology 10/4 (2018) 27 - 37.

© Applied Science Innovations Private Limited, India.

Full Text (OPEN ACCESS): CST-314.pdf

 

Keywords: Epigallocatechin- 3- Gallate; Preprocessing technique; Clustering algorithms; Green Tea

Abstract: In this work, a simple approach of discriminating green tea samples has been proposed using an epigallocatechin-3-gallate (EGCG) sensitive molecular imprinted polymer (MIP) electrode modified with chemically synthesized nickel hydroxide (Ni(OH)2) nanoparticles. The nanoparticles were characterized by powder X-ray diffraction techniques (XRD) and the removal of the template molecule has been ascertained by UV-vis spectroscopy. A three electrode system has been employed to study the electrochemical characteristics of the electrode by means cyclic voltammetry (CV) and differential pulse voltammetry (DPV). Four different kinds of preprocessing techniques, namely – (i) Baseline subtraction, (ii) Autoscale, (iii) Relative scale 1 and (iv) Relative scale 2 were applied on the obtained data set and the best preprocessing technique was optimized. Further, principal component analysis (PCA) and linear discriminant analysis (LDA) were implemented on the preprocessed data set so as to observe the discrimination ability of the electrode on the basis of EGCG content in green tea. The separability index (SI) values for both PCA and LDA plots is calculated and it is observed that baseline subtraction provided the best result with a SI value of 8.72 and 16.01, respectively.