Enzyme immobilization and permselectivity analysis of an interference free peroxide based glucose biosensor

• Main Author: Wong, Fui Ling
• Format: Thesis
• Language: English
• Published: 2008
• Subjects: TP Chemical technology
• Online Access: http://eprints.utm.my/id/eprint/9648/1/WongFuiLingMFKT2008.pdf; http://eprints.utm.my/id/eprint/9648/; http://dms.library.utm.my:8080/vital/access/manager/Repository/vital:1084

The performance of a biosensor sensing layer is mostly determined by the immobilization method and the materials used to immobilize the enzyme. In this work, the performances of four types of glucose oxidase immobilization materials based on poly(vinyl alcohol) (PVA) for peroxide-based glucose biosensor were compared. The matrices of interest were glutaraldehyde cross-linked PVA (GAPVA), freeze-thawed PVA cryogel (FTPVA), tetramethoxysilane sol-gel-PVA hybrid material (TMOS-PVA), and alumina sol-gel-PVA hybrid material (Al-PVA). Only GAPVA showed short period of enzyme leaking and high value of Km app. However, its sensitivity was poor. With the same enzyme loading, the other three types of membranes showed good sensitivity and stability. FTPVA and TMOSPVA, which showed satisfactory sensitivity and adequate value of Km app were quite promising as the support materials for immobilizing glucose oxidase (GOD). The enzyme leakage of FTPVA which had shown highest leaking was improved by reducing the enzyme loading. Even though peroxide-based sensor is very simple and easy to construct, it suffers from electrochemical interferences from common electroactive species present in blood such as acetaminophen. Hence, a photocured poly(hydroxyethyl methacrylate) (pHEMA) layer was investigated as a potential permselective inner membrane to eliminate the interferences based on size exclusion. PHEMA membrane with the cross-linking ratio of 0.043 was found to be able to achieve a selectivity of 10, while maintaining an acceptable degree of hydrogen peroxide response. In a two-layer glucose biosensor, where the FTPVA-GOD and cross-linked pHEMA were combined, acetaminophen and ascorbic acid at 0.2 mM were eliminated completely. 0.2 mM uric acid still gave a bias of approximately 6.6% relative to 5 mM glucose. Overall, the partial glucose biosensor showed promising performance.