The Influence of the Underlying Gold Substrate on Glucose Oxidase Electrodes Fabricated Using Self-Assembled Monolayers D. Losic, + J. J. Gooding,* ++ J. G. Shapter, + D. B. Hibbert, + and K. Short +++ + School of Chemistry, Physics and Earth Science, The Flinders University of South Australia, Adelaide 5001, Australia ++ School of Chemistry, The University of New South Wales, Sydney, NSW 2052, Australia; e-mail: justin.gooding@unsw.edu.au +++ Materials Division, ANSTO, Lucas Heights, NSW 2234, Australia Received: March 23, 2001 Final version: May 24, 2001 Abstract The influence of the underlying gold surface topography on the response of enzyme electrodes fabricated by covalent attachment of glucose oxidase to a 3-mercaptopropionic acid self-assembled monolayer was assessed. Six different gold substrates were prepared. The bare substrates were evaluated for roughness and defects using atomic force microscopy (AFM) and scanning tunneling microscopy (STM). Despite large variations in surface roughness, the shapes of the enzyme electrode calibration plots indicate that approximately the same amount of MPAwas adsorbed on the surface and from electrochemical stripping experiments the same number of active enzyme molecules were attached. However, compared with the smoother surfaces, enzyme electrodes fabricated on surfaces with high microscopic roughness had a smaller linear range but a similar current sensitivity in the linear portion of the calibration plot. Tapping-mode AFM images of the enzyme modified surfaces showed that the enzyme adsorbed onto the surface as clusters of five or six enzyme molecules. The formation of these clusters did not appear to be influenced by the surface roughness. Keywords: Gold surface topography, Glucose oxidase, 3-Mercaptopropionic acid, Self-assembled monolayer, Roughness, AFM, STM 1. Introduction The fabrication of enzyme electrodes using gold electrodes modified with self-assembled monolayers (SAMs) as the base on which to immobilize the enzyme is becoming increasingly popular [1–3]. Part of the attraction of using SAMs is a mono- layer of enzyme can be immobilized close to an electrode surface. Subsequent layers of enzyme can then be assembled onto this initial enzyme layer to provide molecular level control over the spatial organization of the sensing interface [4–10]. Thus the importance in understanding parameters which are influential in defining the reproducibility and structure of the initial enzyme monolayer is apparent. This article is primarily concerned with the effect of the underlying gold surface on the performance of monolayer enzyme electrodes. Most studies in which a monolayer of enzyme is immobilized onto a gold surface using self-assembled monolayers have employed short chain alkanethiols; commonly three carbon atoms in the alkyl chain [1, 5, 11–16]. The length of such SAMs are well within the range of surface roughness of the underlying gold surface. Hence it is conceivable that the gold surface may play a significant role in how the enzyme electrode performs. Surprisingly, there are only two reports we are aware of where the gold surface upon which the enzyme electrode is fabricated is even considered. In the first by Riklin and Willner [5] a gold surface roughened by almagamization was shown to give increased current response over polished bulk gold surface for a given geometric area. However, despite the increased sensitivity the actual current density related to the electrochemically accessible area was decreased. It may be inferred from this decrease in current density that the roughness of the gold elec- trode can have a significantly detrimental effect on the enzyme and hence the performance of the resultant biosensor. In contrast in the other study by Gooding et al. [11] the same current density was observed at a polished bulk gold surface and at a sanded gold surface. The difference between these two observations may reflect the scale of the surface morphology, with the sand roughening forming large macroscopic changes while the amal- gamization may produce roughness of a similar scale to the SAM and the enzymes. More detailed studies on the influence of the underlying gold surface morphology and topography on the integrity of SAMs have concentrated on long chain SAMs [17–20]. These studies all investigated gold surfaces in attempts to form defect-free monolayers. In all three studies there seems to be agreement that the important criteria for obtaining a SAM with minimum defects was a surface with the lowest microscopic roughness. By microscopic roughness the authors mean roughness on the order of the size of the adsorbate. Hence bulk gold surfaces which are rough on a macroscopic scale but contain very few grain boundaries assemble SAMs with less defects than apparently smooth evaporated gold films that contain many more grain boundaries [18, 20]. The role of this microscopic roughness is further highlighted by Porter and co-workers where reductive desorption studies showed that on gold surfaces with many steps, two stripping peaks were observed due to thiols on the flat surface and thiols at the step boundaries [17, 19]. The work of Losic et al. [20] is also consistent with the findings of Porter. Of six different gold surfaces evaluated, the most defect free SAMs were formed on ultraflat gold surfaces fabricated using mica as a template to form the gold surface. In this fabrication method the gold surface originally in contact with the atomically flat mica is used. The resultant gold surfaces revealed that once the mica is stripped from the surface they are possibly the flattest gold surfaces yet fabricated. The previous studies on gold surfaces highlight the importance of smoothness of the gold surface for long-chain SAMs where integrity is an issue. However, the criteria for an enzyme elec- trode are quite different and therefore it is not clear whether the quality of the underlying gold surface will have a significant 1385 Electroanalysis 2001, 13, No. 17 # WILEY-VCH Verlag GmbH, D-69469 Weinheim, 2001 1040-0397/01/1711–1385 $17.50þ.50=0