Optimization of "wired" enzyme O2-electroreduction catalyst compositions by scanning electrochemical microscopy.

“Wired” enzyme electrodes, comprising enzymes connected electrically through redox polymers, which swell to give electron-conducting hydrogels, are already used in experimental subcutaneously implanted electrodes for the continuous monitoring of glucose in diabetics and may be used in future miniature, microwatt producing, membrane-less biofuel cells. The subcutaneously implanted electrodes comprise, in addition to the “wired” enzyme film that transduces the glucose flux to a current, a glucose flux-controlling membrane, which defines the measurable glucose concentration range, and optionally, a bioinert film. The membrane-less miniature biofuel cells comprise a “wired” glucose oxidase anode, and a “wired” bilirubin oxidase or laccase cathode. Optimizing the performance of these devices requires synthesis and characterization of new redox polymers, 5] defining the optimal enzyme–cross-linker–redoxpolymer compositions, and synthesis and optimization of flux controlling and bioinert membranes. A key parameter, which is usually explored first, is the polymer/enzyme ratio. An excessive enzyme weight fraction can decrease the current density, because the enzyme, unlike the redox polymer, is an electronic insulator. Furthermore, if the enzyme weight fraction is high enough for the net negative charge of glucose oxidase to balance the positive charge of the redox polymer, an electrostatic adduct that shows poor electronic conductivity precipitates. Moreover, when the weight fraction of the redox polymer is excessive, the flux of electrons is reduced, because of the smaller number of enzyme molecules. The effort and material expended in the optimization process 11] can be reduced by combinatorial screening techniques exemplified by optimization through scanning electrochemical microscopy (SECM), a technique that has been extensively used for evaluating the activity of enzymatic systems and whose utility is demonstrated herein. Because only optimization of an exemplary parameter, the weight fraction of the multicopper oxidase bilirubin oxidase (BOD) or laccase, of a biofuel cell cathode was performed, the study does not provide as yet a fully optimized electrode, for which additional parameters, including the total loading and the weight% of the cross-linker, would have to be defined, along with the thickness of the electrode. The results obtained establish, nevertheless, that SECM screening of compositionally varying arrays of mm-size spot electrodes yields results identical with those obtained with discreet rotating disk electrodes, but much more efficiently in terms of speed and material required. Ternary mixtures of enzyme, cross-linker, and redox polymer, with compositions ranging from pure enzyme to pure polymer, were prepared from aqueous solutions of BOD from Trachyderma tsunodae (8 mgmL 1 in pH 7.2 20-mm phosphate buffer) or laccase from Coriolus hirsutus (8.6 mgmL 1 in pH 5.0 20-mm citrate buffer), and PAAPVI-[Os(4,4’-dichloro-2,2’-bipyridine)2Cl] 1+/2+ (8 mgmL ) or PAA-PVI-[Os(tpy)(dme-bpy)Cl] (8.6 mgmL ), respectively (PAA= poly(1-carboxy-1,2-ethanediyl, PVI= poly(1imidazolyl-1,2-ethanediyl, tpy= 2,2’:6’,2’’-terpyridine, dmebpy= 4,4’-dimethyl-2’,2-bipyridine). A solution of polyethylene glycol diglycidyl ether at 2 mgmL 1 (PEDGE, Polysciences Inc.) was used as the cross-linker. Arrays of spots containing these mixtures were deposited on glassy carbon (GC) plates 15 B 15B 1mm (Alfa) by using a piezo-based micro-arrayer, a device similar to that used by Schuhmann and co-workers for micro-patterning of enzymatic biosensors. A commercial piezo-dispenser MicroJet AB-01-60 (MicroFab) with an outlet aperture of 60 mm, that dispenses on-demand picoliter-sized ( 100 pL) droplets by application of potential pulses (50 V, 25 ms), was installed onto the head of a digital plotter (Houston Instruments DMP-5) to control its position with a resolution of 100 mm/step. Figure 1 shows the preparation scheme. The dispenser was filled with polymer solution (3 mL), which was dispensed in a programmed number of drops at each site. To test for reproducibility, each composition was prepared in duplicate or triplicate. Thus, the arrays contained 11 rows and two columns of spots. Typically, the first two-spot row contained 10 drops per spot (pure polymer), the next row 9 drops per spot and so on down to and the tenth row with 1 drop per spot. The dispenser was [*] Dr. J. L. Fern ndez, Prof. Dr. A. J. Bard Chemistry and Biochemistry University of Texas at Austin Austin, TX 78712 (USA) Fax: (+1)512-471-0088 E-mail: [email protected]