Abstract—We describe electrochemical recording of catecholamines using gold and platinum nanowire (NW) electrodes with diameters of 30-200 nm and lengths of 1-5 μm. The NW width (30 nm) falls within the range of the size of the synaptic cleft, which is 20-50 nm in length. The electrochemical devices were fabricated by depositing NWs on silica substrates followed by photolithographic patterning of the contact pad for electrical contact. Special care was taken to expose only the NW surface to the catecholamine solution during calibration, by covering the entire surface of the contact pad with photoresist using a second photolithographic step. Calibration curves were generated in both beaker experiments and microfluidic channels to assess the ability of the nanoelectrodes to detect dopamine (DA) and serotonin (5-HT) at predicted synaptic concentrations. Using the NW electrodes, we were able to detect DA and 5-HT in mM concentrations. This sensitivity demonstrates the utility of the device in measuring predicted synaptic concentrations of DA (approximately 1.6 mM) and for investigating the changes in 5-HT concentration during epileptic seizures. The NW recording devices also exhibit calibration parameters comparable to current carbon fiber electrodes (linearity of 0.995). The small electrode size and the ability to record predicted synaptic concentrations point to the possibility of using the NW electrode device to do neural recording from a synaptic cleft. BACKGROUND IN the late 70's, Adams used voltammetry to demonstrate that catecholamines could be oxidized in-vivo by pulsing a constant voltage current through an electrode inserted into brain tissue [1]. The resulting oxidation current was recorded using a potentiostat and was found to be proportional to the extracellular concentration of the monoamine near the tip of the recording electrode. This measurement capability is important to record changes in the extracellular concentration of these monoamines because they have been implicated in several mental illnesses, including attention deficity hyperactivity disorder [2], schizophrenia [3] and epilepsy [4]. Since then, carbon fiber electrodes have become the current standard and are approximately 10 μm in diameter causing minimal damage to the neural tissue during recording. Nevertheless, even 10 µm diameter electrodes can record extracellular neurotransmitter concentrations from only a cluster of neurons. Our goal was to develop a nanosensor that was small enough to fit into the synaptic cleft between two neurons, as a first step to enable single neuron recording. Hence we developed a nanosensor with a diameter in the size range of the synaptic cleft (20-50 nm). CURRENT RESULTS Nanowires ranging from 30 to 200 nm were fabricated by electrodeposition of the appropriate metal in anodized nanoporous membranes as described elsewhere [5]. After electrodeposition, the NWs were released from the membrane by dissolution in sodium hydroxide. The NWs were then rinsed repeatedly with water and ethanol and then stored in ethanol. The NW electrodes were fabricated as shown in Fig. 1. Briefly, the NWs were dispersed on a silica substrate. Using photolithography, a contact pad consisting of a 50 nm chromium adhesion layer and 200 nm copper was aligned with one end partially covering the electrode. Since the surface area of the contact pad was much larger than that of the NW, and the metal of the pad would contribute to the current measured, an insulation layer of photoresist was patterned to completely cover the contact pad while leaving the NW exposed (Fig. 2). Calibration curves were conducted in phosphate buffered saline using a Keithley model 6430 sourcemeter (Keithley Inc., Cleveland, OH) in a two terminal potentiostatic configuration (Fig. 3a). In all experiments, the voltage across the terminals was fixed at 0.55V for DA and 0.45 V for 5-HT and the current was monitored for analysis. For each electrode, data from a minimum of three successive aliquot additions, and two consecutive repeat experiments were then used to generate a linear regression fit (calibration curve, Fig. 3b). The slope of the resulting calibration curves represent the sensitivity of the electrode to DA concentrations. It should be noted that the data points for each aliquot in the calibration curve were determined from the 20-80% range average of the signal to avoid skewing of the data by the initial spike in current that occurs upon the addition of the DA. From the calibration curves, the sensitivity, linearity and their correlation to NW surface area were then determined. The 200 nm Au (sensitivity: 2.37 µA/mM, R2 = 0.992) and Pt (sensitivity: 1.29 µA/mM, R2 = 0.955) electrodes were comparable. The statistical differences observed can be attributed to photolithographic alignment, surface charge, presence of edge plane sites and oxidation of the nanowire surface 17. The 30 nm Au nanosensor however, had a sensitivity of 0.27 µA/mM and had increased linearity over both of the 200 nm nanosensors (R2 = 0.996). REFERENCES [1] R. N. Adams “Probing brain chemistry with electroanalytical techniques,” Analytical Chemistry, vol. 48, pp. 1126A-1138A, 1976. [2] M. V. Solanto “Attention-deficit/hyperactivity disorder: clinical features”. In: psychostimulant drugs and ADHD: basic and clinical neuroscience, pps. 3-30, (Eds. M Solanto, AFT Arnsten, FX Castellanos), Oxford, Oxford University Press pp. 3-30, 2001. [3] L. Linner, C. Wiker, M. L. Wackenberg, M. Schalling, and T. H. Svensson. “Noradrenaline reuptake inhibition enhances the antipsychotic-like effect of raclopride and potentiates D2-blockage-induced dopamine release in the medial prefrontal cortex of the rat,” Neuropsychopharma, vol. 27, pp. 691-698, 2002. [4] W. C. Ziai, D. L. Sherman, and M. A. Mirski. “Target specific catecholamine elevation induced by anticonvulsant thalamic deep brain stimulation,” Epilepsia, vol.. 46, pp. .878-881, 2005. [5] H. Ye, Z. Gu, T. Yu and D. H. Gracias “Integrating nanowires with substrates using directed assembly and nanoscale soldering,” IEEE Trans Nanotech., vol. 5(1), pp. 1-6, 2006.