Evaluation of carbon nanotube fiber microelectrodes for neurotransmitter detection: Correlation of electrochemical performance and surface properties

Abstract

Fibers made of CNTs are attractive microelectrode sensors because they can be directly fabricated into microelectrodes. Different protocols for making CNT fibers have been developed, but differences in surface structure and therefore electrochemical properties that result have not been studied. In this study, we correlated the surface and electrochemical properties for neurochemical detection at 3 types of materials: CNT fibers produced by wet spinning with (1) polyethylenimine (PEI/CNT) or (2) chlorosulfonic acid (CA/CNT), and (3) CNT yarns made by solid-based CNT drawing. CNT yarns had well-aligned, high purity CNTs, abundant oxygen functional groups, and moderate surface roughness which led to the highest dopamine current density (290 ± 65 pA/cm²) and fastest electron transfer kinetics. The crevices of the CNT yarn and PEI/CNT fiber microelectrodes allow dopamine to be momentarily trapped during fast-scan cyclic voltammetry detection, leading to thin-layer cell conditions and a response that was independent of applied waveform frequency. The larger crevices on the PEI/CNT fibers led to a slower time response, showing too much roughness is detrimental to fast detection. CA/CNT fibers have a smoother surface and lower currents, but their negative surface charge results in high selectivity for dopamine over uric acid or ascorbic acid. Overall, small crevices, high conductivity, and abundant oxygen groups led to high sensitivity for amine neurotransmitters, such as dopamine and serotonin. Thus, different surfaces of CNT fibers result in altered electrochemical properties and could be used in the future to predict and control electrochemical performance.

Keywords: CNT fiber; Fast-scan cyclic voltammetry; Microelectrode; Neurotransmitter; Surface properties.

Figure 1

Surfaces of microelectrodes. SEM images of the (A) CA/CNT fiber, (B) PEI/CNT fiber, and (C) CNT yarn sidewall, with the diameters about 20 μm. Scale bar: 10 μm. Top-view SEM images at the tips of polished (D) CA/CNT fiber, (E) PEI/CNT fiber, and (F) CNT yarn microelectrodes. Scale bar: 500 nm.

Figure 2

Electrochemical response to 1 μM dopamine with a waveform of −0.4 V to 1.3 V and back at 400 V/s, 10 Hz. (Left) Background subtracted cyclic voltammograms at (A) CA/CNT fiber, (B) PEI/CNT fiber, and (C) CNT yarn microelectrodes. (Right) Current versus time trace plot for the same electrodes measured at 0.6 V in flow injection experiment. The first arrow indicates the injection of dopamine bolus and the second arrow indicates the switch back to buffer.

Figure 3

Detection of different neurochemicals at CNT fiber microelectrodes. (A) Detection of 20 μM uric acid, 200 μM ascorbic acid, and 1 μM serotonin in PBS buffer at a CA/CNT fiber (top), PEI/CNT fiber (middle), or CNT yarn (bottom) microelectrode. (B) Bar graphs show the ratio of oxidation current for UA, AA, or serotonin compared to the oxidation current of dopamine (n=4 each). * p < 0.05 and **** p < 0.0001

Figure 4

Effect of scan repetition frequency for 1 μM dopamine detection. (A) Example CVs of 1 μM dopamine at the scan repetition frequency of 10 Hz (blue line) and 100 Hz (orange line), with −0.4 to 1.3 V waveform. (B) Peak oxidation current at PEI/CNT fiber microelectrodes (blue circle, n=4), CNT yarn microelectrodes (black triangle, n=5), carbon fiber microelectrodes, (red circle, n=5), and CA/CNT fiber microelectrodes (green triangle, n=4), from top to bottom. Data is normalized to dopamine oxidation signal at different microelectrodes with scan repetition frequency of 10 Hz.