The engineering of carbon nanofibers (CNFs) has recently received extensive attention in the field of biosensors due to their high surface area, rich nanoscale geometries, ability to selectively detect the analyte of interest (e.g., dopamine) in the presence of interfering molecules that exist in biological environment and inherent resistance to biofouling1 - 3. CNFs with the individual fiber length/size adjusted to the diffusion layer thickness can confine the analyte molecules within the nanostructures and result in the formation of a thin liquid layer, increasing their sensitivity 4. This points towards the potential of altering the aspect ratios and growth densities of CNFs to tune their biosensing properties. Thus, here we aim to evaluate the effect of the length, distribution, and microstructure of fibers on their electrochemistry and how this affects the sensitivity and selectivity for dopamine (DA) detection.Herein, CNFs of varying lengths have been grown on a silicon substrate using the thin layers of Ni and Cr as catalysts. Samples with different length and distribution of CNFs have been synthesized by controlling the duration of the growth phase. The length and the detailed structure of the CNFs have been investigated using scanning electron microscopy (SEM) and transmission electron microscopy (TEM). The lengths of CNFs were found to be in the range of 900 nm-1µM, 500-600 nM, 200-300 nM, and <150 nM for 30 mins, 10 mins, 5 mins and 1 min growth times, respectively. Cyclic voltammetry measurements showed that the double-layer capacitance (Cdl) and oxidation current (Ipa) of dopamine increase continuously with the increase in the length of fibers indicating the increase in the surface area (Figure 1A). DA reaction kinetics tend to be reversible for the longest fibers and becomes slightly more sluggish with the decrease in the fiber length. However, it should be noted that carbon nanofiber length (hundreds of nM to 1µm) may match the diffusion layer thickness at certain scan rates, causing the thin liquid layer electrochemical behavior to contribute to the total kinetics. This implies that the variations in the peak separation (ΔEp) can be due to the combined changes in reaction kinetics and geometry. CNFs grown for 30 mins, 10 mins, 5 mins, and 1 min exhibited log Ipa vs v slope of 0.72, 0.63, 0.62 and 0.51, indicating that contribution of adsorption/thin layer formation is coming into effect with the increase in fiber length (Table 1). Washout experiments will be performed to distinguish further between thin layer and adsorption behavior affecting the electrochemistry of DA. It appears that DA reaction kinetics is fully under semi-infinite linear diffusion-control only at 1 min grown CNFs due to the relatively smooth surface of the electrode material. The selectivity of dopamine in the presence of physiological concentrations of ascorbic acid (AA) and uric acid (UA) greatly improves with the increase in the length of the fibers (Figure 1B). Electrodes with 1 min grown CNFs possessing metal particles and exhibiting little CNF growth did not show selectivity towards DA. Interestingly, with the increase in the growth time and consequently the length of the fiber, well-defined oxidation peaks of AA and DA and UA are observed. Moreover, the position of the oxidation potential appears to be affected both by the fiber length and presence/absence of other molecules, and often in the opposite directions (Table 1, Figure 1). The results discussed above point towards the significance of matching the CNF length with the scan rate used for in vivo detection of DA for optimized sensitivity and selectivity. References 1 S. Sainio, T. Palomäki, N. Tujunen, V. Protopopova, J. Koehne, K. Kordas, J. Koskinen, M. Meyyappan and T. Laurila, Mol. Neurobiol., 2015, 52, 859–866.2 S. Sainio, E. Leppänen, E. Mynttinen, T. Palomäki, N. Wester, J. Etula, N. Isoaho, E. Peltola, J. Koehne, M. Meyyappan and others, Mol. Neurobiol., 2020, 57, 179–190.3 A. Kousar, E. Peltola and T. Laurila, ACS omega, 2021, 6, 26391–26403.4 Q. Cao, Z. Shao, D. K. Hensley, N. V Lavrik and B. J. Venton, Langmuir, 2021, 37, 2667–2676. Figure 1