Flexible, diamond-based microelectrodes fabricated using the diamond growth side for neural sensing

The study addresses the challenge of creating implantable neural sensors that can sensitively and selectively detect neurotransmitters (notably dopamine) while also recording neural activity, with long-term stability and biocompatibility. Dysregulated neurotransmitters like dopamine are implicated in disorders such as Parkinson’s disease and schizophrenia, motivating real-time monitoring. Conventional metal and silicon microelectrodes often degrade chronically due to inflammatory responses and mechanical mismatch with soft brain tissue, while polymer-based and carbon fiber solutions face fabrication and durability challenges. Electrochemical detection of dopamine is complicated by high concentrations of interferents such as ascorbic acid with similar oxidation potentials. Boron-doped diamond (BDD) is attractive for electrochemical sensing due to its wide potential window, low double-layer capacitance, and chemical inertness, but its stiffness necessitates integration with flexible substrates. Prior flexible BDD devices typically expose the nucleation surface, which contains more sp² impurities and smaller grains, degrading electrochemical performance. This work proposes exposing the BDD growth surface, hypothesized to enhance sensitivity, selectivity, and stability, and evaluates its electrochemical properties, dopamine sensing, neural recording capability in vitro and in vivo, and biocompatibility.

The paper reviews prior neural sensing and recording approaches, noting limitations of metal microwires and silicon probes in chronic settings and the promise and fabrication challenges of polymer-based and carbon fiber microelectrodes. For neurotransmitter sensing, microdialysis offers high chemical sensitivity but poor temporal resolution, while electrochemical microelectrodes provide real-time monitoring but suffer from interference (e.g., ascorbic acid) near dopamine’s oxidation potential. Among carbon materials (glassy carbon, CNTs, carbon fibers), BDD stands out for wide potential window, low capacitance, and chemical inertness. Film growth physics indicate that nucleation surfaces have smaller grains and higher sp² impurity at grain boundaries, adversely affecting electron transfer and selectivity, while thicker films and growth surfaces have higher sp³ content and improved kinetics. Prior flexible diamond devices transferred BDD structures to polymers but only exposed the nucleation surface, limiting performance. Earlier studies showed thicker BDD films reduce sp² content, enhance electron transfer in Ru(NH3)6 redox, increase dopamine faradaic currents, and improve DA/AA peak separation.

• Device fabrication: Developed a wafer-scale process to integrate boron-doped polycrystalline diamond (BDD) microelectrodes on flexible Parylene C, exposing the BDD growth surface as the sensing interface. Comparative probes were also fabricated exposing the nucleation surface. The proof-of-concept probe used a three-electrode configuration on a single flexible shank with two channels, each comprising a circular working electrode (WE) surrounded by a ring-shaped counter electrode (CE); a square reference electrode (RE) was shared. Exposed BDD areas: WE ~0.0079 mm², CE ~0.028 mm², RE ~0.035 mm².
• Materials characterization: SEM to assess surface morphology and grain size; Raman spectroscopy (532 nm) to evaluate sp³ diamond peak shifts and boron-doping-related peaks; resistivity estimation of BDD films (~5×10⁻³ Ω·cm).
• Electrochemical impedance spectroscopy (EIS): Measured broadband impedance (10 Hz–100 kHz) in 0.1 M PBS (pH 7.4) at room temperature for growth-side vs nucleation-side electrodes.
• Cyclic voltammetry (CV): Assessed potential window and background currents in 1.0 M KCl using Pt CE and Ag/AgCl RE at 0.1 V/s, comparing BDD growth side, BDD nucleation side, and a commercial Au electrode.
• Electron transfer kinetics: CV with 1.5 mM Ru(NH₃)₆²⁺/³⁺ in 1.0 M KCl at 0.1 V/s using BDD electrodes as both CE and RE to determine peak-to-peak separation (ΔEp).
• Double-layer capacitance (Cdl): Background CVs at scan rates 0.1–3.0 V/s in 1.0 M KCl; slope of average current vs scan rate used to estimate Cdl.
• Dopamine sensing: CV in 1.0 mM DA in 0.1 M PBS (pH 7.4) at 1.0 V/s for both surfaces; fouling test by extended soaking in DA solution; square-wave voltammetry (SWV) for DA detection in presence of 100 μM ascorbic acid (AA); calibration across 5–100 μM DA with linear fit and LOD determination.
• Biocompatibility and neural recording: In vitro tests with rat cortical neuron cultures to assess compatibility; validation of neural recording capability both in vitro (cortical neuron culture) and in vivo (rat primary visual cortex, V1).

• Morphology and composition: Growth side exhibited a rougher surface with larger grains (~0.5 μm average) and higher sp³ content; nucleation side showed smoother morphology and a Raman feature near ~1470 cm⁻¹ indicating higher sp²/amorphous content. Boron-doping-related Raman peaks (~473 and ~1209 cm⁻¹) confirmed heavy doping (~10²⁰ cm⁻³); film resistivity ~5×10⁻³ Ω·cm.
• Impedance: At 1 kHz, growth-side electrode impedance ~207.9 kΩ vs nucleation side ~1,123.8 kΩ (≈4–5× lower), attributed to rougher surface and larger grains, implying reduced noise for neural recording.
• Potential window and background: BDD electrodes showed wider water potential windows and featureless low background currents compared to Au; nucleation side displayed background peaks likely from sp² site functional groups.
• Electron transfer kinetics: Ru(NH₃)₆²⁺/³⁺ CV peak separation ΔEp: growth side 80 mV (quasi-reversible, near theoretical 58.5 mV), nucleation side 191 mV (slower kinetics).
• Double-layer capacitance: Growth side Cdl ~10 μF/cm², markedly lower than nucleation side reported previously (~24 μF/cm²), enabling reduced charging currents and lower background noise.
• Dopamine sensing: Growth-side CVs in 1.0 mM DA showed sharp, symmetric oxidation/reduction peaks; nucleation side showed broad oxidation and negligible reduction, indicating sluggish electron transfer due to sp² impurities. Growth side demonstrated better sensitivity, selectivity, and lower surface adsorption/fouling. SWV in presence of 100 μM AA showed DA detection with linear calibration y = 0.9x + 6.0 (R² = 0.989) over 5–100 μM and LOD ~0.83 μM (830 nM).
• Device functionality: Flexible Parylene C-based BDD growth-side electrodes enabled in vitro and in vivo neural recordings; microcrystalline BDD films showed in vitro biocompatibility with rat cortical neuron cultures.

Exposing the BDD growth surface substantially improves electrode electrochemical performance relative to the nucleation surface. Larger grain size and reduced sp² impurity content on the growth side enhance electron transfer kinetics (lower ΔEp), widen the usable potential window, reduce double-layer capacitance and background currents, and lower interfacial impedance, all of which are advantageous for sensitive and selective electrochemical detection in complex biological environments. These improvements translate to enhanced dopamine sensing, including better-defined redox peaks, higher sensitivity with low interference from ascorbic acid, and reduced fouling, addressing key challenges in in vivo neurotransmitter monitoring. Mechanically, integrating BDD on a flexible Parylene C substrate mitigates stiffness mismatch with neural tissue, supporting stable neural recordings demonstrated both in vitro and in vivo. Collectively, the findings support the hypothesis that using the growth surface of BDD yields superior sensing and recording performance suitable for implantable neural interfaces.

The study introduces a wafer-scale fabrication strategy for flexible BDD microelectrodes that expose the diamond growth surface, yielding marked improvements in impedance, potential window, electron transfer kinetics, capacitance, and dopamine sensing performance compared to nucleation-surface devices. The growth-side electrodes exhibit enhanced sensitivity, selectivity, and stability, demonstrate neural recording capability in vitro and in vivo, and show in vitro biocompatibility. This approach advances the development of flexible, diamond-based implantable sensors for multimodal neural interfacing. Potential future work includes scaling to higher channel counts, optimizing electrode geometries for further impedance reduction, and conducting long-term chronic in vivo studies to assess stability and tissue responses over time.