Construction and electrochemical properties determination of carbon nanotubes/porous-network carbon micron tubes/silk fibroin composite film neural electrode

The treatment of central nervous system diseases has been a worldwide problem. The most widely used treatment is drugs, but the effects are limited.¹ Along with the development of tissue engineering and microelectronics technology, neuroprosthetic implanted therapy has been proposed and has shown a remarkable effect in some central nervous system diseases, such as implanted deep brain electrical stimulation treatment of Parkinson’s disease.^2–5 As the link between the implantable neural prosthesis and the bioneural network, the neural electrodes are of critical importance in the performance of neuroprosthetics. Owing to their good corrosion resistance and biocompatibility, Pt, Ir, Au, and other precious metal materials are widely used in neural prosthesis as electrode materials.^6,7 However, due to the lower charge injection capacity (Q_inj less than 0.15 mC/cm²), they cannot reach the high density charge stimulation threshold. So far, much effort has been made on the research of new neural electrode materials.

Currently, the research on improving the electrochemical properties is divided into two methods. One is to increase the faradian capacitor of the electrode with faraday-capable materials such as iridium oxide and conductive polymers.^8–17 Cogan et al. obtained iridium oxide (SIROFS) on the electrode surface by sputtering, and the Q_inj value was measured to be about 4.0 mC/cm².¹⁸ However, the stability of the electrodes is insufficient due to the oxidation reduction reaction of the faraday capacitive material during discharge. In addition, the other improvement method is to increase the double-layer capacitor by raising the surface roughness or by using materials with a larger specific surface area. Wang et al. made a microelectrode based on carbon nanotubes (CNTs), which showed a high Q_inj value of 1–1.6 mC/cm², which was 10 times higher than that of the Pt electrode.^19,20 The neural electrode based on the mechanism of a double-layer capacitor does not have electrochemical reactions in the process of charge transfer, which is beneficial to the stability of the electrode when used, but the charge capacity of the double-layer capacitor is generally not high, and the charge stimulation threshold is low.

Good biocompatibility can avoid inflammatory reactions when the electrode is implanted in vivo. Currently, a hydrogel coating with good biocompatibility or biomaterial doping and self-assembly are generally used to enhance the biocompatibility of the electrode.^21–30 In the present work, the silk fibroin materials with excellent biocompatibility were used to prepare the silk protein film with high porosity on the flaky Pt electrode, and the environmentally friendly reagent phytic acid was used as a dispersant to disperse the CNTs.^31,32 Phytic acid can not only disperse CNTs, but also form strong hydrogen bonds with silk fibroin, which is conducive to elevating the biocompatibility and mechanical properties of materials. The CNTs were then self-assembled on the silk protein film and were irradiated by carbon dioxide laser, and as a result, we obtained carbon nanotubes/porous-network carbon micron tubes/silk fibroin (CPS) electrodes with a carbonized surface of the silk protein film. The composite material presented a good double-layer capacitive performance and high charge injection limits because it combined the distinguished storage capacity of porous-network carbon microtubes (PCMTs) and the unique physical and chemical properties of CNTs, as well as formed a continuous “conductive band.”

Figure 1 shows the fabrication procedure of the CPS electrode. The procedure consisted of three main steps: first, porous SF films (with an aperture of 30–80 μm, porosity ≥90%) were prepared using the freeze-drying method on a 2 × 2 mm² Pt electrode.^33–35 Second, phytic acid was employed as a dispersant in multi-walled CNTs with 10–30 µm in length and 30–60 nm in diameter. The CNTs were uniformly self-assembled on the SF film under ultrasonic conditions to form the CNTs/SF composite film. Finally, CO₂ laser (UL-50-O EM, Universal Laser Systems) was used to irradiate the CNTs/SF composite film under Ar gas protection (the laser wavelength was 10.6 µm, the laser power was 10 W, and the output spot diameter was 4 mm), resulting in CPS electrodes.

The surface morphology and composition of the films were characterized by using a scanning electron microscope (SEM, Hitachi s-4800, Hitachi, Japan) and a Raman spectrometer (Renishaw InVia Raman tin, UK) in each step. As shown in Fig. 2(a), the surface of the CNTs/SF composite film was quite rough, with many channels and cavities.^36,37 In the enlarged view as shown in Fig. 2(b), CNTs with regional distribution on the surface of SF can be clearly observed. After further magnification, it could be seen that the CNTs on the CNTs/SF composite film were well dispersed among each other, with slight entanglement in a few areas, as shown in Fig. 2(c). It can be seen from Fig. 2(d) that after laser irradiation, the SF surface was crimped and carbonized to form PCMTs. A larger version of Fig. 2(e) shows that PCMTs are interconnected into a network, and CNTs on the surface are distributed more evenly and densely. As can be seen from a further magnification in Fig. 2(f), CNTs on the CPS film were fused with PCMTs to be more uniform and smoother. Figures 2(g) and 2(h) present the Raman spectra of CNTs/SF before laser irradiation and CPS films, respectively. We can see three obvious peaks from both figures, two of which are near 1346 cm⁻¹ representing the D peak and near 2659 cm⁻¹ representing the 2D peak, which are due to the defect of CNTs and carbon crystallite structure, the third peak is near 1584 cm⁻¹ representing the G peak, as a result of the graphite base surface of CNTs and the PCMT wall.³⁸ From the two images, it can be seen that after laser irradiation, the strength of the 2D peak and G peak increased obviously, indicating that the graphitization degree of the CPS film was supernal.

A three-electrode system was developed to evaluate the electrochemical properties of the CPS electrodes, in which the Pt electrode and the CPS electrode were used as working electrodes, while the large area Pt electrode and the saturated calomel electrode (SCE) worked as the counter electrode and reference electrode, respectively. 25 °C phosphate-buffered saline (PBS, pH 7.4) was used as the electrolyte. The cyclic voltammetry (CV) performance of the electrodes was investigated by the electrochemical workstation (CHI 625A, Shanghai Chenhua instruments Co., Ltd.). The scan rate was 50 mV/s, and the scanning potential window was −0.6 V to 0.8 V (vs SCE). We can find from Fig. 3 that the cyclic voltammetry (CV) curve of the Pt electrode was a straight line within a small current range. In the case of the CPS film, the area of the electrode’s CV curve was strongly increased, indicating that the charge capacity of the CPS electrode was significantly higher than that of the Pt electrode. In addition to a small peak at −0.4 V, which is probably due to the decrease in adsorbed oxygen,³⁹ the CV curve of the CPS electrode has no apparent redox peak, revealing that the current was transmitted mainly through charging and discharging of the interfacial double layer, which improved the stability of the electrode. The cathodic charge storage capacity (CSCc) of the CPS electrode was ∼864 mC/cm², which was extremely higher than that of the Pt electrode (1.4 mC/cm²). Perhaps, this could be attributed to the fact that the CPS film has a longitudinal porous structure that holds more electrochemical substances and has a larger electrochemical surface area.

The electrochemical impedance spectroscopy (EIS) Bode plot and Nyquist plot of the Pt electrode and the CPS electrode are shown in Fig. 3(b). The Pt electrode impedance value (|Z| value in the lower figure) decreased rapidly vs the frequency. However, in the case of the CPS film, the impedance values remained constant around 222 Ω in the high frequency region (10–10⁵ Hz), and much lower than that of the Pt electrode, demonstrating an increase in the conductivity of the CPS electrodes. We inferred that this may be related to the formation of the conductive network structure of the CNTs and the around PCMTs through laser irradiation. We compared the impedance values of each electrode at 10³ Hz EIS, which is a typical neuronal action potential frequency. The impedance of the CPS electrode was only 5% of that of the Pt electrode, proving that the conductivity of the CPS electrode was significantly enhanced. The Nyquist plot of EIS is shown in Fig. 3(c), in the low frequency region (0–120 Hz), the Zimag values on both electrodes increased vs the frequency. It is noted that the trend of the CPS electrode had a bigger slope than that of the Pt electrode, also implying that the capacitive performance of the CPS electrode is much closer to the ideal state.⁴⁰ 

The maximum safe charge injection limit (Q_inj) represents the ability of the electrode to transmit the charge under the condition of a rapid pulse current and is an important indicator of the stimulating electrode. Potential transient measurements were applied to detect Q_inj of the CPS electrode.²² Q_inj of the CPS electrode was finally acquired at about 5.7 mC/cm², which was much higher than that of the Pt electrode (∼0.15 mC/cm²) and also the commonly used oxidizing crucible electrode (∼4.0 mC/cm²).¹⁷ The high Q_inj value indicated that the electrode had a lower polarization potential while the current pulse passing, which was conducive to enhancing the electrochemical stability of the electrode. This was consistent with the results of cyclic voltammetry and electrochemical impedance analysis, proving that the CPS composite film is an ideal electrode material.

In summary, we fabricated a novel composite electrode by assembling carbon nanotubes with silk fibroin, and the structure of CNTs/SF was reconstructed and carbonized by laser irradiation. We also characterized the electrochemical and constructive properties of our electrode in detail. The electrode was mainly charged and discharged by a double-layer capacitor, and the cathodic charge storage capacity (CSCc) reached 864 mC/cm². At 10³ Hz of EIS, the CPS electrode impedance was only 5.6% of that of the Pt electrode, while the Q_inj of the electrode was 5.7 mC/cm². The electrode in this work exhibited prominent conductive and good capacitive properties, so we believe that it could be an ideal implanted electrode alternative.

This work was financially supported by the Key Scientific and Technological Projects of Beijing Education Commission (Grant No. KZ201910005009) and the Beijing Municipal Natural Science Foundation (Grant No. 4194071).

The data that support the findings of this study are available from the corresponding author upon reasonable request.