• Carbon fibers
• manufacturing process
• Micro Electrodes
• Neural Interface
• PEDOT:PSS

Introduction:
The autonomic nervous system (ANS) is crucial for maintaining the body's internal balance. Afferent nerve fibers detect external and internal stimuli, transmitting this information to the central nervous system (CNS). Efferent nerve fibers transmit signals from the CNS, eliciting a response. ANS dysfunction can lead to imbalanced cytokine production, contributing to conditions such as depression, anorexia, psoriasis, pain, arthritis, organ failure, and tissue damage , . Currently, drugs aim to inhibit or neutralize endogenous cytokine production, impacting the vagus nerve's inflammatory reflex. However, the pharmacological treatment lacks precision, affecting the entire body and potentially causing collateral effects. Bioelectronic Medicine (BM) employs implantable devices for targeted electrical impulses, modulating the peripheral nervous system for therapeutic benefits. By selectively regulating the ANS, we can counteract the damaging effects of disease-associated dysfunctional mechanisms. This involves implanting microdevices, and neural interfaces, into specific peripheral nerves, decoding, and modulating nerve signals. Neural interfaces vary based on selectivity. This refers to the ability to target specific nerve fibers without causing off-target effects. They're classified into extra-neural (less invasive with lower selectivity) and intraneural (more invasive, potentially causing a foreign body reaction but with higher selectivity).
Objective: Spike Cuff Electrode (SpiCE) is a new neural interface, able to target small peripheral nerves (diameter below 1 mm) with high selectivity but less invasiveness than solutions in the literature. It consists of a flexible polymeric structure that wraps around the nerve (Cuff, made of polyimide) and a 3D rigid structure that penetrates the epineurium to reach the most central regions of the nerve. The 3D structure consists of two different components: the Spines and the Links. The former refers to a rigid and biocompatible resin structure designed to secure the Link within the active site of the cuff, aiding in its retention and facilitating penetration into the nerve. It is secured to the Cuff by means of two anchoring feet inserted within the polyimide layer. The Link is a composite material that connects the Cuff’s active site to the inner part of the nerve.
The primary objective of this study was to develop a fabrication method for the Link that met three essential criteria: 1) the ability to penetrate the nerve while maintaining a small size to reduce FBR 2) enabling proper conductive properties to allow stimulation and recording, and 3) increase neural selectivity to target the fibers of interest. The chosen materials included carbon fibers (CFs), poly(3,4-ethylenedioxythiophene) polystyrene sulfonate (PEDOT: PSS), and polydimethylsiloxane (PDMS). CFs are used as the Link cores and have a small cross-sectional area (6-10μm per fiber), which reduces adverse tissue responses by eliminating glial scar formation. A custom solution of PEDOT: PSS and DMSO was used as an intermediate layer to coat the carbon fiber bundle and improve its conductive and mechanical properties. Finally, a film of PDMS, an elastomeric and soft material, was deposited as the outermost layer, to isolate the conductive bundle without altering its mechanical characteristics dictated by the presence of CFs. The materials for the fabrication of the Link were selected according to a structural analysis by Finite Element Modeling (FEM) of the SpiCE electrode. Then, the fabrication process of the Link was carried out. Finally, the Link was electrochemically characterized by performing cyclic voltammetry (CV) and electrochemical impedance spectroscopy (EIS) tests.
Methods:
A structural feasibility study of the Spine was carried out in Comsol®, and two FEM analyses were carried out to evaluate the resistance of the Spine’s anchoring feet to the Cuff when subjected to relative movement and implantation. First, we evaluated a 20µm relative motion between the Cuff and the inner part of the nerve. Then, we simulated the bending model, which considers the manipulation of the device during the implantation. In this case, a nominal force of 4-6N is applied on the side wall of the polyimide. The materials used and their mechanical properties (Young's modulus, Poisson’s ratio, and density) are listed as follows: 1) Polyimide (Cuff): 1MPa, 0.3, and 1100kg/m^3 2) Carbon Fiber (as Link): 240GPa, 0.3, and 1800kg/m^3 3) IP-S (Spine): 5.2GPa, 0.3, 1100 kg/m^3 4) Neural tissue: 96Pa, 0.49, 1000kg/m^3 . Von Mises stress was used to evaluate the stress on the anchor feet.The next project step focused on the Link fabrication, developing a process for consistently producing uniform bundles. Initially, CFs were individually picked up using tweezers and a custom-designed (Fusion360, Autodesk®, California, United States) structure to group them in bundles. The approach was subsequently updated to utilize 33-gauge needles (inner diameter: 108µm) as a supporting structure, forming bundles of carbon fibers with a diameter of 80-100µm. This adjustment streamlined the sorting process by enabling the selection of entire bundles instead of individual fibers with tweezers.
The previously obtained bundles were placed within 100µm, 120µm, and 200µm channels. The smaller channels (100µm, 120µm) were fabricated using PDMS using a molding technique, while the bigger ones were made of polytetrafluoroethylene (PTFE)and are commercial tubing.
The PEDOT:PSS solution with and without DMSO was injected into the microchannels and freeze-dried, inside the microchannels, to observe how the presence of DMSO changed the morphology of the material. The swelling in PBS of PEDOT: PSS +DMSO 3%(w/w) and PEDOT: PSS 3%(w/w) obtained by casting and freeze-drying in a 48-hour time interval was studied. Furthermore, it was observed that pristine PEDOT: PSS at 3% (w/w) in PBS degraded after 2h.
Subsequently, material deposition tests were conducted on the CFs using the various microchannels created. The deposition was carried out using two distinct methods: freeze-drying and casting. PEDOT: PSS double-coated bundles were treated with a 90-minute annealing process at 130°C. Finally, it was chosen to carry out a deposition of PDMS (Sylgard 184, Dow®, Michigan USA) to obtain an external layer to isolate the lateral surface and make only the bundle section conductive. Lastly, CV and EIS were performed for electrochemical characterization with the three-electrode configuration, using a Platinum (Pt) electrode as the counter electrode, a silver|silver-chloride (Ag|AgCl) electrode as the reference electrode, and the Link and a reference sample consisting of CFs coated in PDMS as the working electrode.
In the CV, two types of scan rates (100mV/s and 50mV/s) were performed in a window between -0.6V and 0.8V. From these data, the cathodic charge storage capacity (cCSC) was calculated by calculating the negative area within the curve and then considering only the cathodic current. EIS was performed from 1Hz to 1MHz with 5mV in AC. Then, a statistical comparison was made between the Links and a PDMS-coated CFs bundle by comparing their cCSC and impedance modulus at 1kHz. Tests were conducted to check the gaussianity of the data, followed by a t-test for Gaussian samples or a Mann-Whitney test for non-Gaussian ones.
Results:
In the structural analysis models, the von Mises stress on the anchoring feet was calculated. It was chosen as a reference value because it is an equivalent stress measure, considers both normal and tangential stresses, and is supported by the yield stress of IPS-S: Sy=41 ± 16 MPa. For the verification the worst case of the yield stress 25MPa was chosen. The safety coefficient (CS) of CS=146100 was obtained for the first model, while a value of CS=1.190 was calculated for the second model.
Regarding the devised fabrication strategy, double deposition of PEDOT: PSS+DMSO by casting was chosen as the manufacturing process. This deposition method allowed the fabrication of samples with good swelling properties, and it was possible to wrap the fibers without generating a rough surface.
As the final step in the process, the bundle was dipped in PDMS three times to achieve an insulating outer layer of about 20±6 µm.
Furthermore, it has been proven that the addition of PEDOT: PSS significantly decreases the module impedance compared to a control sample, consisting of CFs and PDMS (on 7 samples at 1kHz for the Links we get a value of 1.79 ± 1.314 kΩ, while for the control samples 20.67 ± 10.6kΩ, p-value = 8.2271e-05). The addition of PEDOT: PSS was shown to significantly reduce module impedance compared with a control sample (p-value = 8.2271e-05), resulting in a more resistive behavior compared to the Link (p = 7.4745e-06).
Conclusions:
The Link fabrication process was defined, and a 100µm diameter bundle was obtained consisting of a CFs core and two layers: a conductive one (PEDOT:PSS) and an insulating one (PDMS). Finally, an electrochemical characterization was carried out. The results from the fabrication strategy and the electrochemical characterization highlight the effectiveness of our approach in creating bundles of CFs. However, we will have to increase the number of samples (n=7) to be tested to increase the robustness of our method.