Poster Z9 - Accelerated Long-term Chemical Stability Evaluation of Multichannel Thin-film Implants

Introduction: Approximately 1 billion people globally are affected by neurodegenerative diseases (WHO, 2005). In response, neural engineering labs are working to fabricate thin-film implants that will record and stimulate neural signals as a means for researching new treatment methods. To find the best suited device for the neural tissue environment, different device compositions and materials need to be tested and compared in an environment mimicking the conditions in neural tissue. Testing in vivo takes significant time and resources and poses risk for failure, potentially causing damage or loss to either the device or its carrier. For a more practical and low-risk method to save time, money and reduce the use of animal subjects, researchers are utilizing a method called reactive accelerated aging. This method artificially accelerates the degradation of devices by using naturally harmful accelerant factors, to mimic failure mechanisms of devices implanted in vivo. Evaluating the performance of these devices in this way, allows us to study the predicted in vivo behavior without risking the in vivo environment.

Materials and

Methods: Using reactive accelerated aging, this study tests the overall fabrication quality and durability of two types of neural interface devices: type 1 with plain gold microelectrodes and type 2 with gold microelectrodes and an added iridium oxide (IrOx) coating. Each device type has two substrate types, one insulated with a solely polyimide (PI) substrate and one with an amorphous silicon carbide (a-SiC) layer in addition to the polyimide substrate. Similar a-SiC thin film devices were shown to improve electrical insulation and liquid ingress resistance (Deku et al, 2018) over PI only devices. Initial device measurements were recorded in phosphate buffered saline (PBS) at room temperature, then devices were assigned to one of three test conditions for accelerated aging: at room temperature (all devices), at 37°C (all devices), then at 67°C for PI only devices, or at 87°C for PI/a-SiC devices. Aging tests were performed in an Isotherm oven with the electrodes submerged in PBS. At 14 day intervals, impedance and charge storage capacity of each electrode were recorded until device failure. The expectation is that heat is a key accelerated aging factor, with higher temperatures leading to faster device failure. Our goal is to find potential failure mechanisms in our devices early, and facilitate fabrication of revised devices with improved performance and longevity to remain effective in vivo for a patient's lifetime.

Results, Conclusions, and Discussions: Our first experiments showed that devices with a layer of a-SiC in addition to PI were less susceptible to failure than devices with solely polyimide substrate. For polyimide-only devices, the polyimide substrate layers delaminated from internal metal layers during aging in the 37°C and 67°C ovens and the electrode materials were degraded. The added layer of a-SiC:H indeed improved resistance to liquid ingress and reduced delamination and changes in impedance over time.However, we did observe sporadic fractures of the a-SiC layers, and future work plans to explore the origins of these fractures in more detail. Additionally, devices with IrOx deposited onto the electrode sites had lower impedance and higher charge storage capacity than those with gold only.
Apart from temperature, other factors within the neural tissue environment contribute to failure mechanisms, such as varying chemical concentrations, electrical activity, and ROS presence. And so, additional factors need to be considered to ensure a material has maximum biocompatibility, especially for different characteristics like device charge-transfer properties, fluid ingress resistance, flexibility, size, etc. For the next steps in this project, we plan to explore the combined use of electrical pulsing, chemicals like hydrogen peroxide (a source of ROS), and other strategies in order to improve our accelerated aging experimental conditions and determine the best aging techniques for mimicking the neural tissue environment. Expanding in accelerated aging methods will allow for research to effectively test different fabrication methods and materials to determine the most suitable designs for any targeted therapeutic use.

Acknowledgements (Optional): I would like to thank CURE, BMEs and the Deku Lab for this extraordinary experience.