COATING STAINLESS STEEL ELECTRODES WITH IRIDIUM FOR APPLICATIONS IN FUNCTIONAL ELECTRICAL STIMULATION

James S. Walter, Lisa Riedy, Paul Zaszczurynski, Stuart F. Cogan*, Ying-Ping Liu* Rehabilitation Research and Development Center, Hines VA Hospital, Hines IL, USA; EIC Laboratories, Norwood Mass. USA

ABSTRACT

Iridium coated single strand 316LVM wire was evaluated for corrosion resistance under high charge injection conditions. Both positive- and negative-first charge injection were conducted for test periods of approximately 240 hours. Charge balanced pulsing was conducted at 60 pps with 100 or 200  _s pulse duration. Charge injections densities ranged from 80 to 1280 _C/cm2. Electrical transients were periodically obtained to evaluate the stability of the electrode. Optical and scanning electron microscopy evaluation of the electrodes was also conducted. Corrosion was not apparent with any of the injection protocols as viewed with light microscopy. However, scanning electron microscopy revealed some surface disruptions and corrosion at the highest charge injection densities. The results indicate that the iridium coated 316 LVM electrodes can be used with considerably higher charge injection densities than uncoated electrodes.

INTRODUCTION

The longevity of an FES electrode system is a major concern for clinical applications of chronic neural prosthetics (2,3). Depending on the application, the stimulation demands on the implanted electrodes will vary but the overall goal remains the long term avoidance of corrosion and adverse tissue reaction. Stainless steel 316LVM electrodes are commonly used in neuroprosthetic applications due to their general good corrosion resistance, adequate mechanical strength, and ease of manufacturing. As an electrode material, 316LVM stainless steel may be susceptible to pitting corrosion under certain conditions which are initiated by the breakdown of the passivation layer (1,2,5). The maximum limit of safe charge injection is generally reported to be 40 _C/cm2 and only with cathodal-first pulsing. However, we have observed tarnishing with balanced wave forms using anodic- and cathodic-first charge injections of 20 _C/cm2 after one year of continuous pulsing (4). As a result of the limited availability of high charge injection electrodes for bipolar stimulation, we have evaluated coatings of sputtered iridium films on 316 LVM substrate wires. Results for both anodic- and cathodic-first pulsing over a wide range of charge injection densities are presented for ten days of continuous pulsing. The films were periodically characterized with electrical transients and light and electron microscopy.

METHODS

Wire electrodes of 316 LVM stainless steel (7 mil) were coated with Ir metal by DC magnetron sputtering. A thin film of titanium was sputtered onto the 316 LVM as an adhesion layer prior to iridium deposition. The sputtering was accomplished with a DC argon plasma at a pressure of 10 millitorr for the Ti deposition and 22 millitorr for the Ir deposition. For pulse studies, coated wire was insulated with #20 thermoplastic polyester heat shrink tubing such that the final exposed length was 5 mm. Electrodes were pulsed in a bicarbon-ate buffered electrolyte solution purged with 5% CO2/6% O2 gas to a pH of 7.4 representative of interstitial fluid [1,4,5]. A stimulator was used to control a stimulus pulse amplifier of our own construction that provided square wave pulses. Charge balanced pulses were obtained with a 0.47 _F capacitor in series with the stimulating electrodes. The capacitor was discharged during interpulse periods through a 1.8K_ shunt resistor across the output of the stimulator. Monophasic anodic- and cathodic-first current pulses were applied in a 3-electrode test cell comprised of the test electrode, a large surface area Pt counter electrode and a saturated calomel reference electrode (SCE) in close proximity to the test electrode. Pulsing at each charge injection level was conducted continuously at a rate of 60 pps for 10 days. Different pulse durations were used because of the limited current available from our stimulator. A 100 _s pulse duration was used for charge injection densities of 80, 160 and 320_C/cm2; a 200_s pulse duration was used for 640_C/cm2; and a 400_s pulse duration was used for 1280_C/cm2. A 100 MHz storage oscilloscope with 10 M_ impedance was used for recording electrical transients as the potential between the test electrode and the SCE electrode. The current was also recorded using a 100 _ resister in series with the test electrode. The access resistance (Ra) was obtained from the voltage transient by dividing the access voltage (Eacc) by the instantaneous current. During the pulse the maximum excursion potential [Eex] was obtained by subtracting the access voltage (Eacc) from the maximum voltage vs SCE. The commonly used maximum potential (Emax) was obtained by adding the Eex to the interpulse potential [Eipp]. The interpulse potential (Eipp) was measured just prior to the onset of the current pulse. The Eipp values were found to be identical to the electrode potential measured with an electrometry immediately after the pulsing was interrupted. Electrodes were soaked in distilled water to remove salt crystals before evaluation under the light microscope. Scanning electron microscopy (SEM) of the electrode surface was conducted at the end of pulsing. The appearance of the electrodes under light and electron microscopy are described.

RESULTS

The electrical transient components, Eacc, Emax and Eipp, indicated a low impedance to charge transfer of the electrode/electrolyte interface and potential excursions. Eacc is the initial electrical transient representing the initial resistance to pulsing (the iR drop). Absolute values of Eacc for both charge injection protocols were less than 2.1 V for 80 _C/cm2 charge densities indicating an Ra of less than 84 ohms. Eacc increased to a maximum value of 6.2 V at the highest charge densities of 640- and 1280 _C/cm2; however, Ra values remained low at less than 62 ohms throughout pulsing. Emax is the maximum potential excursion during the pulse. Values of Emax were less than +0.95 V va SCE for all charge injection densities. In addition, Emax values for each electrode varied little during the 10 days of pulsing. Eipp reflects the open circuit potential of the test material relative to an SCE electrode as well as discharge characteristics of the pulsed electrodes. The open circuit potential before pulsing was positive at approximately 300 mV. Initial Eipp` values for both positive and negative charge injection protocols were positive with values less than 0.8 V. However, at the high anodic pulsing of 640_C/cm2, small but negative Eipp voltages were recorded. The relatively small voltages recorded for the electrical transients over the 10 days of pulsing were associated with little corrosion of the iridium coating. Before stimulation, electrodes had a shiny surface without surface disruptions. During pulsing observations with light microscopy continued to show a shinney surface except at the highest charge injection of 1280_C/cm2 where a single brown spot was observed during the second day of pulsing and remained a small spot during the remaining 8 days of pulsing. At the end of pulsing the electrodes were also observed with SEM. Cathodic- and anodic-first pulsing above 320 _C/cm2 revealed surface disruptions that appear to be delamination of the TiIr coating at the Ti/316LVM interface.

DISCUSSION

Our principal finding was that the iridium coated stainless steel electrodes are resistant to corrosion at charge injection densities, as high as 320 _C/cm2 for anodic and cathodic-first pulsing. Active surface disruptions indicating corrosive processes were present at higher charge injection densities. These charge injection densities are much higher than the maximum of 40 _C/cm2 for uncoated 316 LVM stainless steel electrodes [1,4]. Access resistances remained low throughout these studies from 40 to 90 ohms. These resistances are much lower than we previously reported for uncoated 316LVM stainless steel wire electrodes [4,5]. The initial access resistance during 10 days of pulsing was 136 ohms and the final resistance was 246. The Emax values are also low for all of the charge injection protocols conducted here. For example, for the iridium electrodes pulsed cathodally at 80 _C/cm2, Emax values ranged from -0.16 to +0.12 V whereas our previous report with stainless steel under similar pulsing conditions had Emax values ranging between -0.36 to -1.73 V [5; note, these values are converted from reported Eex values]. The less negative potential transients reflect the larger and more available charge injection availablewith Ir compared to 316LVM. Iridium as a charge injection site is characterized as having multiple oxidation states suitable for charge transfer. For charge densities of 320 _C/cm2 or less, the Eipp remained close to the open-circuit potential of electrodes prior to pulsing, indicating that the discharge of the capacitor during the interpulse period is effectively complete. The unremarkable values of Eipp and the similar open-circuit potential measurement observed when pulsing was interrupted also suggested that no unexpected electrochemical processes are occurring on the electrodes. The resistance to corrosion and modest potential excursions observed with charge injection may provide important improvements in electrodes for neuroprosthetic applications. These electrodes would appear to be suitable for high charge injection applications as well as bipolar stimulation. Several groups have raised the potential of applying neuroprosthetics for controlling skeletal muscles, autonomic organs and central nervous system stimulation [1,2,3,6], and iridium coated electrodes may help to provide electrodes for these applications.

REFERENCES

1. Cogan, SF, Jones, GS, Hills DV, Walter, JS, Riedy, LW. Comparison of 316LVM and MP35N alloys as charge injection electrodes. 28: 233-240, 1994.

2. Handa Y, Hoshimiya A, Iguchi I, and Oda T. (1989) Development of Percutaneous Intramuscular Electrode for Multichannel FES System. IEEE Trans. Biomed Eng. 36: 705.

3. Lobe GE. Neural Prosthetic Interfaces with the Nervous System. Trans. Neurosci. 12:195, 1989.

4. Riedy, J.S. and Walter J.S. Effects of low charge injection densities on corrosion responses of pulsed 316LVM stainless steel electrodes. IEEE Trans. Biomed Eng. Submitted.

5. Riedy, L.W., Walter, JS, Comparison of electrical transients and corrosion responses of pulsed MP35N and 316LVM electrodes. Ann.

Biomed. Eng. 22:202-211, 1994. 6. Walter JS, Wheeler JS, Cogan SF, Plishka, M, Riedy LW, Wurster RD. (1991) Evaluation of direct bladder stimulation with stainless steel woven eye electrodes J. Urol. 150:1990-1996, 1993.

ACKNOWLEDGEMENT

This work was supported by the Rehabilitation Research and Development Service of the U.S. Department of Veterans Affairs (Merit Review, B658-R)

James S. Walter, Ph.D. Hines VA Hospital Rehabilitation R&D Center (151L) P.O. Box 20 Hines, IL 60141 Phone: (708) 343-7200, Ext. 5805 Coating Stainless Steel Electrodes with Iridium for Applications in FES