Date of Award

Spring 2007

Document Type

Thesis

Degree Name

Master of Science in Electrical Engineering - (M.S.)

Department

Electrical and Computer Engineering

First Advisor

Mesut Sahin

Second Advisor

Durgamadhab Misra

Third Advisor

Dentcho V. Ivanov

Abstract

Present day neural prostheses require electrodes with high current densities. Traditional electrodes are not capable of delivering such high current densities. Titanium nitride as an electrode material and other techniques like reactive ion etching, platinization of titanium were studied in this thesis towards improving charge density of electrodes.

Titanium nitride (TiN) was sputtered in a custom designed pattern of electrodes with silicon as substrate, at a deposition rate of 2 A°/sec. Atomic Force Microscopy (AFM) analysis of TiN film showed a smooth surface for a film thickness of 1 µm. X-Ray Diffraction (XRD) analysis of the film showed the presence of TiN and Ti on the substrate.

Reactive Ion Etching (RIB) of the electrode surface with CF4 and SF6 for different combination of chamber parameters gave a peak CIC of 65.2 µC/cm2. Platinization of Ti in chloroplatinic acid (H2PtCl6) provided a maximum Charge Injection Capacity (CIC) of 2.6mC/cm2.

Electrodes made at University of Michigan were used as reference for all measurements conducted on NJIT patterned electrodes. Other methods to investigate CIC dependencies showed that CIC is not scalable with size, although CIC is calculated per unit surface area. Large surface area electrodes (4000µm2) had higher CIC per unit surface area and it decreased for smaller electrodes (1 250µm2 and 1 77µm2). Electrodes tested within the water window of hydrolysis showed CIC was dependent on bias voltage and pulse width.

An extended voltage limit in the cathodic cycle increased CIC of TiN coated electrodes significantly. The maximum injectable charge was 4.45mC/cm2 for a bias voltage of-O.8V.

It can be concluded that electrodes with rough surface had higher charge injection capacity and the charge injection capacity dependencies show that simple elements are not enough for modeling the electrode-electrolyte interface.

Share

COinS