Date of Award

Spring 2007

Document Type


Degree Name

Doctor of Philosophy (PhD)


Micro and Nanoscale Systems

First Advisor

Robert Szlavik


There have been numerous studies presented in the literature demonstrating proof of principle neural-electronic circuitry. Some of these studies involve simulations of neural detection using synthetic electronic circuitry, while others involve simulations of neural excitation using external electronics. A common feature of these studies is the simplicity of the overall circuit topology. Some of these studies implement the circuit equations in conventional numerical ordinary differential equation solvers. This process involves the algebraic manipulation of the circuit equations which is a tedious process for all but the simplest circuit topologies. As the overall complexity of the network topology increases, the numerical solver approach quickly becomes intractable necessitating an alternate implementation strategy. SPICE implementations of the Hodgkin-Huxley neuron model have sought to remedy this problem. There have been multiple studies associated with implementing the Hodgkin-Huxley model in the open source circuit simulator, SPICE. In this dissertation, a novel implementation of a portable SPICE device model developed using the Hodgkin-Huxley active membrane model is implemented using the code-level modeling functionality of an open source version of SPICE. The model is validated by comparison with standard Hodgkin-Huxley model simulations including gating variable dynamics simulations, accommodation, anodebreak excitation, and others. A further validation study is carried out demonstrating two blocking phenomenon described in the literature. The device model fully parameterizes the Hodgkin-Huxley membrane model to include temperature, internal and external concentrations used in the Nernst equations, and other user specified parameter values. This parameterization allows for making changes to the underlying neuron model rapidly and with minimal implementation complexity.

The novelty and robustness of the modeling approach described herein is based on the ease of implementation. A wide variety of active membranes can be simulated using this code model approach. These biologically realistic components can be integrated with artificial electronic components allowing for the simulation of hybrid neuralelectronic circuitry under the SPICE simulation platform. These types of hybrid circuit simulations are not currently achievable using other neural simulators such as NEURON or GENESIS. While this implementation uses the Hodgkin-Huxley neuron model with its known limitations, the process of developing the device model can be used to implement any neuron model which can be described mathematically.