John Stout and Dr. Aaron Hawkins, Department of Electrical Engineering
The silver flash project is an effort to create a reliable, MEMs based memory device. As a joint electrical and mechanical structure these devices offer several advantages over the traditional NAND flash memory currently found in phones, mp3 players, and USB memory drives. The silver flash project uses standard silicon-processing materials and processes to build a device that should have a long lifespan and continue to function under extreme conditions.
The first step in completing the silver flash project was to design the physical structure of the device. The device was to be fabricated on top of a silicon wafer using standard microfabrication processes. A total of five separate layers were required to leave the precise patterns of the electrodes, the hollow chamber that would later contain silver nanoparticles, and the holes in the protective top layer of glass to make contact with the electrodes.
The fabrication of the devices was completed by introducing a solution with silver nanoparticles into the channel and letting the solution dry out, leaving behind the silver. After the solution had evaporated the channels would be full of silver nanoparticles. After the hollow channels contained enough silver particles the electrical tests would begin. Two different solutions have been tried, and a third set of tests is in progress.
The objective of each test was to show that the devices were programmable; that is that there were two stable states for the device and that the state of the device could be reliably changed and detected. Ideally, the devices would have an infinite resistance between the two main electrodes when the device was in the ‘off’ state, and a resistance of zero when the device was turned ‘on’. The state would be set by applying a voltage to a ‘gate’ electrode that would be situated above, but not touching, the two main electrodes. The tests would also need to test how permanent the state of the device was. The results from these tests changed in response to the different silver solutions that were tested.
The three different solutions that were used in testing yielded very distinct results. The first solution was one with silver nanoparticles in a citrate solution. No switching was observable when this solution was use. After some research we decided to try a colloidal silver mixture. We ordered a small bottle of some cheap colloidal silver. This bottle had a very high concentration of silver, 1000 ppm. Later, we discovered that the reason that it could come in such a high concentration was because there was protein mixed in with the silver. The protein that was introduced into the channels with the silver changed the nature of the device from a mostly resistive device to a mostly capacitive one. We saw that the state of the devices could reliably be changed, from one capacitance to another.
We are in the process of testing the devices with a solution of pure silver in pure water. This solution should provide the change in conductance that we are looking for. After we have proved that the devices function we have a number of modifications that we would like to make to improve the device. Some of the changes will be to modify the layout of the devices to make testing them easier. Other ideas we have been developing pertain to improving the behavior of the device. One of our recent theories about the device is a way to simplify it. We believe that the device will function if we use the upper electrodes as the gate, thus eliminating one electrical contact. This theory seems to be consistant with the material taught in the MEMs class here at BYU.
This project has helped me see the process of research and appreciate the amount of work that it requires. I have appreciated the opportunity of doing research as an undergraduate and I look forward toward more research opportunities and further work on this project as a graduate student next year.