Aerospace and Mechanical Engineering
Adjust Font Size: Normal Large X-Large

AEM faculty spotlight:

Ryan Elliott

As technology advances, engineers are being asked to design smaller, more efficient, and more reliable structures and devices. One strategy for achieving these goals is to replace old multiple-piece systems with new one-piece systems made from high-tech. "shape-memory materials." For example, currently the shape of a jet engine's exhaust nozzle is controlled by a system of hydraulic actuators, but if the exhaust nozzle were constructed from a shape-memory material there would be no need for the hydraulic actuators. A design with shape-memory materials and without hydraulic actuators would be more reliable (fewer failure modes) and lighter (better fuel efficiency). However, the safe use and development of high-tech. materials for the aerospace industry requires a deep understanding of their behavior. In order to obtain the necessary level of knowledge and understanding for these materials, AEM Associate Professor Ryan S. Elliott is working to develop a type of computer simulation, called first principles materials modeling, that holds the potential to simplify the discovery and custom-design of new high-tech. materials. What follows is a discussion with Professor Elliott on his techniques and what the future may bring for shape-memory materials.

Elliott
Ryan Elliott

What type of research do you do?
My work attempts to model shape-memory materials and fundamentally understand why they exhibit their surprising and unusual behavior. I am doing this by starting at the smallest scale and studying the atomistic interactions. I want to see if we can develop models that will accurately predict this type of behavior. If so, we could come up with new materials that have better properties in this way. My research group has been exploring a number of different ways to do this. The techniques we are studying - which I call branch-following and bifurcation (BFB) methods - allow you to efficiently explore how the predictions of a first principles material model change in response to adjustments of temperature or some other stimulus. It's a nice, systematic way to determine the behavior of different materials. The difficulty, of course, is trying to understand the materials and build into the model enough fundamental physics so you can get accurate behavior predictions.

Do you have an example of an unusual application of shape-memory materials on a current technology?
Every year a large number of people visit the automobile repair shop to have a rattle in their car fixed. It's been suggested that the amount of time and money you need to spend for the auto mechanic to take your car apart, find, and then tighten the loose bolt could be reduced significantly by the use of shape-memory materials. The idea is to use a shape-memory washer on the bolts that hold the car's components together and to connect it to electrical leads so that if the nut starts rattling, it can be detected. Then, that particular shape-memory washer can be heated so that it automatically tightens the bolt. This could be done without ever having to disassemble any part of the car.

What initially excited you about this field?
The first thing is the unusual behavior of these materials. The examples that you can show, the demonstrations you can do are really intriguing. For instance, if you have a piece of this material, say a wire, and you wrap it around your finger, it forms a spring shape, as would a piece of steel (such as a paper clip). But if you drop this spiraled shape-memory wire into a cup of hot water, it will actually straighten out, like before you deformed it. It is a very unintuitive thing; it is not something we encounter in our day-to-day lives. I definitely want to understand why this behavior occurs and how it happens, all in terms of basic principles of physics. Once I have a good idea of the how and why, then I can build a material model based on that idea.

What is a model?
A material model is a set of equations or a computer program for which I can specify the "operating conditions" (e.g., the temperature, force, torque, etc.) and given these conditions the model returns its prediction of the material's response (e.g., its shape, hardness, toughness, etc.). There are a lot of input parameters, like temperature, that I could specify, and the model will give different responses for different sets of input. It's too big a problem to feed the model every possible set of data and look at the outputs. The BFB methods I have developed are a systematic way to scan through this huge set of input data in order to pull out the information we are most interested in and to ignore the superfluous data. A good model equals good response predictions for real materials.

Is there a particular metal of interest right now?
The shape-memory alloy nickel titanium is currently of great interest to academia, government, and industry. It is used extensively in the medical-device industry to make vascular stents and surgical tools. Stents are little wire meshes wrapped into a cylindrical tube. Making stents from shape-memory alloys allows, at body temperature, for it to be completely crushed, put into the body, and then to regain its original tube shape. The shape-memory material is useful because it is biocompatible, can regain its shape and also can deal with cyclic deformation associated with the body's circulatory pulse. It has moderately good fatigue life, but it is not perfect. Improvements to all of these properties would lead to wider and more successful medical applications of shape-memory materials and, consequently, a healthier population. My research aims to use first principles materials models to discover and design materials with these improved properties.

What does the future hold for you in terms of research?
In the immediate future I am going to be continuing the development and improvement of my BFB techniques. They need to be automated so that we can run large simulations without a lot of user interaction. I will continue modeling materials and working to predict shape-memory material behavior. The longer-term goal would be to develop a tool that could be made available to not only researchers in academia but those in industry. They would use this tool to help understand and design new materials. The modeling tools would be able to predict the behavior of new materials before they are built in the lab.


Last Modified: Wednesday, 05-Oct-2011 13:19:42 CDT -- this is in International Standard Date and Time Notation