Recent earthquake damage has exposed the vulnerability ofexisting structures to strong ground movement. At the Georgia Institute ofTechnology, researchers are analyzing shape-memory alloys for their potentialuse in constructing seismic-resistant structures.

“Shape-memory alloys exhibit unique characteristics that youwould want for earthquake-resistant building and bridge design and retrofitapplications: they have the ability to dissipate significant energy withoutsignificant degradation or permanent deformation,” said Reginald DesRoches, a professorin the School of Civil and Environmental Engineering at Georgia Tech.

Georgia Tech researchers have developed a model thatcombines thermodynamics and mechanical equations to assess what happens whenshape-memory alloys are subjected to loading from strong motion. The researchersare using the model to analyze how shape-memory alloys in a variety ofcomponents -- cables, bars, plates and helical springs -- respond to different loadingconditions. From that information, they can determine the optimalcharacteristics of the material for earthquake applications.

The model was developed by DesRoches, School of MechanicalEngineering graduate student Reza Mirzaeifar, School of Civil and EnvironmentalEngineering associate professor Arash Yavari, and School of Mechanical Engineeringand School of Materials Science and Engineering professor Ken Gall.

A paper describing the thermo-mechanical model was publishedonline Feb. 3 in the InternationalJournal of Non-Linear Mechanics. This research was supported by theTransportation Research Board IDEA program.

To improve the performance of structures during earthquakes,researchers around the world have been investigating the use of “smart”materials, such as shape-memory alloys, which can bounce back afterexperiencing large loads. The most common shape-memory alloys are made of metalmixtures containing copper-zinc-aluminum-nickel, copper-aluminum-nickel ornickel-titanium. Potential applications of shape-memory alloys in bridge andbuilding structures include their use in bearings, columns and beams, orconnecting elements between beams and columns. But before this class ofmaterials can be used, the effect of extreme and repetitive loads on thesematerials must be thoroughly examined.

“For standard civil engineering materials, you can usemechanics to look at force and displacement to measure stress and strain, butfor this class of shape-memory alloys that changes properties when it undergoesloading and unloading, you have to consider thermodynamics and mechanics,” explainedYavari.

The Georgia Tech team found that the generation andabsorption of heat during loading and unloading caused a temperature gradientin shape-memory alloys, which caused a non-uniform stress distribution in thematerial even when the strain was uniform.

“Shape-memory alloys previously examined in detail werereally thin wires, which can exchange heat with the ambient environment rapidlyand no temperature change is seen,” said Mirzaeifar. “When you start to examinealloys in components large enough to be used in civil engineering applications,the internal temperature is no longer uniform and needs to be taken intoaccount.”

To predict the internal temperature distribution ofshape-memory alloys under loading-unloading cycles, which could then be used todetermine the stress distribution, the researchers developed a model that usedthe surface thermal boundary conditions, diameter and loading rate of the alloyas inputs.

The team included ambient conditions in the model becauseshape-memory alloys for seismic applications could operate in a variety ofenvironments -- such as water if used in bridge structures or air if used inbuilding structures -- which would produce different rates of heat transfer. Theresearchers used a thermal camera to record the variation in surfacetemperature of shape-memory alloys experiencing loading and unloading.

Using their model, the researchers were able to accuratelypredict internal temperature and stress distributions for shape-memory alloys. Themodel results were verified with experimental tests. In one test, they foundthat a shape-memory alloy loaded at a very slow rate had time to exchange theheat created with the ambient environment and exhibited uniform stress. If it wasloaded very rapidly, it did not have enough time to exchange the heat, leadingto a non-uniform stress distribution.

“Our analytical solutions are exact, fast and capable of simulatingthe complicated coupled thermo-mechanical response of shape-memory alloysconsidering temperature changes and loading rate dependency,” said Mirzaeifar.

In future work, the researchers plan to examine morecomplicated shapes and the effects of combination loading -- tension, bendingand torsion -- to optimize shape-memory alloys for earthquake applications.

This project issupported by the Transportation Research Board of the National Academies (AwardNo. NCHRP-147). The National Academies has rights to the data and the contentis solely the responsibility of the principal investigators and does notnecessarily represent the official views of the National Academies.

Research News & Publications Office
Georgia Institute of Technology
75 Fifth Street, N.W., Suite 314
Atlanta, Georgia 30308 USA

Media RelationsContacts: Abby Robinson (abby@innovate.gatech.edu; 404-385-3364) or JohnToon (jtoon@gatech.edu; 404-894-6986)

Writer: AbbyRobinson