Power law creep

Nanoindentation creep

Thin films and surface layers Nanoindentation near edges negative stiffness          

Nanoscale fracture

Bulk Metallic Glasses

Corrosion of Fasteners Creep fatigue of solders Mechanical properties and chemistry of wood cell walls         HOME PAGE
Bulk metallic glasses

Plastic deformation in BMGs is poorly understood.  Current models of plastic deformation hold that the quanta of deformation are shear transformation zones (STZs), collections of 10-100 atoms that undergo shear under applied stress. Macroscopic deformation takes place as indivudual STZs link together in a process of percolation across large distances.  Our group studies STZs and plasticity using a combined approach of mechanical measurements and atomic structural characterization.

We develop nanoindentation techniques to characterize deformation mechanisms in BMGs. Graduate student Jon Puthoff implemented broadband nanoindentation creep (BNC) as a method for obtaining the hardness-strain rate properties across 5-6 orders of magnitude of strain rate. BNC offers the potential to characterize the activation barrier for deformation, which is likely to be sensitive to medium range order (MOR) in the alloys. We work in collaboration with Paul Voyles' group, which characterizes MRO using fluctuation electron microscopy.



Amorphous metals

Bouncing ball bearings and the coefficient of restitution

Glassy and crystalline regions in a mixture of bubbles

Plastic deformation



Corrosion of fasteners in wood

Some woods like cedar and black locust are inherently decay resistant and can be used outdoors where they stand up to attack from fungi and insects. Decay-resistance is imparted by naturally occuring compounds inside the wood. Most other kinds of wood, when used in exterior applications, must be protected, and often this done with preservatives that penetrate into the structure of the wood to help it to resist decay.

Waterborne preservatives are usually comprised of metallic or organometallic salts which serve to protect the wood; but they also increase corrosion for metals embedded in or in contact with the treated wood. If sufficiently rapid this corrosion can be a problem for fasteners like nails and screws used to hold preserved wood structures together. The corrosiveness of waterborne preservatives has been documented in the literature for well over 80 years, and has not been regarded as a problem because Chromated Copper Arsenate (CCA), the major wood preservative of the past 50 years, did not greatly accelerate corrosion of metals in contact with wood, and CCA had a long track record of durable service.

With the voluntary withdrawal of wood treated with CCA for use in residential construction in January 2004, the corrosiveness of treated wood has become a concern because the replacements, which include Copper Azole (CuAz) and Alkaline Copper Quaternary (ACQ), are more corrosive than CCA. Just how much more corrosive they are, and the mechanisms by which corrosion takes place, remain a subject of scientific study.

Student Sam Zelinka works at the USDA Forest products laboratory, where he has been studying corrosion of fasteners in treated and untreated wood. He is trying to learn about the mechanisms of corrosion including identifying the relevant reduction reactions and rate-controlling kinetic steps. Sam is also developing protocols for using acceleration methods like electrochemical testing to help predict how long different fasteners will last in wood and to help assess the relative corrosiveness of the preservatives. Sam has been able to demonstrate that wood "coffees" made by seeping treated wood sawdust in water often demonstrate the same level of corrosiveness as the wood itself. Sam is currently in the process of creating synthetic coffees to identify the roles played by the different chemical components present in the coffees. Ultimately, Sam's goals include to help create less corrosive preservatives, design more corrosion-resistant fasteners, and construct corrosion mechanism maps to help predict the service lives of fasteners.



Anthocyanins as PH Indicators


Wood Handbook



Water and electrical conductivity of wood


For most materials temperature is the most important environmental variable: properties like strength and electrical conductivity vary enormously depending on temperature. But in wood, water replaces temperature as the most important environmental variable. Water greatly affects the strength of wood, and its presence or absence in wood can change over many orders of magnitude things like electrical conduction and the rate of decay. Electrical conductivity measurements are a standard method for assessing water content of wood because electrical conductivity is so sensitive to the presence of water.

Our interest in water in wood stems from ionic transport in connection with corrosion of fasteners (nails, screws) in treated wood. In our studies we have had to reassess the long-standing models for electrical conduction in wood. Interestingly, the first models used to describe electrical conduction in cellulosic materials, developed about a century ago, were much like the one we propose. Those earlier models involved "conduction pathways through free water." This sounds very much like our present percolation model; but at the time those models were invented the mathematical tools (i.e., percolation theory) necessary to fully advance them were yet to be invented. The models were ultimately abandoned in the middle of the 20th century.

Student Sam Zelinka has re-evaluated the theory of conduction in wood by employing a percolation model to describe electrical conduction in terms of overlapping paths of loosely bound or capillary water (Type II water). Given the importance of water to a variety of phenomena like decay and strength in wood, this new model has wide ranging implications.




Electrical Conductivity



Wood Handbook


Ionic conductivity

Nanoindentation creep

Nanoindentation is the most commonly used method for characterizing mechanical properties at the micrometer length scale. But most nanoindentation measurements are limited to the more basic mechanical property indices like hardness and modulus.

We attempt to extend nanoindentation beyond those kinds of measurements into a kind of mechanical spectroscopicopy probe capable of distinguishing deformation mechanisms. Using finite element analysis, Abdelmageed Elmustafa at Old Dominion University has run parametric studies of indentation creep to help us examine how to interpret hardness-indentation strain rate data. Students Joseph Jakes and Jon Puthoff have invented broadband nanoindentation creep as a method for generating hardness data across 4-6 orders of magnitude in strain rate. Studies on the effects of edges and heterophase interfaces on nanoindentation have helped us learn how to isolate local properties in heterogeneous specimens so that we can truly employ the BNC method as a local probe of deformation mechanisms.





Nanoscale Fracture

In collaboration with Walt Drugan in Engineering Physics, we investigated fracture at the nanoscale. Walt's student, Mike Starr, performed mechanics analyses of dislocation emmission from stressed, nanoscale cracks located near surfaces. Mike found that dislocation emmission is radically affected by the presence of a nearby interface or free surface.

Mike and undergrads Maria Lopez-Garcia and Ryan Webster studied nanoscale fracture using bubble rafts, analog models for crystals. Mike invented a "rocksalt" crystal structure bubble raft which behaved in a more brittle fashion than the usual hexagonal bubble rafts.




Bubble raft movies of fracture

Mechanical properties and chemistry of wood cell walls


There has been a resurgence of interest in the mechanical properties of wood, motivated in large part by concerns having to do with the environment and environmental sustainability.  Structural applications of wood usually entail the use of adhesives and additives, many of which are prone to giving off volatile organic comounds or VOCs deemed bad for air quality.  Scientists are seeking ways of replacing these chemicals with others that give off lower levels of VOCs, but in their search for replacements they also try to understand how existing adhesives and additives work -- that is, how they interact with wood and change its properties.  Environmentally friendly replacements can then be designed to target similar mechanisms. As another example, economic incentives are being sought for harvesting "undesirable" trees including non-native, invasive trees and small diameter trees in managed forests.  Normally, the wood from these trees has little economic value, but if high-quality, engineered products could be manufactured from the wood of these trees, then those products would help to provide economic incentives for harvesting the "undesirable" wood. Such products might also help to take some of the pressure off old growth forests. Key to improving the properties of these woods is the ability to modify the chemistry and structure of wood at the cellular level. 

In collaboration with Chuck Frihart and Jim Beecher at USDA Forest Products Laboratory, student Joseph Jakes has been studying the effects of chemical additions on mechanical properties of the cell walls in wood.  To tackle this problem, Joseph has had to overcome a number of obstacles associated with the mechanics of nanoindentation in cell walls.  First, Joseph invented a method for making reliable specimens without having to embed the wood in a supporting media such as Spurr's epoxy (the usual practice in these kinds of studies) and thereby avoid contamination of the wood with the components of the embedment.  He then found a protocol for generating and interpreting nanoindentation data to remove artifacts caused by the heterogeneity of the cellular structure inclidng the lumena, or holes, that  run along the length of the cells.  Lastly, because of the polymeric makeup of the cell wall, interpretation of nanoindentation data from wood has been open to question:  the relationship between hardness and viscoelastic and viscoplastic deformation propertes in polymers is poorly understood.  To solve this problem Joseph helped to invent perfected broadband nanoindentation creep (BNC) as a method for determining hardness across 4-6 orders of magnitude in indentation strain rate.  The rate effects allow us to decipher the connection between hardness and flow stress in polymers.  Joseph is using BNC to generate mechanical spectroscopy data for characterizing the kinetic signatures of deformation mechanisms in the wood cell wall. 




indents in wood


Wood Handbook





Thin films and surface layers


In this body of research we apply nanoindentation techniques to explore low-temperature deformation mechanisms in thin films, multilayers, and surface layers.

Graduate student Karl Yoder investigated hardening mechanisms in evaporated molybdenum films on silicon.  The films, with grain sizes down to 22 nm, extended the Hall–Petch relation out to hardness values between 6 and 12 GPa.  Analysis of nanoindentation creep, load relaxation, and rate change data show that thermally activated glide of dislocations is rate controlling, and that grain size has little effect on the underlying rate processes even down to 22 nm.  X-ray diffraction gave evidence of the e-beam evaporated films containing radiation damage in the form of high densities of interstitial loops or other lattice-expanding defect clusters resulting from argon ion bombardment during deposition.  Analysis of the creep data indicated that these defects affect the thermal activation of dislocations.  

Graduate student Francis Tambwe studied indentation creep signatures of deformation mechanisms in Nb-Cu nanolayer composites and found that the layer interfaces are weak barriers to dislocation motion. Down to 3 nm layer thickness the kinetic signatures of deformation appeared to be controlled by bulk-like dislocation glide mechanisms.  Similar experiments in Cu-Ni nanolayer composites revealed the layer interfaces in those materials to be strong barriers. 

Mageed Elmustafa studied the indentation size effect (ISE) in aluminum and alpha brass.  The indentation size effect (ISE) is where the hardness of a material (usually a soft metal) increases at shallow depths.  The study employed rate effects to examine the fundamental mechanisms responsible for the ISE. We noticed both hardness and derivative of hardness with respect to indentation (strain) rate exhibit an ISE. The trend of the derivative vs. hardness as a result of the ISE is consistent with the trend resulting of work hardening.  This result suggests that a dislocation mechanism is responsible for the ISE.  However, for shallow indents, the trend of the hardness data suggests that existing strain gradient plasticity models should be modified. 

A recurrent theme is the difficulty of making accurate measurements from layered specimens, where the substrates introduce artifacts into the measurements. We are aided by our ability to model the stiffness of contact between indenter and specimen for any arbitrary stack of layers. Simulations for indentation against films on silicon substrate can be found at another site





Power law dislocation creep


The concepts of scaling and similarity have always been fundamental to our description and understanding of dislocation mechanisms in deformation. Scaling laws relating stress to microstructural features, such as the dislocation density and average subgrain diameter, have been central to the development of models for plasticity and creep. Kuhlmann-Wilsdorf [1985] has championed the idea of ‘‘similitude’’ in work hardening, that is, the idea that dislocation structures arising from different levels of work hardening are similar to each other because the low-energy configuration at one level of stress is the same as at another level of stress differing only in length scale. Similitude justifies the use of the ‘‘average’’ dislocation to model the behavior of the entire dislocation network. On the other hand, the ability of Weertman [1957], in his archetype model of power law dislocation creep, to come up with a power law exponent of 4.5 instead of 3 was due to the implicit assumption that similarity is broken: as the stress is increased, the distance h between parallel glide planes of neighboring sources decreases in inverse proportion to stress, while the source density, M, remains constant.

Load relaxation data are often found to exhibit a scaling between stress level and characteristic strain rate.  Graduate student Thawatchai Plookphol studied the microstructural basis for this scaling along with the relationship between load relaxation and steady state creep. Thawatchai showed that as the dislocation structure evolves during transient creep, it passes through a series of states each of which can be represented by a load relaxation curve at constant strain.  Graduate student Jon Putoff modeled load relaxation using a cellular automaton to replicate a dislocation percolating through a crystal. 


Creep-fatigue interaction in solders


Graduate student Seong-Min Lee investigated the role of grain (colony) boundary sliding in the creep-fatigue interaction in Pb-Sn eutectic. He found that at room temperature grain boundary sliding takes place at strain rates below 10-3/s. Based on an analysis using a mechanical analog model to interpret the width of the sliding-no sliding transition, Seong-Min estimated the power law exponent which governs sliding. Seong-Min also measured the growth and coalescence of cracks along colony boundaries during strain cycling. 


Nanoindentation near edges, heterophase interfaces, and in compliant specimens




Nanoindentation measurements placed near free edges and heterophase interfaces are affected by the elastic singularities represented by these discontinuities.  It turns out that the effects can be lumped into a structural compliance that is independent of the size of the indent.  Grad student Joseph Jakes has shown how to account for the artifacts in nanoindentation measurements placed near edges.   



Negative stiffness composites


See Rod Lakes' homepage   



Extreme materials

Composites stiffer than diamond

Hooke's Law