Our research group specializes in the development of x-ray techniques to solve challenging problems like this, primarily using synchrotron radiation.
Reactive materials consist of two solid materials that are capable of producing a strong exothermic (heat-producing) chemical reaction. A well-known example is the thermite reaction in which aluminum reactive with iron oxide producing large amounts of heat. Thermite has a long history of technological uses, especially in welding of metal components in place, such as railroad tracks.
In the 21st century we can control the structure of materials on the nanoscale, creating new opportunities to create materials with useful properties. For reactive materials, for instance, we can produce foils consisting of alternating layers of different elements, each just a few nanomaters thick:
This level of control permits new applications of reactive materials. For example, reactive multilayers are used commercially for bonding materials that are difficult to join in other ways, such as bonding ceramics to metals where the difference in thermal expansion can lead to cracking.
With such thin layers the reaction can happen extremely quickly. Here’s movie showing a reaction in a multilayer consisting of aluminum and nickel; the reaction is initiated by a small spark, with a dramatic result:
Looking at one frame from the movie shows that the reaction proceeds as a self-propagating front that moves across the foil:
A variety of investigations have revealed that the reaction front has a width about equal to the thickness of a human hair (less than 50 microns), moves across the foil at up to ten meters per second, and experiences a heating rate of up to ten million Kelvin (degrees Celsius) per second.
The characteristics of the phase transformations in these reaction fronts are not the same as those that occur under less extreme conditions, but how can we study them? What is needed are characterization techniques with both spatial resolution on the order of micrometers (the width of the reaction front) and temporal (time) resolution on the order of microseconds (the time it takes the reaction front to pass a fixed point).
Our research group specializes in the development of x-ray techniques to solve challenging problems like this, primarily using synchrotron radiation. For studying reactive materials we focus the x-rays, using either x-ray capillaries or x-ray mirrors, to produce an intense x-ray beam as small as ten microns across:
Then, either with a very fast x-ray detector or by shuttering the x-ray beam to produce a short pulse of x-rays, we can record diffraction patterns in as little as one microsecond as the reaction front passes by. For example, here are four diffraction patterns recorded during the passage of reaction fronts in aluminum-nickel reactive multilayers similar to that in the movie above:
The first pattern shows three diffraction peaks from the unreacted aluminum and nickel. As the reaction front passes a new peak appears as a shoulder on the middle diffraction peak; this new peak is from the product of the reaction between aluminum and nickel, which is an intermetallic phase with the composition AlNi. At the same time there is a broad increase in the background, which is due to melting of the aluminum. By the time of the third diffraction pattern (55 microseconds later) the original Al and Ni are gone, leaving only the liquid and the AlNi intermetallic. Much later (128 ms) these two react to produce the final phase, Al3Ni plus a small amount of a vanadium-containing phase.
These in situ experiments reveal that the reaction proceeds by melting of aluminum, with some diffusion of nickel atoms into the liquid. The new AlNi intermetallic phase nucleates from the liquid, presumably at the interfaces with the (unmelted) Ni layers. It is worth pointing out that this sequence of events is completely different from what happens under at lower heating rates, showing the unique power of these in situ experiments to reveal details of processes that happen under extreme conditions.
A somewhat different technique shows the reaction front itself. Instead of doing diffraction, we use synchrotron radiation to produce images of the reaction front as it propagates. Here’s a movie of a reaction front in aluminum-nickel multilayers, similar to that above:
(Each pixel is about 2 microns and the frame rate is 7.2 microseconds per frame.) Notice that the reaction front moves uniformly and is apparently very narrow, just a few pixels in extent. Compare this with a similar reaction in an aluminum-zirconium multilayer:
Now the reaction front is not uniform at all! Instead, a series of “steps” move across the reaction front, each about 25 microns wide. The overall forward motion of the front is due to the successive passage of many such steps. This kind of insight into the nature of these reactions is only possible with advanced, in situ characterization techniques such as those developed by our groups.
Research on reactive materials in our group has been funded by the Office of Naval Research (ONR), the Department of Energy (DOE), and the National Science Foundation (NSF). Ongoing research in this area is conducted in collaboration with the group of Prof. Tim Weihs, who receives funding from the Defense Threat Reduction Agency.