Going with the Flow

by Eric Goldscheider and Aubin Tyler

Robert Thompson and graduate student Laila Abu-Lail prepare samples for the gas chromatograph. They are looking to see how well molecular sieves known as zeolites are able to remove harmful organic compounds from drinking water.

With sophisticated computer models and modern analytical equipment, WPI researchers study the dance of bubbles, the interplay of blood flow and arterial plaque formation, and the action of molecular sieves.

Flow has been the subject of scientific inquiry for centuries. More than two millennia ago, Archimedes invented a spiraling contraption that made water flow against gravity. In the 17th century, Galileo sought to understand the complex physics of swirls, whorls, and vortices. At the start of the 20th century, aviation pioneers explored the flow of air over wings, searching for the secrets of lift. In the last few decades, powerful computers running advanced algorithms have opened whole new realms in the study of how things flow from one place to another.

Seen broadly, flow research is not just about the motion of liquids and gases. At WPI, it encompasses the study of phenomena as diverse as the migration of pollutants through groundwater, the transport of ions across cell membranes, the streaming of data across open space in wireless networks, and the carefully choreographed dance of materials and products in an industrial supply chain.

The common denominator is motion in the context of the pathways and principles that determine how things get from here to there. Some of the work on flow under way at WPI is highly theoretical. Some is practical. Some has life-or-death implications. As these labors show, the products of a life at the bench come from going with the flow of the cooperative tango between pure and applied research.

When Flows Collide

If you were to say Gretar Tryggvason has devoted his career to bubble theory, you wouldn't be far off the mark. He calls his area of scientific inquiry multiphase flow, which is another way of describing what happens when two flowing substances, such as air and water, come into contact. When they do, they often produce bubbles, or a spray of drops.

Among the questions Tryggvason and his team seek to answer are: how big are the bubbles or drops, how do they move and interact, how quickly do they circulate, and are they more or less likely to combine with other bubbles or drops? The answers are of great interest to industry, as they can affect everything from the rate at which fuel combusts in a furnace to how readily sugars ferment to produce ethanol to how easily one can control any number of chemical processes.

Tryggvason, professor and head of WPI’s Mechanical Engineering Department, is responsible for some of the computational breakthroughs that have allowed scientists to glean information about multiphase flows using powerful computers that crunch vast amounts of numbers. He says it is not uncommon for him to book weeks or even months on a supercomputer to accomplish calculations aimed at modeling exactly what happens when bubbles are bubbling. "The days when we would build these models using paper-and-pencil computations in the privacy of our cubicles are gone," says Tryggvason.

"The challenge is to understand the range of scales. To do that, it is not sufficient to know what happens to one bubble; you need to know how several interact."

To get an idea of the magnitude of the mathematics involved in capturing the complexity of these inherently unstable systems, imagine the movement of a column of bubbles represented in 100 million grid points, each of which can have several values that are updated 1,000 times per second. "The challenge is to really understand the range of scales,” says Tryggvason. “To do that, it is not sufficient to know what happens to one bubble; you need to know how several interact. With each interaction, the complexity increases."

Watch a video about WPI's multifaceted flow research.

Stemming Heart Attacks

Dalin Tang uses sophisticated mathematical models and data from patient-specific MRI studies to understand how arterial plaques form and rupture, leading to heart attacks and strokes.

Like Tryggvason, Dalin Tang, professor of mathematical sciences, is interested in the complex interactions that take place at interfaces-in this case, the interplay of blood flow and mechanical forces inside tissues that can lead to the formation and growth of atherosclerotic plaques. On his computer monitor, Tang displays a computerized 3-D image of a diseased blood vessel. The outer portion of the vessel is lime-green mesh; inside, a segment of red mesh indicates plaque, its composition determined, perhaps, by splashes of yellow (lipid) or purple (calcium). By adding a cross-sectional view of what's happening inside the plaque to standard imaging, Tang hopes to help physicians noninvasively assess and predict a patient's vulnerability to heart attack or stroke-before it happens.

Lambert's team will try to direct the growth of neurons on an artificial surface, such as glass, gold, or silicone, so their axons (the long, thin fibers that carry electrical impulses) extend along channels etched in the materials. The team will try to achieve predictable neuron growth and axon myelination (fully developed axons are covered with a sheath of myelin, a substance that insulates them, much like the plastic coating that insulates electrical wires). "This is very basic research, looking particularly at what factors and substrates are important for the growth of neurons," Lambert says. "Connecting to the nervous system is a complicated problem; we are very much in the infancy of this effort."

The work calculates blood flow in an artery and stress and strain inside plaque based on patient-specific MRI and mathematical models. "Heart attack and stroke are often related to a rupture of plaque-under very critical flow or stress conditions," Tang says. "We are trying to develop advanced techniques and software so that we can predict potential ruptures."

An early-warning system for the heart wins entrepreneurship award

WPI's Kalenian Award is designed to encourage innovation and entrepreneurship among students, faculty, and alumni. The 2008 award was presented to mathematical sciences professor Dalin Tang for his work on a noninvasive diagnostic tool that could give physicians the ability to identify and diagnose cardiovascular disease earlier than is currently possible. Tang hopes to commercialize the software, which is based on his research.

Today, Tang notes, a 70 percent narrowing or blockage of an artery by plaque is considered an indication for surgery. Yet not all blockages-even of that size-require intervention. In fact, of 20 surgeries performed, perhaps only one plaque rupture may be prevented, he points out. "We are trying to develop better prediction methods to reduce unnecessary surgeries."

Tang and his collaborators are pushing to further stratify the morphological classifications of plaque vulnerability introduced by the American Heart Association in the 1990s. "Plaques that have certain features, such as a large lipid core with a very thin cap, are more vulnerable than those with stable structures. Plaque vulnerability assessment should be performed with morphology and mechanical forces combined."

The Tiniest Filters

The connection of flow to human health is also a key concern of Robert Thompson, professor of chemical engineering. This interest has led him to study flows at a very small scale that can have large implications for the quality of drinking water. Thompson and his colleagues work with molecular sieves called zeolites. These naturally occurring and synthesized minerals have been found to have special filtering properties (Thompson calls them "smart filters"). "The most useful ones are hydrophobic and organophilic," he says, explaining that they repel water but can attract, or scavenge, some of the hazardous organic compounds that can find their way into water supplies.

The effluents include organic compounds poured down the drain, flushed down toilets, and spread on the land, including pharmaceuticals and antibiotics, pesticides, solvents—even caffeine.

Flow comes into the picture when one examines the pores of various zeolites, Thompson says. The size and structure of these pores help determine how quickly the compounds can sequester organic molecules. Variations in the pore structure are measured in nanometers (a nanometer is a billionth of a meter). These differences cannot be detected except with the most powerful microscopes, so Thompson and his team use computer simulations to help decipher the results of laboratory studies and paint a more complete picture of how materials move through these "nanoporous environments."

He says he believes the WPI group, which includes John Bergendahl, associate professor of civil and environmental engineering, and Nikolaos Kazantzis, associate professor of chemical engineering, along with four off-campus colleagues, including Jennifer Wilcox at Stanford University and James Hauri Jr. at Assumption College, is the only research team using this combination of theory and experimental studies to investigate transport in nanopores.

Additonal Readings

A negative correlation between human carotid atherosclerotic plaque progression and plaque wall stress: In vivo MRI-based 2D/3D FSI models
Tang, D., C. Yang, S. Mondal, F. Liu, G. Canton, T. Hatsukami, and C. Yuan, Journal of Biomechanics, vol. 41, no. 4, pp. 727-736, 2008.

Adsorption of disinfection byproducts on hydrophobic zeolites with regeneration by advanced oxidation
Koryabkina, N., J. Bergendahl, R. W. Thompson, and A. Giaya, Microporous and Mesoporous Materials, vol. 104, pp. 77-82, 2007..

Much of the impetus for this work comes from a massive study by the U.S. Geological Survey of all 50 states, which showed that pollution in waterways is not limited to obvious culprits, such as factory waste and sewage. The effluents include organic compounds poured down the drain, flushed down toilets, and spread on the land, including pharmaceuticals and antibiotics, pesticides, solvents-even caffeine. "Many of these are toxic, carcinogenic, or obnoxious-some in very low concentrations," according to Thompson, who adds that most are difficult to remove with conventional filtering systems.

Thompson's research isn't limited to the flow of water through zeolites. He is also interested in finding out how to use these molecular sponges to build highly effective filtration systems, which involves binding them with materials, such as clay, that won't clog their pores. "Our fundamental interest is making drinking water safer by removing harmful organic compounds," he says.

Whether the ultimate goal is improving public health, heading off a deadly disease, or advancing all manner of industrial processes, flow research at WPI will continue to be marked by an unending stream of new ideas and innovation, as new discoveries in the lab cascade into real-world applications aimed at meeting the rising tide of critical problems.

A New Generation of Tools for Studying Flow

In nanofluidic and microfluidic devices, gases and liquids move through channels or over physical features as small as a few nanometers (billionths of a meter). Since only a small number of molecules can fit side by side in such minute passages, the interactions of those molecules with the surfaces over which they flow come to dominate the fluid’s behavior. The multiplicity of physical scales present in such devices breaks down the traditional computational tools that scientists use to study flow behavior at the atomic or macroscopic scale.

With grants from the Air Force Office of Scientific Research, Nikos Gatsonis, professor of mechanical engineering and director of WPI's Aerospace Engineering Program, is developing a new generation of computational methods that can accurately describe nanoscale fluid flow and predict fluid behavior in devices that involve nanoscale to microscale (millionths of a meter) characteristics. These advanced models are validated with a wealth of laboratory experimentation, including work conducted in Gatsonis's lab.

With growing interest in nanoscale fluidic devices and processes, there is a critical need for these new tools. Among the devices of interest to the Air Force are tiny thrusters that will propel micro- and nanospacecraft. Gatsonis's models can predict how flows change as they move from the interior of a nozzle (as narrow as 250 nanometers, see image at left ) to the plume (measured in centimeters). "Such multiscale simulations provide unparalleled insights on the operation of nanodevices and provide directions for their optimization," Gatsonis says.