... update pending.....


Leukocyte Transport

Blood cells

A marginated white cell (blue) in a small blood vessel

We are investigating the transport of white cells (leukocytes) toward the wall of microvessels, which is an important component of the inflammation response. Interactions with red cells is a key aspect of this process. In this work we employ a numerical model we developed specifically for simulating large numbers of deforming blood cells flowing and interacting in the microcirculation. It is a boundary integral solver for the Stokes flow inside and outside the arbitrarily deforming shell membranes that make up the cells. This work is funded by NSF.

animations: 1 2 3 4

Blood Rheology

Blood cells

The non-monotonic effective viscosity of blood flowing in small tubes

This cell simulation tool is also being used to study the red-cell dynamics leading to its peculiar non-monotonic effective viscosity, as seen in the plot to the left. In the smallest diameter vessels, cells are well known to flow in a regular single-file line; in tubes slightly large than the cell dimension, this breaks down and the cells interact in an apparently random fashion. The minimum effective viscosity occurs near this transition.

animations: 1 2


Jet Crackle

Direct numerical simulation of a Mach 1.92 turbulent jet and its near acoustic field.

High specific thrust jet engines, such as those on certain military aircraft, emit a distinctive sporadic raspy sound known a crackle. It is what gives such aircraft a distinctive acoustic signature --- they are not simply louder, but the sound has a different character than civilian transports. It is loud enough to damage hearing of nearby personnel, and its sporadic character makes it a particularly annoying community noise. It seems that supersonic advection of the turbulent eddies in the flow emit Mach-wave-like sound, but what leads to its specific character or how to reduce it remains unclear. We are using advanced simulation tools coupled with analytical methods for nonlinear acoustics to understand the mechanisms crackle and identify means of suppressing it. This work is funded by AFOSR.

animations: 1 2

Adjoint-based Optimization and Control

Adjoint schematic

Adjoint-based optimization for jet noise control

Recently, we reported a novel method that employs adjoint-based optimization to control free shear flow noise, jet noise being the principal application. The adjoint of the linearized and perturbed compressible flow equations is used to provide the sensitivity of the noise to changes in control near the nozzle lip. This was demonstrated for a model two-dimensional mixing layer flow by Wei & Freund (JFM, 2006). The current work, which is funded by NASA and is in collaboration with Prof. Bodony here at Illinois, is bringing this approach to control noise in a turbulent jet using plasma actuators. These actuators have been demonstrated effective for noise control by our collaborator Prof. Samimy at OSU.

animations: 1

Resonance in Engine Test Cells

WH Solution

A Weiner-Hopf solution of vortex-sheet waves scattering from the sharp-edged jet shroud exit.

Jet engines are tested at altitude conditions in low pressure test cells, such as those at the Arnold Engineering Development Center's. Under certain, difficult to anticipate conditions, a strong (170dB pressure) resonance occurs and can disrupt testing. We have (1) developed tools for predicting this resonance in complex geometries, (2) completed simulations in model geometries to identify the mechanisms leading to the resonance, and (3) mathematically analyzing interactions that might close the feedback loop leading to resonance using Weiner-Hopf methods. This work was supported by AFOSR via the Testing & Evaluation program.

animations/illustrations: 1 2 3 4 5

Anechoic Chamber for Nozzle Evaluation and Mechanistic Investigations

WH Solution

Anechoic jet-noise facility.

Collaborating closely with Prof. Joanna Austin and Prof. Greg Elliott, we have designed and built an anechoic chamber for the testing of complex geometry jet nozzles for their effects on noise and for investigating the root mechanisms of jet noise. It is capable of continuously running supersonic one-inch diameter nozzles and has reproduced accepted jet-noise spectra from other laboratories. We are currently investigating the connection between nozzle conditions and far-field sound. This work is funded by Gulfstream and Rolls Royce.


Papilla Model

Mechanical model of the renal medulla.

Shock-wave lithotripsy is a common treatment for kidney stones and other urinary tract calculi in which shocks, generated outside the body, are focused on the stone to break it up. It has been found that more injury accompanies treatment than initially thought, the mechanisms of which are unclear. We are investigating the mechanisms of this injury. We have proposed a novel cumulative shear mechanism. A model of the tissue of the inner medulla, which is where injury seems to begin, was constructed and simulated. It has been shown that shock shear can accumulate where the interstitial volume fraction is large. We have also studied the effect of tissue confinement on cavitation bubble dynamics (Freund, 2008). Cavitation certainly play an important role in cases of wide spread injury. We collaborate with Dr. Evan's and Dr. McAteer's groups at IUPUI, Dr. Bailey 's group in the Applied Physics Lab at UW, Dr. Cleveland's group at BU and Dr. Colonius's group at Caltech. Funding is from NIH.

illustrations: 1 2 3


Dynamics of atomically thin liquid films

Papilla Model

A moving solid-liquid-vapor tri-junction.

Many applications involves a liquid evaporating off a hot solid and many of these involve a solid-liquid-vapor tri-junction. The standard, continuum governing equations do not seem to apply in the atomic-scale neighborhood of this tri-junction, suggesting that some new physics comes into play. We are undertaking atomistic simulations of evaporating menisci and interpreting these in the context of asymptotic continuum models. Agreement is remarkably good in some circumstances, but the existing models considered also require refinement in other cases. A challenge in taking this approach is the hugely disparate time scales of the fluid and atomic velocities. This work was funded by NSF.

illustrations: 1 2

Ion Bombardment and Surface Nonastronomical of Silicon

Papilla Model

Nanometer-scale ripples formed by ion bombardment of silicon.

Bombardment of silicon with medium energy (around 1keV) argon ions leads to the formation of nanometer-scale surface patterns. The mechanism for this has long been thought to be due to ion induced sputtering. However, statistical results accumulated over thousands of ion impacts simulated with atomistic methods show that there is significantly more mass rearrangement than ion induced sputtering. For non-normal incident angles, the mean change in the surface is the same type of asymmetric catering observed in macroscopic impacts. Using angle dependent crater functions determined in this way in a stochastic model for surface evolution leads to the type of ripple formation observed in experiments, matching in both wavelength and long-time ripple amplitude. This work is collaborative with Prof. Harley Johnson and is funded by NSF.

illustrations: 1 2


A strong shock passing through a model propellant.

I currently serve as the research director of the Center for Simulation of Advanced Rockets (CSAR). Over the past 12 or so years, this DOE/NNSA-funded center has completed its mission of developing a large-scale multi-physics integration for the coupled mechanics (solid and fluid) and combustion processes of solid rocket motors. Current efforts are focused upon applying this tool to forward-looking rocket designs and well as expanding our propellant modeling capability to address safety issues. My own work in the center currently focuses on the shock sensitivity of propellant blends. This is collaboration with Dr. Jackson, Prof. Pantano, and Prof. Austin. Current efforts are also funded by AFOSR.

illustration/animation: 1 2


Healing and curing agents filling a model crack.

This project concerns materials that when cracked can heal themselves autonomously. In some fashion or other this involves a healing agent flowing into the crack and solidifying. This has been demonstrated in various settings by the AMS Team. Our effort focuses on the simulation of this processes, which is complex. It involves the drawing of the healing agent into the crack plane, usually by capillary forces, its mixing with agents that trigger its solidification, and its eventual solidification. The objective is to understand the detailed mechanisms at play in existing systems and optimize the process for new healing systems. This work is funded by AFOSR.

animations: 1 2


A model of a plume on Io showing density gradient and Mach number contours. Gravity couples direction with the gas dynamics at these conditions.

Volcanoes on Io, the moon of Jupiter, play a key role in its overall atmospheric dynamics, drive the rapid resurfacing, and also provide clues concerning its interior work sings. Collaborating with Prof. Susan Kieffer at UIUC and Prof. David Goldstein at UT Austin, we have developed a continuum simulation methodology for simulating these strange volcanoes. Because of their scale and because they vent into near vacuum, these volcanoes represent an interesting and unusual set of flow conditions: gravity interacts in an order-one fashion with the flow inertia leading to fountain-like gas-dynamic flows like that shown. Rarefied-gas effects, which are modeled by the UT Austin group, are also important. Funding is from NASA.