Bruce Finlayson Presented the Eighth Leland
Lecture
Bruce Finlayson presented the Eighth Leland lecture
at 4 PM on April 11, 2002 in McMurtry Auditorium. The title of
his talk was: "Continuum Mechanics on the Nanoscale: Modeling
Ferrofluid Movement under the Influence of a Time-Varying Magnetic
Field."
 Bruce
A. Finlayson is the Rehnberg Professor of Chemical Engineering at
the University of Washington. He is internationally known for his
papers and books, which have made major contributions in modeling
and numerical simulation of a variety of phenomena and processes.
His research has earned him numerous honors including the William
H. Walker Award of the American Institute of Chemical Engineers
in 1983 and election to the National Academy of Engineering in 1994.
He is also well known for his service to the profession. A Fellow
of the American Institute of Chemical Engineers, he served as its
President for the year 2000.
Professor Finlayson received B.A. and M.S. degrees
from Rice University and was a student in Professor Leland’s
thermodynamics course. He received his PhD degree from the University
of Minnesota. Following two years with the Office of Naval Research,
he joined the chemical engineering department of the University
of Washington, where he has spent his entire academic career. He
was chairman of the department from 1989-1998. He has held Visiting
Professorships in the U.S., Europe, and South America.
The endowed lectures honor the memory of Professor
Thomas Leland, a distinguished researcher and teacher who had been
a member of our department from the early 1950s until his death
in 1986.
THE T.W. LELAND, Jr. LECTURE
IN CHEMICAL ENGINEERING
Bruce A. Finlayson
Rehnberg Professor of Chemical Engineering
University of Washington, Seattle
"Continuum Mechanics on the Nanoscale:
Modeling Ferrofluid Movement under the Influence
of a Time-Varying Magnetic Field"
Thursday, April 11, 2002
4:00 P.M.
McMurtry Auditorium, Duncan Hall
Rice University
Visitors should use Entrance 12 from Rice Blvd. or Entrance 8
from University and park in the stadium lot. Shuttle service
from the parking lot to Duncan Hall runs every 15 minutes
For more information about this event,
please call Diana Thomas-Walker at (713) 348-4902.
ABSTRACT
Ferrofluids are suspensions of magnetic particles
(size 10 nanometers) in a carrier fluid, with a surfactant added
to keep the particles from agglomerating. The resulting particle
is about 25 nanometers in diameter and responds to a magnetic fluid.
Ferrofluids are used as sealants in pumps and computer hard drives,
because a magnet can seal the lubricating channel permanently; they
are also used in high-end stereos to remove heat and dampen oscillations.
The research our group is conducting attempts to model the flow
of ferrofluids using computational fluid dynamics techniques.
Early in my academic career I identified the conditions
under which convective instability of ferrofluids would take place.
In recent years, the study of ferrofluids has grown, and that paper
has supplied a starting point for many analyses by others. A few
years ago I began a research program looking at modeling ferrofluids
if they were used as coolants in electrical transformers; in that
application the transformer already had the time-varying magnetic
field, and the theory admitted the possibility of enhanced convection.
The same theory would make possible magnetic convection in space,
where natural convection would be absent or extremely small. In
the transformer problem, it soon became apparent that under standard
operating conditions the flow was either turbulent or in the regime
in which only a transient, laminar flow solution was possible. Since
little was known about turbulent flow of ferrofluids, that became
the new focus.
Working with Tahir Cader and Stuart Knoke at Energy
International, we devised experiments and calculations to model
the flow of fluids in pipes in the presence of an oscillating magnetic
field. In order to model the fluid, there are a variety of new physical
properties that are needed: vortex viscosity, spin viscosity, and
time constant of the fluid. These are all influenced to some degree
by the degree of agglomeration of the particles, which might lead
to groups of particles 50-100 nm in diameter. Because the fluids
are dark, it is difficult to use optical techniques for their study.
The talk will show simulations of laminar and turbulent
flow in an oscillating magnetic field and show that the theory is
correct. The theory involves a non-symmetric stress tensor and the
angular momentum equation. One problem is that the appropriate boundary
conditions are not known. The turbulent flow calculations are done
using a k-epsilon model under the hypothesis that the scale of the
particles is much smaller than the scale of the turbulence. In addition,
we have done simulations to compare with data for heat transfer
in stationary magnetic fields, where the magnetic field can be used
to offset gravity, or even completely reverse the direction of the
net body force. Finally, simulations of heat transfer with a rotating
magnetic field show how magnetic stirring can enhance heat transfer.
These calculations lead up to our current interest for the future:
modeling turbulence using direct simulation techniques to validate
our assumptions that the turbulence is not changed by the ferrofluid,
except in that it changes the viscosity.
|
| |
|
|
 |
|
