LLNL Customizes Industrial Process
with Electrophoretic Deposition
February 21, 2017
Livermore National Laboratory (LLNL) researchers have published a
precise computer model of a deposition process using electrically
charged particles that will provide scientists and engineers with
unprecedented insights into the method.
The process of electrophoretic deposition (EPD) uses an electric field
to drive colloidal particles suspended in a liquid from a solution onto
a conductive substrate. Commonly used to apply paint to cars, EPD also
is utilized to coat ceramics, metals and polymers with a range of
materials and for 3D printing objects.
Developed using a particle dynamics framework and run on the Vulcan
supercomputing system at LLNL, the newly published model tracks every
single particle during the entire EPD process -- each particle is about
200 nanometers wide, roughly the diameter of the smallest bacteria. The
research is published in the Dec. 20 issue of the journal Langmuir.
The new electrophoretic deposition
model developed at Lawrence Livermore tracks every single particle
during the entire EPD process, each particle being roughly the diameter
of the smallest bacteria.
"This gives us more information than
any model before and fresh insights that were previously inaccessible,"
said the study's LLNL postdoctoral researcher Brian Giera. "Within this
particle dynamics framework we were able to get really detailed
information. In terms of understanding the EPD process in detail, this
is a first-of-its kind."
Over a period of two years, the team, led by principal investigator Todd Weisgraber, a researcher from LLNL's Materials Engineering Division,
developed the model and ran several dozens of different simulations,
changing the strength of the electrical field and the concentration of
salt in the system. Not only does the strength of the electrical field
affect the development of crystals, Giera said, but salt concentration,
surprisingly, also plays a key role. Giera said the model could be used
to better understand deposition kinetics, determine how fast to build
and anticipate resulting crystallinity, which could impact how armor is
produced, and how coatings are applied using the EPD process.
LLNL researchers (from left) Todd
Weisgraber, Luis Zepeda-Ruiz and Brian Giera developed the EPD model
over a period of two years.
"The model is poised to take on a lot
of questions," Giera said. "It gives us more predictive information to
optimize the system."
Luis Zepeda-Ruiz, a scientist in the Lab's Materials Science Division,
built the initial model containing all the essential mechanisms before
Giera took over the work. He said the model can be augmented to allow
for virtually any type of material, extending the science to a broad
range of applications.
"Our computational model can access details that are extremely difficult
to observe in real experiments," Zepeda-Ruiz said. "It also can be used
when experiments fail to reproduce results, when the solution ages and
changes its chemistry. Now we have a pure, reproducible means for doing
EPD, and that's a benefit."
The model has been so well received by the scientific community that it
was selected to be presented in a keynote speech by Giera at the
international Electrophoretic Deposition Conferences Series held in
South Korea in October.
researcher Andy Pascall, an expert in EPD, helped define the model's
initial parameter choices and is working on validating it for future
implementation. Pascall said the model will be particularly useful to
the field of photonics science, which requires precise control over
"Photonic crystallization is interesting to the scientific community in
general. The way this has been done before in the lab has been through
trial and error," Pascall said. "It's fair to say this is the only
particle-based EPD model out there. Having a model that can be
predictive allows you to run hundreds of virtual experiments that would
take us months to do in the lab."
Next, Giera will study how the colloidal particles re-suspend and, more
importantly, tailor the model to account for particles of different
The study was part of a process optimization project conducted with
funding from the Laboratory Directed Research and Development (LDRD)