apply to be a Lawrence fellow, knowing your chances were less than
1 in 100 of being accepted? For the applicants, the stakes are high.
But the payoff is great for both the fellows and the Laboratory.
postdoctoral program is formally known as the Lawrence Livermore
Fellowship Program. Informally, it is called the Lawrence Fellowship
in tribute to Ernest O. Lawrence, the cofounder of the Laboratory,
who cultivated creativity and intellectual vitality in the scientists
who worked with him. Lawrence Livermore National Laboratory strives
to do the same.
Laboratory has always been a place where postdoctoral fellows thrive.
They can work on state-of-the-art equipment with leaders in their
field, performing research in areas of high demand. While all postdoctoral
fellows pursue independent research, most are hired by a particular
program, usually to perform research for a specific project. Lawrence
fellows have no programmatic responsibilities and are given the
opportunity to select the group in which they want to work. The
allure of freedom and an atmosphere that cultivates creativity,
coupled with a competitive salary and Livermores extensive
resources, make the Lawrence Fellowship Program a prestigious opportunity.
In exchange, it brings to Livermore some of the most sought-after
Ph.D.s in the world.
fellows produce remarkably creative research during their tenure.
Many stay on as full-time career employees, continuing their work.
Some leave Livermore to take positions at other institutions. But,
as one fellow says, The ones who leave are ambassadors for
Livermore for the rest of their careers.
Solution to a Challenge
The Lawrence Fellowship Program was the brainchild
of Jeff Wadsworth, former deputy director for Science and Technology.
He initiated the program in 1997 in an effort to reverse the effects
of the dot-com boom, which was leading many young scientists
to choose the remuneration offered by private industry over employment
with Department of Energy laboratories.
To help persuade the best
and the brightest to come to Livermore, the Lawrence Fellowship
offers an attractive salary and considerable research freedom. It
was modeled after the J. Robert Oppenheimer Postdoctoral Fellowship
Program at Los Alamos National Laboratory. In both programs, non-U.S.
citizens may apply. Lawrence fellows are hired by the Directors
Office, in cooperation with Livermores University Relations
The new program was first
announced in the fall of 1997. Although some Lawrence fellows learn
the program through contacts with Laboratory employees, most applicants
find out about it through advertisements in journals such as Science
and Nature or on the Web at either fellowship.llnl.gov/
We are interested
in finding people who werent necessarily thinking about coming
to Livermore or who didnt know about Livermore initially,
says Harry Radousky, chair of the Lawrence Fellowship Program committee.
The fellows are chosen for
3-year appointments by a selection committee consisting of a representative
from each of the Laboratorys scientific directorates. The
criteria for acceptance are rigorous. Out of 1,849 applicants in
the first 4 years of the program, only 15 have been accepted. More
recently, 282 applications were received for the programs
fifth year, and 2 applicants have been invited to participate.
Each application is read
by the selection committee, which looks primarily for leadership
of stellar research projects. Applicants must have received their
Ph.D. within the last 5 years. The applicant pool is eventually
reduced to 6 individuals who undergo a 2-day interview. On the first
day, the fellowship finalist gives a seminar on his or her area
of interest; has lunch with the committee, which serves as a question-and-answer
session; and then meets with current fellows in the afternoon. On
the second day, applicants have the opportunity to talk to Laboratory
scientists with whom they might be interested in working.
The goal of this process
is to find people who will succeed at the Laboratory. The likelihood
of success is measured in several ways: by matching an applicants
field of interest with those of the Laboratory, examining the applicants
academic record and publications, and analyzing the research projects
the applicant has initiated and the level of innovation those projects
Were not looking
for management skills but at scientific leadership, says Radousky.
The object of the fellowships is to encourage intellectual
vitality at the Lab and to recruit the best people in the world,
he continues. What weve discovered is that the application
process is an excellent way to attract people to all kinds of positions.
Many applicants who dont get into the Lawrence Fellowship
Program are awarded postdoctoral fellowships to work in Laboratory
programs or are hired as full-time employees.
Of the 15 individuals who
have received Lawrence Fellowships thus far, 3 are now career employees,
2 left to become professors at the Massachusetts Institute of Technology
(MIT), 1 went to the National Institute for Standards and Technology,
another returned to his native Belgium, and the remaining 8 are
still Lawrence fellows.
Results of Freedom
Freedom to work on projects
and with mentors of their choice is what most current Lawrence fellows
say attracted them to the program. This freedom, coupled with the
Laboratorys interdisciplinary atmosphere, also permits many
fellows to move outside their initial area of specialization and
investigate other scientific fields.
Wei Cai, for instance, a
current Lawrence fellow from China, earned his Ph.D. from MIT. Midway
through his graduate work, mentor Vasily Bulatov left MIT for the
Laboratory. Bulatov encouraged Cai to apply for the program. Cai
was a successful fellowship applicant and has worked not only with
Bulatov but also with Malvin Kalos, the father of quantum Monte
Carlo simulations. With Kalos, Cai has been investigating how to
use Monte Carlo simulation codes more efficiently for modeling the
microstructures of materials. Cai has amended some of Kaloss
techniques and applied them to small-scale problems with great success.
Now, together with Kalos, Bulatov, and other Livermore researchers,
Cai is working on a project funded by the Laboratory Directed Research
and Development (LDRD) program to apply these techniques to larger,
more complex systems. Cai has also been working on a new massively
parallel computer code for modeling dislocation dynamics. What
happened here has a lot to do with the academic freedom the fellowship
provides, Cai attests.
This freedom also allowed
Cai to work on a particularly exciting project far removed from
his usual line of research. At the suggestion of Giulia Galli, leader
of the Quantum Molecular Dynamics Simulations Group, Cai tried to
solve a problem that Gallis group was facing: adding a means
of modeling a magnetic field to the electronic structure simulation
codes regularly used to model condensed matter systems. Cai devised
a code that successfully modeled in two dimensions the behavior
of small systems, such as isolated hydrogen atoms and molecules,
under an arbitrary magnetic field. The next step will be to apply
this method with the more powerful electronic structure codes used
for large-scale calculations, such as the modeling of magnetic field
effects on the dynamics of fluid hydrogen.
Cai notes that the freedom
allowed in the Lawrence Fellowship Program can be almost disconcerting
at times. You need discipline and must be able to make decisions
at critical times about what you want to study.
|Lawrence fellows Jeffrey Grossman
and Andrew Williamson are using quantum Monte Carlo simulations
to research the characteristics of nanostructures such as these
silicon quantum dots. (a) A 71-atom silicon quantum dot. Hydrogen
atoms (white) bonded to the surface make the material less reactive.
(b) When a more reactive oxygen atom replaces two hydrogen atoms,
the electron charge cloud (purple) is drawn toward the oxygen
atom, dramatically changing the optical properties (wavelength)
of the silicon quantum dot.
at the Nanoscale
computational physicists became a team as Lawrence fellows. Jeffrey
Grossman, a Ph.D. from the University of Illinois at Champaign-Urbana,
and Andrew Williamson, a Ph.D. from the University of Cambridge
in England, had known each other for years and both were interested
in working with Giulia Galli. Almost immediately after arriving
at Livermore as fellows, they applied for LDRD funding to use quantum
Monte Carlo simulations to learn more about the characteristics
of nanostructures, atomic-scale dots 1,000 times smaller than the
width of a human hair. (See S&TR, April
Simulations Tell the Atomic-Level Story.)
in nanotechnology centers around one very simple concept,
says Grossman. When you make something really small, its characteristics
change. At the nanoscalejust a few hundred atomsa materials
properties start changing and become really interesting. Those differences
and the ability to control the size of the structures mean that
all kinds of new devices could be madenew ways to deliver
drugs, storage systems for hydrogen fuel, detectors that can recognize
microscopic amounts of anthrax in the air.
were a major draw for this duo because quantum Monte Carlo simulations
are computationally intensive. With Livermores computers,
they can do work that they couldnt do at most places.
Another selling point was
that Gallis group was beginning a new project on nanoscience
when Grossman and Williamson joined the Laboratory. Part of
what makes the Lawrence Fellowship Program so attractive,
says Williamson, is the opportunity to create something new
and shape the direction that research takes, rather than trying
to come in and fit into a slot that was shaped by someone else.
Experimental biologist Julio
Camarero, who is also working at the nanoscale, saw the Lawrence
program advertised in Science and Nature while a postdoctoral fellow
at Rockefeller University in New York City. Camarero received his
Ph.D. from the University of Barcelona.
At Livermore, he started
out in the Biology and Biotechnology Research Program (BBRP) but
moved to the Chemistry and Materials Science Directorate, where
he continues to perform biological experiments. He is a member of
a team that aims to use dip-pen nanolithography to create and probe
ordered arrays of proteins and colloids. One of the many uses for
dip-pen nanolithography is to create tiny sensors that will detect
biological warfare agents.
The Lab is interested
in applying science and technology to create tools for national
security, notes Camarero. I think that the technology
we have developed is very powerful and has many applications, not
the least of which is protecting us from biological terrorism.
In dip-pen nanolithography,
the tip of an atomic force microscope is dipped into either an organic
or inorganic substance (the ink) and then is used to
write on the surface of an inorganic substrate. (See
Science Gets to the Heart of Matter.) As the tip moves across
the surface, it creates a precise, orderly pattern, or template,
of material that is in chemical contrast to the substrate surface.
The goal of Camareros
research is to form specific chemical patterns less than 10 nanometers
wide on silicon dioxide and gold surfaces. The chemicals in this
template will react with proteins, thus making the template a sort
of molecular Velcro to which the proteins bind in ordered
arrays. Use of these templates allows for total control over the
orientation of the proteins.
|Lawrence fellows Julio Camarero
and Aleksandr Noynow a full-time Laboratory employeeare
pursuing research using dip-pen nanolithography. This technology
uses the tip of an atomic force microscope (AFM) dipped in molecules
to write on an inorganic substrate. The molecules
react with the substrate to create a pattern of nanostructures
attached to the substrate. These nanostructures have a variety
of scientific uses.
Kim was at the University of Cambridge as a Wellcome Trust fellow
in the Applied Mathematics and Theoretical Physics Department when
he learned about the Lawrence Fellowship Program from colleagues
at the University of California at Berkeley and from Livermores
Web site. Kim works in BBRPs Computational and Systems Biology
Division, led by Michael Colvin. Traditionally, biology has
been a qualitative discipline, Kim says. But mathematics
can play an important role in the biological sciences by providing
a precise and powerful language to clarify underlying mechanisms
and reveal hidden connections between seemingly disparate systems.
Mathematical modeling may allow biology to become a predictive science
alongside physics and chemistry.
is applying the mathematical methods of statistical mechanics to
the study of the astonishingly complex interactions and collective
behavior of biological systems. He has studied the collective behavior
of interacting bodies (inclusions) in an elastic medium (a cell
membrane). The mathematical model that describes this behavior can
be used to investigate the mechanism that causes protein inclusions
in cellular membranes to distribute themselves into large, stable
aggregates as a function of their global shape. This research illustrates
the rich interplay between geometry and statistical mechanics that
underlies biological and other complex systems.
Kim is also developing a
mathematical model for gene regulatory networks. In a gene network,
the protein encoded by a gene can regulate the expression of other
genes, which in turn control other genes. A protein can also regulate
its own level of production through feedback processes.
This network of interacting
genes is another concrete example of collective behavior exhibiting
an amazing degree of complexity at many spatial and temporal scales,
Olgica Bakajin of Yugoslavia
is yet another fellow working at the nanoscale. Bakajin had completed
her Ph.D. at Princeton University and was on her way to the National
Institutes of Health (NIH) when Livermore called to inform her that
she was a successful Lawrence fellow applicant. Since arriving at
Livermore, she has worked on several projects related to the development
of novel microstructures and nanostructures. She is designing and
fabricating a fast microfluidic mixer for the study of proteins.
Just 10 micrometers widea human hair is 80 micrometers widethe
mixer can cause proteins to fold and unfold when solution conditions
in the mixer are changed quickly and precisely. Bakajin will be
using the mixer to examine the kinetics of fast protein folding
reactions (an LDRD-funded project) and to investigate the kinetics
of the folding of single-protein molecules (a collaboration with
Working with former Lawrence
fellow Aleksandr Noy, Bakajin is using carbon nanotubes in microfabricated
devices to separate biological molecules. In the future, these microdevices
could be used as detectors of chemical and biological warfare agents.
The interdisciplinary atmosphere at the Lab has provided me
with lots of research opportunities, says Bakajin. Right
now, I have more ideas for interesting projects than I have time
to pursue them.
|Olgica Bakajin is designing
and fabricating this fast microfluidic mixer used for researching
the kinetics of protein folding.
Three former fellows are
now full-time Laboratory employees, having exchanged some of the
freedom of the Lawrence Fellowship for a staff position.
Theoretical biologist Shea
Gardner, who studied population biology at the University of California
at Davis, worked initially on several computational biology projects,
one of which was a mathematical model to tailor chemotherapy treatments
for individual cancer patients. Treatment strategies are based on
the kinetics of the patients particular tumor cells. Gardner
has filed a provisional patent for this modeling approach and has
been contacted about commercially developing the software.
Gardner also worked on biostatistics
for the analysis of gene microarrays. A microarray is a glass microscope
slide covered with spots, each occupied by a different
gene. (See S&TR, March
Down Virulence in Plague.) The entire slide is exposed to a
stimulus such as a chemical or a change of temperature, and scientists
note how each gene responds to the stimulus. With microarrays,
you can see the expression of over 12,000 genes at once, in a single
run, Gardner notes. Previously, you could look at just
one gene at a time.
is now participating in bioinformatics work for the National Nuclear
Security Administrations Chemical and Biological National
Security Program, computationally identifying DNA signatures that
could be used to detect biological pathogens. She hopes to continue
with this research. Mathematical modeling, biostatistics,
and bioinformatics are really different, she says. Where
else would I have had the opportunity to work on all three?
Noy, a physical chemist from Harvard University, came to Livermore
in 1998 to work on high-resolution microscopy. To that end, he developed
a new microscope system that combines the topographic capabilities
of the atomic force microscope with the spectroscopy capabilities
of a confocal microscope. (See S&TR, December
Science Gets to the Heart of Matter.)
My interests morphed
from just looking at tiny things to fabricating them and using them
for nanoscience applications, he says. Shifting focus
like that would not have been possible if I had not been a Lawrence
fellow. Noy has worked on several nanoscience projects, including
some that use carbon nanotubes in unique ways. Much of his research
requires his new microscope to make the results visible.
He now leads a group that
is fabricating electroluminescent nanostructures by dip-pen nanolithography.
The researchers write with a conjugated polymer that
emits light when a voltage is applied. Nanowires made of conjugated
polymer poly [2-methoxy, 5-ethyl [2´ hexy(oxy)] para-prenylene
vinylene], or MEH-PPV, may some day serve as light-emitting nanodiodes.
MEH-PPV nanowires are also highly sensitive to light and can serve
as tiny optoelectric switches, which today are typically 1,000 times
larger than tomorrows MEH-PPV nanowires will be.
Plasma physicist Robert Heeter
heard about the Lawrence Fellowship Program from Paul Springer,
a group leader in Livermores Physics and Advanced Technologies
Directorate, who performs laboratory astrophysics experiments. Heeter
has been working with Springer since coming to Livermore in 1999.
While at Princeton University
earning his Ph.D., Heeter worked in England at the Joint European
Torus, a magnetic fusion energy facility. But because of funding
cuts, magnetic fusion research had fewer opportunities when Heeter
was about to graduate. He was also interested in astrophysics, so
he decided to apply for a Lawrence Fellowship at Livermore, which
had active programs in both astrophysics and fusion energy.
Heeter became a Lawrence
fellow and almost immediately got involved in photoionization experiments
on Sandia National Laboratories Z Accelerator in Albuquerque,
New Mexico. Today, he continues his photoionization research. Ive
also been doing other experiments in high-energy-density plasma
physics, he adds. Ive stayed in the same group
and in the same field that I was in as a fellow. High-energy-density
physics experiments have numerous applications: in stockpile stewardship,
in inertial fusion, and in astrophysics. And theres a lot
of fundamental science to explore that hasnt been done before.
||One of the research interests
Shea Gardner pursued as a Lawrence fellow, which she continues
today as a Laboratory employee, is modeling the DNA signatures
of viral pathogens. These simulations contribute to technologies
for detecting agents of biowarfare.
Not all Lawrence fellows
stay on as full-time Laboratory employees. The most recent one to
depart was metallurgist Christopher Schuh, who left in the summer
of 2002 to become a professor at MIT. After completing his Ph.D.
at Northwestern University, he came to Livermore to work on grain
boundary engineering, in which conventional metallurgical processing
is tailored to produce better metals. Grain boundarieswhere
crystals with different orientations come togetherare the
weak link in any material. Schuh examined ways to manipulate the
orientation of crystals at grain boundaries to create metals with
desirable properties such as less cracking, corrosion, and cavitation.
Schuhs research also
took him beyond grain boundaries to the individual atoms in the
crystals. If you disturb the atoms in metals so much that
the crystal structure no longer looks anything like that of traditional
metals, the metals will have very different properties, says
Schuh. Were trying to understand how these changes affect
the physics of the metal.
Schuh notes that postdoctoral
fellows typically join a program with the understanding that they
have been hired to work with someone on a certain project. For
Lawrence fellows, he says, theres no such obligation.
That gives you complete freedom and a lot of latitude.
received his Ph.D. from MIT and immediately joined Livermore as
a Lawrence fellow, deferring a teaching appointment at MIT for a
year. While at Livermore, he helped to develop a code that extended
the use of direct Monte Carlo calculation from the simulation of
dilute gases to the simulation of dense fluids. With this code,
Livermore researchers can simulate for the first time the phase
change characteristics of a van der Waals fluid.
Joel Ullom, who completed
his Ph.D. at Harvard, focused on the development of cryogenic detectors,
which are small electrical circuits that produce a current or voltage
pulse when hit by a photon or particle. The detector must be cooled
to temperatures between 0.1 and 1 kelvin, so that the energy of
a single photon will produce measurable heating. Ullom used cryogenic
detectors to weigh the protein fragments dislodged from bacterial
spores by a pulse of laser light. He also developed refrigeration
technology to produce the ultralow temperatures needed for cryogenic
detectors. Ullom became a Laboratory career employee before leaving
for a position at the National Institute of Standards and Technology.
Luc Machiels, a native of
Belgium, received his Ph.D. from the Swiss Federal Institute of
Technology. After a postdoctoral position at MIT, he came to the
Center for Applied Scientific Computing, where he solved problems
in continuum mechanics. With colleagues at MIT, he developed a new
finite-element error control strategy for the version of the NavierStokes
equation that describes the motion of an incompressible fluid. The
technique, which is both accurate and efficient, calculates lower
and upper limits for the output of a system, such as the temperature
bounds at the surface of an electronic device. Before leaving Livermore,
he also developed new techniques for the solution and modeling of
partial differential equations.
Radousky has only good things
to say about the Lawrence Fellowship Program. Weve learned
that we can attract really top people to the Laboratory, he
says. This program has attracted the best young scientists
to the Lab and promoted university collaborations. It is also an
excellent way to do general recruiting.
When the program first started,
more fellows were engaged in traditional physics research, while
today more are studying biology and nanoscience. This shift is consistent
with changes throughout the scientific community. Biological research
leaped to the foreground with the success of the Human Genome Project.
Many experts predict that the 21st century will be remembered for
a revolution in biotechnology and medicine comparable to the advances
made during the last century in physics.
Nanoscience is a similarly
hot research topic. As all kinds of devices in our world
become smaller and smaller, nanostructures of all types will find
in all, the Lawrence Fellowship Program has been a resounding success
in bringing new talent to the Laboratory and encouraging creativity
and exciting science.
Laurie Powers and Katie Walter
Key Words: Lawrence
fellows, Lawrence Fellowship Program, postdoctoral positions.
For further information contact Harry Radousky (925) 422-4478 (firstname.lastname@example.org).
For information on the Lawrence Fellowship Program and other fellowship
opportunities at the Laboratory, see these Web sites: