I work on several topics, the common thread being
geometry. Most of my problems involve
nontrivial patterns in space; frequently the problem
is to determine the qualitative nature of the ground state
(or equilibrium phase).
Assisting staff in Physics 214
(access course website via Blackboard)
In past years, I've assisted in introductory courses
Physics 101-102, 207-208, 214, and (once) P213.
Also, I used to teach undergraduate Solid State (P454).
Undergrad projects:
I have ideas for projects in theoretical solid-state physics,
not using quantum mechanics and
related to my quasicrystals, biological physics, or
frusrated antiferromagnets research (see "Research supported
by the DOE", below). For more details,
click here
Grad research opportunities:
see here . Postdocs: Sorry, I don't usually have funding
for a fulltime, two-year postdoc.
Typically on the computational-analytic borderline, with a
focus on: what are unbiased ways to get information out of
the (experimental or numerical) data?
There are two threads currently:
(1) How to characterize the (maybe exotic)
ground state of a quantum model from the exact diagonalization
of a finite-sized system. (Recent theses: Nai-gong Zhang '02,
Siew-Ann Cheong '06.)
(2) Phenomenological modeling to describe
STM (scanning tunneling microscope) experimental data from
high-Tc cuprates, taken in the Seamus Davis group at Cornell.
(Current grad : Sumiran Pujari)
Past projects under (1):
Spinless fermion model --
Naigong Zhang (PhD '02) and I studied the "t-V" model in two dimensions,
the poor man's Hubbard model: fermions of
only one spin flavor (as might be realized in an
extreme magnetic field) hop on a square lattice,
with a large (for simplicity, infinite) repulsive
interaction between those on nearest-neighbor sites.
This is comparatively tractable (numerically and
analytically) and had interesting charged domain
walls called stripes.
Paul Fendley has devised a similar model which
-- for a special value of an
added pair interaction -- is supersymmetric;
it would be interesting to understand the phase
diagram next to that special point (which Fendley
conjectures to be multicritical).
It's worth noting that perhaps the likeliest way
to realize spinless fermions experimentally is with
cold, dilute atoms in optical lattices, which
is part of the research of Prof. Erich Mueller
(LASSP theory, Cornell).
Quantum renormalization group --
Siew-Ann Cheong (Phd '06) and I studied the density matrix for a
block of sites in a ground state of (e.g.) a fermion model.
The DM may be useful in guiding the truncation in
a two-dimensional real-space renormalization
(such as "contractor renormalization").
A separate idea, "correlation density matrix", is a way
to discover what kind of correlations is dominant in a particular
system, or to prove it has no hidden correlations.
This category includes
(1) magnetic ordering in frustrated vector antiferromagnets,
and (2) the statistical physics of discrete
spin models that map to rough interface models
(related to conformal field theories).
(1) Statistical physics (classical and quantum)
of frustrated antiferromagnets
on the Kagome and related lattices.
These complex antiferromagnets have nearly degenerate ground states,
and it is challenging to figure out how small perturbations
single out one of them as being the true ground state, or
produce an exotic disordered state which is a superposition of
many configurations.
(Recent theses, Uzi Hizi '06;
current undergrad, Sophia Sklan '10).
In the systems I study, the spins have length > 1/2. Thus the naive approach
(as works in most magnetic systems) is to expand around the classical ground state
-- but which ground state? since they are highly degenerate.
Hence it's necessary to set up a perturbation theory to expand around an
unspecified state. A common phenomenon in the business is "order by disorder":
that means the degeneracy gets resolved (and long-range order develops)
precisely due to the strong fluctuations associated with the degeneracy, or perhaps
due to quenched randomness (e.g. substitution of the moments by nonmagnetic ions).
Uzi Hizi (PhD '06) worked out how quantum fluctuations
resolve the degeneracy
of the spin-ordering pattern of pyrochlore antiferromagnets.
(2) Height models --
I've also studied 2D discrete models with "height" (interface)
representations, which connects to the theory of exact solutions and to
conformal field theory (past collaborators Dr. Jane Kondev and Dr. Chen Zeng.)
Such models are currently used in toy models of
highly frustrated and/or exotic kinds of order, in
(a) quantum systems [See
"From classical to quantum dynamics at Rokhsar-Kivelson points"
,
or (b) three-dimensional systems
Quasicrysstals are complex metal alloys with highly ordered, yet non-periodic
structures.
We want to determine their structure and understand why
they form.
Thus our work breaks down into
(1) atomic structure fitting and structural energies;
(2) random tiling ensembles;
(Senior collaborator: Dr. Marek Mihalkovic, Slovak Acad. Sciences.
Prof. Richard Hennig (Cornell Materials Science Dept.) did a
project with me as part of his Ph.D. thesis (2000) from Washington U.
Past undergrads: Nan Gu '05, Sejoon Lim '08; current undergrad,
Amulya Bhagat '10)
My colleague, Prof. Veit Elser, was formerly active in quasicrystals
about quasicrystals.
(1). Atomic modeling of quasicrystals
On the atomic scale, we try to use microscopcially-derived
effective potentials to resolve details of the atomic arrangements
which would be unclear from diffraction fitting.
With Dr. Marek Mihalkovic (from Slovakia and Chemnitz, Germany)
and Prof. Mike Widom (Carnegie-Mellon University),
we have built a successful computer model of decagonal
AlNiCo in this fashion: see the preprint
"Total-energy-based prediction of a quasicrystal structure"
, by Mihalkovic et al., or a
related preprint
by C.L. Henley, M. Mihalkovi\v{c}, and M. Widom,
In 2004, this work was extended to the Co-rich
phase of d(AlCoNi) by Nan Gu (undergrad '04) (see an
example image of the idealized atomic pattern
found by this simulation.)
(2) Statistical physics of quasicrystals,
We are pursuing the "random tiling models".
Dr. Marek Mihalkovic and I are have simulated
random-tiling like models with Hamiltonians
that, in the low-temperature limit,
approach a particular random tiling of the tiles called "canonical cells".
I work on two major topics in biological physics
They involve statistical mechanics plus spatial patterning.
(1) Assembly of virus shells ("capsids")
Every virus encodes a protein, many copies of which
form a shell, called the "capsid", that encloses
the viral DNA or RNA until it reaches its host cell.
The local pattern is well approximated by a triangular lattice,
except at certain "disclinations" the coordination is reduced
from 6 to 5. Many viral species (very high) icosahedral symmetry
and are relatively large.
How (or to what extent) does the capsid manage to reach such
a structure in its non-equilibrium growth process?
We've focused on retroviruses (of which the
best known is HIV, the AIDS virus) which form an
ensemble of irregular structures; we are in contact
with the retrovirus lab of Prof. Volker Vogt at Cornell.
This year, we're trying to extract effective spring constants
from molecular dynamics simulations which will let us
model the energetics at the coarse-grained scale of the
above-mentioned triangular lattice.
(Student Steve Hicks.)
(2)
Macroscopic Left/Right asymmetry (handedness) in animals and plants
By what physical
mechanism did life break chiral symmetry?
Obviously, an ancient symmetry-breaking led to the
handedness of biological molecules, but it's nontrivial
how this gets expressed at the level of a multicellular
organism.
Examples are (1) bilateral asymmetry in vertebrate animals
(why isn't your heart on the right side?), and (2)
spiral growth in plants.
Past projects
. Structural energy of elemental Boron
As an offshoot of our interest in (plausible) hypothetical
quasicrystalline Boron
-- real boron, in all its polymorphous forms,
is built from icosahedral clusters --
my Ph. D. student Wei-Jing Zhu
in his thesis, extracted an empirical potential (the first) for elemental
Boron, from a database of LDA total energies
of hypothetical structures.
Click
here for a version of this work in preparation for Phys. Rev. B.
Nonlinear dynamics
In the past I worked on statistical physics topics
of self-organized criticality and
on percolation (that includes a bit on cluster-update algorithms).
Semiclassical approaches to spin systems
We applied semiclassical techniques to relate the
eigenstates of quantum spin systems to their classical dynamics
(with Paul Houle PhD '98 and N.-G. Zhang PhD '02).
Click
here
for some reprints on semiclassical dynamics.
(Supported by NSF grant.)
Surface growth models
I've been interested in novel mathematical characterizations of rough surface
morphologies, and in the relationship of microscopic to
continuum growth models.
See
"Nonlinear Measures for Characterizing Rough Surface Morphologies" ,
by Jan\'e Kondev, C. L. Henley, and David G. Salinas,
Phys. Rev. E 61, 104-125 (2000).
(Supported by my NSF grant)
Materials Research
I belong to the Cornell Center for Materials Research.
I was once active in a research group within the CCMR
on "Energetic Surface Processing", which involved models
of the growth of crystals by atom deposition.
(see above)