..
Greater Boston Area Theoretical Chemistry Lecture Series
2010-2011 Speaker Schedule
Dynamics of biomolecules at the single-molecule level: lessons from theory and computer simulations.
09/29/10 4:00pm
MIT Building 56, Room 154
Dmitrii Makarov
Observation and manipulation of complex dynamics of biological molecules through single-molecule techniques has
been a major breakthrough in molecular biophysics. The time resolution of such techniques however remains to be
their important limitation. In the last few years, several developments have pushed the time resolution of
single-molecule fluorescence measurements towards microseconds, sometimes even nanoseconds. Those advances pose
unique theoretical challenges. For example, whereas classical chemical kinetics is commonly concerned with the
rate of a chemical reaction, which is related to the frequency of transitions between “reactant” and “product”
states, recent single-molecule measurements of protein folding dynamics with sub-millisecond time resolution
brought to spotlight the time a protein spends in transit between its native and unfolded states. This transit
time, while being a fundamental property of any conformational rearrangement, has never been calculated by
traditional rate theories. Likewise, recent single-molecule studies of spatio-temporal correlations within
unfolded proteins promise to shed light both on the physical nature of the unfolded state and on the timescales
that govern protein folding. Do unfolded proteins behave like excluded-volume random coils and obey dynamic
scaling laws predicted by polymer theories? Is there any transient residual order and what would be its signature
on the observed protein dynamics? This lecture will describe some of our recent efforts to address these issues
and will focus on two topics: (1) Properties of transit times in complex molecular rearrangements and (2)
Universal scaling relationships observed in the intramolecular dynamics of unstructured biopolymers such as
unfolded proteins and single-stranded DNA.
Excitations and Dynamics in Time-Dependent Density Functional Theory: The Fun of Functional Development
10/20/10 4:00pm
MIT Building 56, Room 154
Neepa Maitra
TDDFT has become a method of choice for the calculation of
electronic excitations, response, and spectra in molecules,
clusters and solids. The theory applies also to electron dynamics in strong fields,
where it is particularly of interest, given that wavefunction methods to solve the dynamics
of more than two electrons is prohibitively expensive. However, state of the art
functional approximations have difficulties in capturing some of the
most interesting dynamical phenomena, including electronic quantum control, and non-sequential
double-ionization. Although in the response regime TDDFT has achieved an unprecedented balance
between accuracy and efficiency, here there are also certain
applications for which the commonly used functional approximations fail, for example charge-transfer
excitations, and multiple-excitations.
In this talk, we review the fundamentals behind the theory,
discuss some topical applications and striking successes, point to
the current challenges and failures for the usual functional
approximations in both the linear response and strong-field
regimes, and discuss the development of improved functional
approximations.
Quantum and Semiclassical Path Methods: From Dissipation and Decoherence to Superfluid Dynamics
11/17/10 4:00pm
MIT Building 56, Room 154
Nancy Makri
I will start this lecture with an introduction to the path integral formulation of time-dependent quantum mechanics,
its semiclassical limit, and the application of the path integral to quantum statistical mechanics. I will also describe
the numerical issues associated with the evaluation of the real-time path integral, known as the sign problem.
Next, I will discuss the application of the path integral to system-bath Hamiltonians, which leads to influence functionals and non-Markovian
dynamics, and present an iterative path integral methodology for propagating the density matrix in quantum dissipative systems that is stable
and accurate over very long times. I will describe representative applications to proton transfer rates, charge transport and multi-time correlation
functions using harmonic bath models to mimic the dissipative effects of condensed phase environments.
Shifting attention to the dynamics of liquids, I will show how decoherence can be used to eliminate the oscillatory phase in the semiclassical representation
of time correlation functions. The resulting forward-backward semiclassical approximation provides an efficient methodology for simulating the dynamics of
condensed phase processes. I will present applications to such low-temperature fluids as supercritical argon, neon, para-hydrogen, and helium across the
lambda transition. Calculated dynamic structure factors are in excellent agreement with experimental results and with fully quantum mechanical results obtained via
short-time propagator techniques. The forward-backward semiclassical calculations provide novel insights into the separate roles of quantum dispersion and Bose statistical effects on superfluid dynamics.
Last, I will describe a numerically exact and completely general path integral approach to quantum dynamics. This method consists of an iterative Monte Carlo
evaluation of the discretized path integral expression, propagating the density matrix in a stepwise fashion on a grid selected by a Metropolis Monte Carlo procedure.
This approach circumvents the exponential growth of statistical error with increasing propagation time, while realizing the advantageous scaling of importance sampling in
the grid selection and integral evaluation. Numerical results on complex- and real-time correlation functions in multidimensional model systems over long propagation times
illustrate these features.
Modeling Permeation through Biological Ion Channels: A Physico-Chemical Perspective
12/01/10 4:00pm
MIT Building 56, Room 154
Rob Coalson
Ion channel proteins embed in the cell membrane in such a way that they establish an aqueous pore which spans the membrane. Most ion channels undergo conformational
changes, induced by precise physiological stimuli, between a configuration which allows ions to flow through the aqueous pore (the “open” state) and one that prevents
such flow (the “closed” state). It is challenging to compute, from basic physico-chemical principles, the current of ions that flows down the electrochemical gradient
presented to them when the channel is “open”. This is a complex many-body problem in non-equilibrium statistical mechanics. In principle, the atoms in the protein
channel, mobile ions, water solvent, and lipid bilayer membrane are all dynamical variables, and the time scale is long on the scale of all-atom MD simulations (order
of microseconds to collect full statistics of the permeation process). Fortunately, in many cases, the ions move slowly compared to fluctuations of the surrounding
protein, water and membrane atoms, so a Langevin description of the ionic motion is appropriate. Furthermore, prescriptions exist for computing the inputs into such
a simulation, e.g., the effective potential felt by the mobile ions, ionic diffusion constants, etc.
In the first part of this talk (“pedagogical”), we will discuss the strategy above for “integrating out the bath” (non-ion degrees of freedom), and then pay particular
attention to computational algorithms for propagating the multi-ion primary system via Langevin Eq. simulations. We will also discuss alternative strategies for
computing ion flux through an open ion channel system, including adopting a continuum (density field) description of the ions (Poisson-Nernst-Planck theory), and
discrete state kinetic models of such processes. The latter approach requires that the ion channel present a series of binding sites to the permeant ions, a condition
which is met in many specific cases. If so, states + rates models offer considerably flexibility: they are not restricted by long time scales, other mechanistically
relevant processes such as channel gating can be incorporated into the kinetic scheme, etc.
In the second (“current research”) part of the talk, we will present Langevin dynamics calculations of the type sketched above which have been performed on a recently
crystallized bacterial pentameric ligand-gated ion channel, whose eukaryotic counterparts play an important role in neuronal function. Observations on possible functionally
important effects of the binding of anesthetic molecules to the channel upon its function will also be proffered.
Chemically-Powered Nanomotors
12/08/10 4:00pm
MIT Building 56, Room 154
Raymond Kapral
Biological systems make frequent use of molecular motors to
perform tasks such as active transport of material in the cell,
cell locomotion and biochemical synthesis. Recently, chemists
have fabricated a variety of synthetic nanomotors that use
chemical reactions to effect self propulsion. Because of their
potential applications, such synthetic nanodevices are being
investigated actively. Like their biological counterparts,
these nanomotors operate in the regime where they are subject
to strong molecular fluctuations from the environments in which
they move, and their motion is governed by viscous forces. The
first talk will describe recent work on various types of synthetic
nanomotors, the means by which they move and some of their
possible uses. The second talk will focus on chemically-powered
nanodimer motors. In particular the following topics will be considered:
simulations of their dynamics, microscopic mechanisms for their motion,
how to design motors that beat fluctuations, nonomotor efficienecy and
their collective motions.
Modeling coherent excitation energy transfer in photosynthetic light harvesting
2/9/11 4:00pm
MIT Building 56, Room 154
David Coker
Recent 2D photon-echo experimental evidence suggests that the excitation energy transfer in light harvesting systems occurs coherently rather than via an
incoherent hopping mechanism proposed in many earlier models of the process. More surprisingly, Scholes and co-workers have found evidence for coherent
transfer even at ambient temperature in photosynthetic marine algae [E. Collini et. al, Nature 2010, 463, 644-647]. In this talk we outline an iterative
linearized density matrix (ILDM) propagation approach that can be converged to provide the exact evolution of the multi-state density. We demonstrate the
approach in applications to various system-bath models that include tens of quantum states and hundreds of bath modes. We report a recent study of the coherent
exciton transfer dynamics in phycocyanin PC645 from Chroomonas CCMP270 under ambient conditions (T=294K) with a multi-state system-bath dissipative model
hamiltonian. The numerical results indicate that the oscillatory population beating lasts more than 400 fs and shows strong coherence between the DBV dimer
and DBV-MBV bilin chromophores, an observation that agrees well with the experimental findings. Moreover, the quantum beating survives for nearly ten periods,
and this long lived coherent superposition is likely to be responsible for providing a mechanism for the system to avoid excitation trapping and localization,
providing sufficient time for the excitation to explore the entire complex and reach the acceptor, and thus has the potential to enhance the harvesting efficiency.
Our calculations explore the influence of high and low frequency structures in the model environmental spectral density on the persistence of quantum coherence
in these systems. We also explore the influence of various models of correlation between bath modes and the effects of such correlations on the coherence decay time.
TBD
2/16/11 4:00pm
MIT Building 56, Room 154
Ulrich Hansmann
Bose Einstein condensation of Exciton/Polaritons in organic thin film quantum wells: theory and experimental prospects
4/06/11
Harvard, Pfizer Lecture Hall (12 Oxford St)
Eric Bittner
Recent experiments on thin-film microcavities give evidence of Bose condensation of exciton-polariton states. Inspired by these observations, we consider the possibility that such exotic "half-light/half matter" states could be observed in thin-film organic semiconductors where the oscillator strength is generally stronger than in inorganic systems. In my talk, I present a theoretical model and simulations of macroscopic exciton-polariton condensates in anthracene thin-films sandwiched within a micro-meter scale resonant cavity and establish criteria for the conditions under which BEC could be achieved in these systems.
Electronic and Optical Processes in Organic Semiconductors: The Case of Organic Solar Cells
05/04/11 4:00pm
MIT Building 4, Room 149
Jean Luc Brédas
This presentation seeks to provide a basic understanding of the most
important electronic and optical processes taking place in devices based on
organic semiconductors, by taking organic solar cells as an example.
We will address in particular issues related to:
(i) photon absorption and exciton migration;
(ii) exciton dissociation and charge separation at the organic-organic
interface; and
(iii) charge transport.
Understanding and predicting materials for energy: Insight from quantum simulations
05/11/11 4:00pm
MIT Building 4, Room 149
Guilia Galli
The understanding and prediction of fundamental properties of materials and molecular systems from the basic equations of quantum mechanics is an important component in the design of materials for energy applications. However the field of quantum simulations is still in its infancy and formidable theoretical and computational challenges lay ahead. After a general introduction of current first principles theories and techniques to describe molecules and condensed phases, we will discuss recent progress in predicting optical and thermoelectric properties of nanostructured materials, as well as some deceivingly simple fluids, i.e. water and hydrocarbons. We will then address open problems in quantum simulations of matter, especially the complex interplay between theory, computation, and experiment.
Past Schedules