Download C500 Projects

Iyengar, 1
SRINIVASAN S. IYENGAR
Molecular structure and motion determines properties including spectroscopy and reactivity. We study
properties of fundamental chemical systems1,2,3,4 of interest in atmospheric0 and biological5,6 chemistry,
and nano-technology using theoretical and computational chemistry. We also specialize in the
development of new computational methods7,8,9,10,11 and tools to facilitate efficient and accurate study of
chemical dynamics in such large systems. This puts our research efforts on the interface of chemistry,
computational physics and applied mathematics.
Simulating the dynamics of large biological and chemical systems using computational chemistry has a
long history. It is worthwhile to briefly overview this area here, before we embark into a detailed
exposition of our specific research interests. We start with describing a computational tool known as
molecular dynamics. Here, the molecular geometry (or conformation) is allowed to naturally evolve in
time using the inter-atomic forces between the various constituents of the system. The physical law that
governs such a dynamics is given by the Schrödinger equation. However, this equation is very difficult to
solve even on today’s ultra-fast computers. This has lead researchers in the past to develop models to
describe the inter-atomic potential energy for large systems, and these models, being easier to compute as
compared to the Schrödinger equation, have been used to study the structure and dynamics of molecules
and determine their properties. However, these models have a number of limitations that severely restrict
their applicability to systems where chemical reactions do not occur. Since studying chemical reactions
and their feasibility is an important part of understanding chemistry, the applicability of such methods is
rather limited.
We specialize in using the Schrödinger equation, and more specifically the concepts of quantum
mechanics, to computationally perform molecular dynamics. Such approaches that use the Schrödinger
equation in their development are called ab initio molecular dynamics (AIMD) approaches. (The words
ab initio are derived from Latin: ab stands for ‘from’ and initio means ‘first-principles’; thus the term ab
initio is used to characterize computational methods that are derived from first-principles.) And due to the
rather intricate and computationally expensive methodology involved in solving the Schrödinger
equation, we develop approximations and alternate approaches to facilitate the study of large systems.
Specifically, our group has recently developed a new approach to ab initio molecular dynamics.7-9
Our computational methodology7-11 provides a computationally efficient framework to accurately model
chemical reaction dynamics in large systems. In quantum mechanics, the wavefunction describing the
electrons and the state of the nuclear degrees of freedom,
which could be described using the Cartesian coordinates of
the nuclei or quantum description of the same, together
represent the structure of the molecular system. Our approach
helps propagate in time the complete state of a molecular
system, in a computationally efficient fashion and hence
makes the treatment of large,5,6,9,10,12 have been successfully
treated using this approach. This approach is called Quantum
Wavepacket Ab Initio Molecular Dynamics (QWAIMD). The
approach allows a massively parallel-computational
implementation; computational treatment of simultaneous
dynamics of electrons and nuclei in medium sized chemical
systems can now be treated over 100s of computer processors
Figure 1: Active site of lipoxygenase
leading to an efficient computational methodology.
Furthermore, our approach allows the flexibility to treat a
Iyengar, 2
subset of the nuclei in a quantum mechanical fashion while simultaneously studying the dynamical
evolution of the electrons with the majority of nuclei treated in a classical fashion. (See Refs. 1-3 for
details on the theoretical aspects.)
Why would one want to treat "some nuclei using quantum mechanics"? It turns out that this problem is of
relevance in many biological enzyme problems and also in atmospheric chemical problems, the study of
which we are now actively pursuing. As an example, of our current study, Figure 1 the represents the
active site of the enzyme "lipoxygenase" where the quantum mechanical nature of a transferring hydrogen
atom dictates the kinetics. Such interesting chemistry is also found in many atmospheric chemical
problems as a result we are currently applying this methodology to problems in biological chemistry and
atmospheric chemistry. In addition, for problems where a quantum mechanical treatment of nuclei is not
necessary, as part of a collaborative effort we have invented another powerful new ab initio molecular
dynamics approach that enables the study of chemical dynamics in medium-to-large size systems.
Systems with more than 100 atoms can be routinely handled within this scheme at an accurate level of
theory. The approach is based on the simultaneous dynamics of the electronic degrees of freedom along
with the nuclear degrees of freedom (treated classically). The method is called Atom-centered Density
Matrix Propagation (ADMP). (See for example Refs. 10-13 for details on the theoretical aspects.) The
atom-centered density matrix propagation (ADMP) approach13,10 is a new (and already established)
approach to ab initio molecular dynamics that was pioneered by us and is being used by a number of
groups around the world. In our group, we work on further development of such approaches to facilitate
the study of cutting-edge chemical problems as described in the following sections.
We are also currently modifying and generalizing the Quantum Wavepacket Ab Initio Molecular
Dynamics (QWAIMD) approach described above to treat the transport of excess electrons and flux of
electron densities through molecular wires and other nano-scale devices. Key theoretical challenges
remain here which we are currently in the process of surmounting. Some of the important issues that we
are currently focusing our attention on here, include the accurate description of an electronic flux through
a molecular wire subjected to open system boundary conditions.
The QWAIMD also presents many interesting problems from a scientific computing point of view which
allows graduate students and post-doc in our group to become familiar with state of the art ideas in
numerical analysis. We develop our own numerical methodologies and associated computational models.
Another active area of research in our group involves the use
of ab initio molecular dynamics (with and without nuclear
quantum effects) to study the spectroscopic properties of
small molecular clusters. As part of a team of collaborators
we have recently noted that the hydrated proton in a water
cluster has an amphiphillic (hydrophobic+hydrophillic)
character. These results were obtained originally from ab
initio and empirical molecular dynamics and later confirmed
by experiment. These will lead to profound implications in
biological and atmospheric problems where proton transfer
plays a critical role. We have also shown that dynamics can
play a critical role in determining the vibrational spectra of
such small "in-flux" clusters.
Figure 2: Amphiphilicity of the hydrated
proton
Thus, in our group, a graduate student will have the
opportunity to learn and understand many different aspects of theoretical chemistry, including
quantum mechanics, classical mechanics, molecular dynamics, applied mathematics and
computational algorithm development. Many of the projects that are described below surmount
Iyengar, 3
traditional boundaries of chemistry and physics, and will expose the student to important concepts in
theoretical/computational physics and applied mathematics. You will also learn to apply these important
concepts to develop new computational and theoretical methods resulting in state-of-the-art
computational tools. Below a mixed set of fundamental and applied projects are provided.
1. Quantum Dynamical Effects in Biological Enzyme
kinetics: The complete and accurate study of enzyme
reactions is an important problem in biological sciences. As
noted in the introductory section above, the quantum
mechanical nature of nuclei can contribute significantly to
the kinetics. Experimentally, this is seen through an
unexpectedly large primary and secondary kinetic isotope
effect14,15,16 and these have been thought to be due to
nuclear quantum effects, such as hydrogen tunneling.14
What does all this mean? General chemistry tells us that
there exists an exponential relation between temperature
and the reaction rate constant. The so-called Arrhenius rate
expression. In some enzymes, such as the one in Figure 1
Figure 3: The existence of “proton
above, what is experimentally found is that the rate
orbitals” in lipoxygenase can be
constant does not depend on temperature at all. The
studied using our QWAIMD
hydrogen atom is an extremely light atom and shows the
approach. (The pink and green p-like
propensity to display quantum behavior, more so than any
orbital is a proton and not an
other atom, and it is believed that this might be the reason
electronic p-orbital!!)
for this experimental result. However, the sheer size of
enzyme systems makes time-dependent quantum dynamical studies impossible. Hence, approximate
methods to examine such critical problems are essential. Our preliminary results indicate that indeed
quantum mechanical nature of the hydrogen atom can contribute to this result and we find that
“proton orbitals” such as the one in Figure 3 do exists. These orbitals facilitate “hydrogen tunneling”
thus leading to unexpected experimental results. The student will have the opportunity to study this
problem using our newly developed QWAIMD approach and its variations. The computational
method is based on a partitioning scheme and will involve QM/MM (quantum
mechanics/molecular mechanics) generalization of our procedure. This involves the seamless
extension of our current formalism into the QM/MM scheme and will involve understanding
the current formalism followed by appropriate generalization to treat large systems.
2. Molecular electronics through quantum dynamics: The possibility of
silicon based electronic chip technology coming close to its inherent
physical limitations has given rise to the new area of molecular
electronics. Here single molecules are connected across electrodes and
treated as potential transistors, capacitors and molecular switches, that is
potential components of a novel, futuristic quantum computer. As part of
a C500 project the student will have the opportunity to make important
contributions and participate in critical steps required to calculate the
conductivity of electrons across molecular wires, connected across
electrodes using generalizations of our QWAIMD approach.
Figure 4: A carbon
nanotube molecular wire
3. Quantum dynamical and ab initio dynamical effects in critical atmospheric chemical problems:
Topics on ozone control: The projects discussed below are fundamental in atmospheric chemistry
and are related to ozone production. These projects are in collaboration with the Philip Stevens
Iyengar, 4
Group and the Caroline Jarrold Group. As part of the C500 project, the students will have the
opportunity to study fundamental atmospheric chemistry problems using state of the art theory.7,8,10,11
The problem of photochemical air pollution remains a serious threat to human health and welfare.
Ozone, the primary component of photochemical smog, forms from chemical reactions involving
nitrogen oxides (NOx) and volatile organic compounds (VOCs) emitted from transportation, energy
production, and industrial activities. Despite years of control legislation, many areas in the United
States have consistently exceeded the National Ambient Air Quality Standard for ozone. Hence,
recent strategies on ozone control are focusing on regional ozone production. The dominant natural
emission of isoprene from trees can contribute significantly to the production of ozone through its
high reactivity with the hydroxyl radical (OH), the main oxidant in the atmosphere. Ozone reduction
strategies depend on an accurate understanding of isoprene chemistry under a variety of conditions.
As part of this C500 project the reactivity and spectroscopy of OH radical with isoprene in the
presence of water will be studied in detail using QWAIMD. A side-project that automatically arises
here is as follows:
a) Treatment of hydroxide water clusters: This project deals with the structure, stability
energetics and spectroscopic properties of hydroxide water clusters. We have recently found
that the hydroxide ion can display non-standard bond arrangements when surrounded by
water molecules. The central oxygen atom displays an anamolous propensity to be
pentavalent. However, we also find that properties such as the vibrational frequencies which
can be determined experimentally, are susceptible to the fact that there exists many other
geometries (with similar energetics) that support a central tetrahedral oxygen; hence the
measurable spectrum is an average of all these “similar” geometries. Dynamics plays an
important role in this spectral averaging process. We are in the midst of predicting the
spectroscopic properties of such clusters and make important connections to experiment. The
student will have an opportunity to investigate this further using ab initio dynamics.
4. Quantum dynamical effects on spectroscopy: NMR and IR spectroscopy: As noted in the
introductory section, we have recently discovered1-3 that the protonated species in water (that is, the
hydronium ion) has “hydrophobic” and “hydrophilic” sites. The hydrophobic end of the hydronium is
close to its lone pairs. Due to this reason, we have found that a hydroniuim ion tends to reside on the
surface of a cluster of water molecules and the surface of a water/vacuum interface. This is in contrast
to conventional chemical wisdom where the protonated species is expected to be solvated at the
center of a cluster of water molecules. Figure 2 shows a 21-water protonated cluster that is found in
upper stratosphere. The protonated species here is an H2O-H3O+ that is a solvated hydronium and is
called a Zundel ion. Note that this protonated species is on the surface of the cluster. As one can
imagine this is an extremely fundamental observation and has important implications which has lead
to many experiments (by other groups) designed to test this phenomenon.
In addition, we are one of the first groups to point out that this system has unique spectroscopic
properties1 which are essentially dictated by the dynamics of the water molecules. As part of the C500
project, the student will have the opportunity to conduct similar studies on NMR coupling constants.
NMR is an extremely fundamental tool in chemistry and in biochemistry. Do quantum nuclear effects
and dynamical effects play a critical role in NMR coupling constants? Specific systems that will be
considered are DNA base pairs and protein anti-parallel beta sheets.
5. Non-adiabatic Reaction Dynamics of Complex Systems. In many biological and chemical
systems, reactions can involve multiple electronic states. This is particularly the case for
photochemical reaction. For example, in rhodopsin (Figure 5), which is found in the mammalian eye,
the cis-trans isomerization of a protonated Schiff base called retinal causes a conformational change
leading to vision. This is among the fastest known chemical reactions. Detailed quantum mechanical
study of such processes constitutes an unsolved problem in computational biophysical chemistry. The
incoming graduate student will have the opportunity here to first fully understand the concepts behind
Iyengar, 5
our quantum ab initio dynamics formalism. This project
is a development project and will involve training in
computation, mathematics and chemical-physics.
6. Generalization of ab initio quantum dynamics to more
accurate electronic structure treatments. The student
involved in this project will have a great opportunity to
study electronic structure theory and dynamics, two
major areas of quantum chemistry. The basic idea behind
this project is as follows. In many important and practical
problems, such as the search for high-energy low-density
rocket fuels, theoretical treatments are required to predict
Figure 5: Active site of Rhodopsin
properties with great accuracy. Currently, ab initio
dynamics can provide accurate results in the vicinity of 1 kcal/mol, but to go beyond this a
fundamentally different approach is required. We will develop first-principles dynamics approaches
that will use highly accurate post-Hartree-Fock methods in our dynamics scheme. This is project
involves a good deal of theoretically development, and the interested student will have a great
opportunity in participating in the development of a novel computational formalism.
7. Quantum-mechanics/molecular mechanics (QM/MM) and continuum model generalizations for
Quantum Wavepacket Ab Initio Molecular Dynamics. In the framework described above and in
project 1, only a portion of the full enzyme can be treated due to the computational complexity of
QWAIMD. As a result, the bulk of the enzyme system will not be treated this way. The student will
have an opportunity to modify this and construct more realistic models to the enzyme. This project
can be coupled with project 1 above, but that is not a requirement.
Selected Publications
1
The Properties of Ion-Water Clusters. I. The Protonated 21-Water Cluster, S. S. Iyengar, M. K.
Petersen, C. J. Burnham, V. E. Teige and G. A. Voth, "". J. Chem. Phys. 123 , 084309 (2005)
2
The Hydrated Proton at the Water liquid/vapour interface Behavior, M. K. Petersen, S. S. Iyengar,
T. J. F. Day, and G. A. Voth, J. Phys. Chem. B 108 , 14804 (2004).
3
On the Amphiphilic Behavior of the Hydrated Proton: An Ab Initio Molecular Dynamics Study,
S. S. Iyengar, T. J. F. Day, and G. A. Voth, Int. J. Mass Spectrometry 241, 197-204 (2005).
4
Dynamical effects on vibrational and electronic spectra of hydroperoxyl water clusters, S. S.
Iyengar, J. Chem. Phys. 123 , 084310 (2005)
5
A Molecular Dynamics and Ab Initio Molecular Dynamics Study of Ligand Binding in Ancestral
hemoglobins in Archaea, James S. Newhouse, Tracey Allen K. Freitas, Maqsudul Alam, Gregory A.
Voth and Srinivasan S. Iyengar, Proc. Natl. Acad. Sci., In Preparation.
6
Hybrid Ab-Initio/Empirical Molecular Dynamics: Combining the ONIOM Scheme with the
Atom-centered Density Matrix Propagation (ADMP) Approach, N. Rega, S. S. Iyengar, G. A. Voth,
H. B. Schlegel, T. Vreven and M. J. Frisch, J. Phys. Chem. B 108 4210-4220 (2004).
Iyengar, 6
7
Quantum Wavepacket Ab Initio Molecular Dynamics: An approach to study quantum dynamics
in large systems, S. S. Iyengar and J. Jakowski, J. Chem. Phys. 122 114105 (2005).
8
Ab Initio Dynamics with wave-packets and density matrices, S. S. Iyengar, Theo. Chem. Accts.
Special issue on "New Perspectives in Theoretical Chemistry” In Press. (URL:
http://www.indiana.edu/%7Essiweb/papers/TCA-final.pdf)
9
Computational Improvements to Quantum Wavepacket Ab Initio Molecular Dynamics using a potentialadapted, time-dependent deterministic sampling technique, J. Jakowski, I. Sumner and S. S. Iyengar, Journal of
Chemical Theory and Computation Submitted. (URL: http://www.indiana.edu/%7Essiweb/papers/qcadmp2.pdf)
10
Effect of time-dependent basis functions and their superposition error on Atom-centered Density
Matrix Propagation (ADMP): Connections to Wavelet theory of multi-resolution analysis,
Srinivasan S. Iyengar and Michael J. Frisch, J. Chem. Phys. 121, 5061 (2004).
11
Ab Initio Molecular Dynamics: Propagating the Density Matrix with Gaussian orbitals. II.
Generalizations based on Mass-weighting, Idempotency, Energy Conservation and
Choice of Initial Conditions, S. S. Iyengar, H. B. Schlegel, J. M. Millam, G. A. Voth, G. E. Scuseria, M. J. Frisch,
J. Chem. Phys. 115, 10291 (2001); Atom-centered Density Matrix Propagation (ADMP): Generalizations using
Bohmian Mechanics, S. S. Iyengar, H. B. Schlegel, G. A. Voth, J. Phys. Chem. A. 107, 7269 (2003). (Festschrift in
honor of Professor Donald J. Kouri)
12
Ab Initio Molecular Dynamics: Propagating the Density Matrix with Gaussian orbitals. III.
Comparison with Born-Oppenheimer Dynamics, H. B. Schlegel, S. S. Iyengar, X. Li, J. M. Millam, G.
A. Voth, G. E. Scuseria, M. J. Frisch, J. Chem. Phys. 117, 8694 (2002).
13
Ab Initio Molecular Dynamics: Propagating the Density Matrix with Gaussian Orbitals, H. B.
Schlegel, J. M. Millam, S. S. Iyengar, G. A. Voth, A. D. Daniels, G. E. Scuseria, M. J. Frisch, J. Chem.
Phys. 114, 9758 (2001).
Enzyme dynamics and hydrogen tunnelling in a thermophilic alcohol dehydrogenase, A.
Kohen, R. Cannio, S. Bartolucci, and J. P. Klinman, Nature 399, 496 (1999).
14
A link between protein structure and enzyme catalyzed hydrogen tunneling, B. J.
Bahnson, T. D. Colby, J. K. Chin, B. M. Goldstein, and J. P. Klinman, Proc. Natl. Acad. Sci.
U.S.A. 94, 12797 (1997).
15
Hydrogen Tunneling in Enzyme Reactions, Y. Cha, C. J. Murray, and J. P. Klinman, Science
243, 1325 (1989).
16