Thomas Miller

Postdoctoral Associate Chandler and Miller Research Groups Room 15, Pitzer Center for Theoretical Chemistry Department of Chemistry University of California at Berkeley Berkeley, CA 94720 Phone: (510) 643-1169 Fax: (510) 643-1566 Email: tfmiller@berkeley.edu

Exciting news:

Research Interests

Nature exhibits dynamics that span extraordinary ranges of space and time. In some cases, these dynamical hierarchies are well separated, simplifying their understanding and description. To aim a bowling ball, for example, it is not necessary to explicitly consider the motion of its electrons. But chemistry and biology are replete with examples of dynamically coupled scales. Electron- and proton-transfer reactions couple intrinsically quantum-mechanical and classical-mechanical motions. Molecular motors convert atomic-scale reactions into nano-scale motion and work. And signaling pathways use molecular recognition processes to regulate activities at the cellular level. Understanding processes that bridge dynamical hierarchies is a fascinating and ongoing challenge. My research focuses on the development of new theoretical methods to simulate and understand complex dynamics in chemistry and biology.
• 

  Simulating complex dynamics in soft matter systems

  In a recent collaboration with Eric Vanden-Eijnden, we applied solvent coarse-graining and the string method to study the hydrophobiccollapse of a hydrophobic chain.Using computer simulations of more than 100,000 atoms, we obtained the mechanism for the hydrophobic collapse of an idealized hydrated chain by projecting the atomistic water molecule positions onto a cubic grid with angstrom-level resolution. With this coarse-grained representation for the water, the string method was used to compute a path of maximum likelihood (minimum free energy) for the collapsing chain. These simulations provided the first atomistic demonstration that the rate-limiting step in hydrophobic collapse is a collective solvent motion that leads to the formation of a liquid-vapor interface at the hydrophobic surface. Furthermore, by showing that both the energetics and dynamics of the atomistic water in this system is quantitatively reproduced with our grid-based representation, we obtained a proof of principle for the coarse-graining of water dynamics in biophysical simulations.

  The figure at right shows the minimum free energy path for the collapse of an 8.6 nm hydrophobic chain in water, obtained in over 130,000 dimensions by coarse-graining all-atom simulations and using the string method in collective variables. Monomers in the idealized chain are shown in red, and regions of low water density are shown in white against the blue background. Above, a single peak in the free energy profile is seen at configuration 22. Below, the configurations of the path in the vicinity of the free-energy peak are shown with configuration numbers indicated in white text.

  Read more at:

  T. F. Miller, III, E. Vanden-Eijnden, and D. Chandler, PNAS, 104, 14559 (2007). pdf
  

• New methods for long timescale and rare event dynamics

  Transition path sampling (TPS) is a general technique for characterizing the dynamics of rare events. However, in its standard implementation (i.e. the shooting algorithm) TPS utilizes real-time molecular dynamics trajectories to propose new reactive paths. This strategy does not efficiently address systems with meta-stable intermediates, and it does not naturally leverage parallel computing resources to access long simulation timescales.

  In a collaboration with Cristian Predescu, I have developed a path integral formulation of TPS that overcomes these difficulties. Dynamical trajectories are represented as chains-of-states and sampled using standard Metropolis Monte Carlo, which naturally eliminates the difficulty of meta-stable intermediates. Large-scale parallelization is achieved with the aide of an additional technique that we call Sliding and Sampling (S&S). The two step S&S process is illustrated in the movie, where the dynamical trajectory is shown as a sequence of positions as a function of time. In the first step, the ensemble of trajectories is divided into time-segments that are independently sampled by different computer processors. Then, the time-segments are redistributed among the various processors by sliding the sampling domains along the time coordinate. The S&S algorithm is ergodic and relies only on the Markov property of the real-time dynamics. It enables different processors to sample different regions of time, thus enabling large-scale computers to probe dynamics on exceedingly long timescales.

  We have used this path-integral transition path sampling approach to study the phase-change dynamics of the 38-atom Lennard-Jones cluster, and we are pursuing its application in a variety of biological and material science contexts.

  Read more at:

  T. F. Miller, III and C. Predescu, J. Chem. Phys., 126, 144102 (2007). pdf
  

• Quantum dynamics in the condensed phase

  As part of the LBNL Helios project, we have recently been focused on simulating coupled electronic and nuclear dynamics in condensed phase systems. This task requires a model that exhibits both of the following properties. Firstly, it must provide a suitably accurate description for fast, highly quantum mechanical dynamics of the electrons. Secondly, the model dynamics must rigorously preserve the quantum Boltzmann distribution so that long simulations can be performed to access the slower dynamics of the nuclei. Encouraging progress has been made in this effort, and more details will be posted shortly.

Education

• 2002-2005: D. Phil., Chemistry, University of Oxford
  
National Science Foundation Graduate Research Fellow
Thesis: Quantum simulation of biological molecules
Adviser: David C. Clary
Collaborators: David E. Manolopoulos and Anthony J. H. M. Meijer
• 2000-2002: M. Phil., Chemistry, University College London
  
British Marshall Scholar
Thesis: Torsional path integral method for the quantum simulation of large molecules
Adviser: David C. Clary

• 1996-2000: B. Sc., Chemistry and Mathematics, Texas A&M University
  
Research adviser: Michael B. Hall

Publications

• [19] "Solvent coarse-graining and the string method applied to the hydrophobic collapse of a hydrated chain"
T. F. Miller, III, E. Vanden-Eijnden, and D. Chandler, PNAS, 104, 14559 (2007). pdf

• [18] "Sampling diffusive transition paths"
T. F. Miller, III and C. Predescu, J. Chem. Phys., 126, 144102 (2007). pdf

• [17] "Quantum simulation of a hydrated noradrenaline analog with the torsional path integral method"
T. F. Miller, III and D. C. Clary, J. Phys. Chem. A, 110, 731 (2006). pdf
Special issue in honor of Prof. Donald G. Truhlar.

• [16] "Sum rule constraints on Kubo-transformed correlation functions"
B. J. Braams, T. F. Miller, III and D. E. Manolopoulos, Chem. Phys. Lett., 118, 179 (2006). pdf

• [15] "Quantum diffusion in liquid water from ring-polymer molecular dynamics"
T. F. Miller, III and D. E. Manolopoulos, J. Chem. Phys., 123, 154504 (2005). pdf

• [14] "Collision-induced conformational changes in glycine"
T. F. Miller, III, D. C. Clary, and A. J. H. M. Meijer, J. Chem. Phys., 122, 244323 (2005). pdf

• [13] "Torsional anharmonicity in the conformational thermodynamics of flexible molecules"
T. F. Miller, III and D. C. Clary, Mol. Phys., 103, 1573 (2005). pdf
Special issue in honor of Prof. John P. Simons.

• [12] "Quantum diffusion in liquid para-hydrogen from ring-polymer molecular dynamics"
T. F. Miller, III and D. E. Manolopoulos, J. Chem. Phys., 122, 184503 (2005). pdf

• [11] Comment on "A centroid molecular dynamics study of liquid para-hydrogen and ortho-deuterium" [J. Chem. Phys. 121, 6412 (2004)]
T. F. Miller, III, D. E. Manolopoulos, P. A. Madden, M. Konieczny, and H. Oberhofer, J. Chem. Phys., 122, 057101 (2005). pdf

• [10] "Quantum free energies of the conformers of glycine on an ab initio potential energy surface"
T. F. Miller, III and D. C. Clary, Phys. Chem. Chem. Phys., 6, 2563 (2004). pdf
Special issue on bio-active molecules in the gas phase.

• [9] "Army ants algorithm for rare event sampling of delocalized nonadiabatic transitions..."
S. Nangia, A. W. Jasper, T. F. Miller, III and D. G. Truhlar, J. Chem. Phys., 120, 3586 (2004). pdf

• [8] "Predicting conformations of biomolecules: Application to a noradrenaline analogue"
T. F. Miller, III and D. C. Clary, J. Phys. Chem. B, 108, 2484 (2004). pdf

• [7] "Torsional path integral Monte Carlo method for calculating the absolute quantum free energy of large molecules"
T. F. Miller, III and D. C. Clary, J. Chem. Phys., 119, 68 (2003). pdf

• [6] "Symplectic quaternion scheme for biophysical molecular dynamics"
T. F. Miller, III, M. Eleftheriou, P. Pattnaik, A. Ndirango, and G. J. Martyna, J. Chem. Phys., 116, 8649 (2002). pdf

• [5] "Torsional path integral Monte Carlo method for the quantum simulation of large molecules"
T. F. Miller, III, and D. C. Clary, J. Chem. Phys., 116, 8262 (2002). pdf

• [4] "Excitation of the symmetry forbidden bending mode in molecular photoionization"
J. S. Miller, E. D. Poliakoff, T. F. Miller, III, A. P. P. Natalense, and R. R. Lucchese, J. Chem. Phys., 114, 4496 (2001). pdf

• [3] "Structural and bonding trends in platinum-carbon clusters"
T. F. Miller, III, and M. B. Hall, J. Amer. Chem. Soc., 121, 7389 (1999). pdf

• [2] "Linear semibridging carbonyls. 6. Structure and bonding in the dimers of 17-electron tantalum..."
T. F. Miller, III, D. L. Strout, and M. B. Hall, Organometallics, 17, 4164 (1998). pdf

• [1] "Structure and stability of palladium-carbon cations"
D. L. Strout, T. F. Miller, III, and M. B. Hall, J. Phys. Chem. A, 102, 6307 (1998). pdf

Back to the Chandler Group homepage

Last changed: 2/3/2008

For Josh