Greater Boston Area Theoretical Chemistry Lecture Series
2022-2023 Speaker Schedule
Making sense of polarons and electron localization in solids
11/2/22 4:00 pm EST at MIT building 32 room 124
Feliciano Giustino
Polarons are fascinating realizations of emergent quasiparticles resulting from the interaction between fermions and bosons [1]. In crystals, polarons form when electrons or holes become dressed by phonons in the form of lattice distortions. In the presence of weak electron-phonon interactions, polarons behave like conventional Bloch waves with heavier effective masses. In the presence of strong interactions, on the other hand, polarons become localized wavepackets and profoundly alter the transport, electrical, and optical properties of the host material. In this talk I will describe recent explorations of polaron physics from the point of view of first-principles atomic-scale calculations [2,3]. In the first part of the talk, I will provide a general introduction to the electron-phonon problem, and show that we can now perform predictive non-empirical calculations of many materials properties relating to electron-phonon physics using density functional theory and many-body Green's functions methods [4,5]. In the second part, I will present a recently-developed ab initio theory of polarons and its applications to simple ionic insulators. I will make the connection with historical developments such as Feynman's polaron theory, and discuss implications on our current understanding of electron localization in solids.
References
[1] F. Giustino, Rev. Mod. Phys. 89, 015003 (2017).
[2] W. H. Sio, C. Verdi, S. Poncé, and F. Giustino, Phys. Rev. Lett. 122, 246403 (2019).
[3] J. Lafuente-Bartolome, C. Lian, W. H. Sio, I. G. Gurtubay, A. Eiguren, and F. Giustino, Phys. Rev. Lett. 129, 076402 (2022).
[4] S. Poncé, W. Li, S. Reichardt, and F. Giustino, Rep. Prog. Phys. 83, 036501 (2020).
[5] M. Zacharias and F. Giustino, Phys. Rev. Res. 2, 013357 (2020).
Rare Conformational transitions in Biomolecular Systems
12/7/22 4:00 pm EST at MIT building 32 room 124
Benoît Roux
Classical molecular dynamics (MD) simulations based on atomic models play an increasingly important role in a wide range of applications in physics, biology and chemistry. The approach consists of constructing detailed atomic models of the macromolecular system, and having described the microscopic forces with a potential function, using Newton's classical equation, F=MA, to literally "simulate" the dynamical motions of all the atoms as a function of time. The calculated trajectory, though an approximation to the real world, provides detailed information about the time course of the atomic motions, which is impossible to access experimentally. While great progress has been made, producing genuine knowledge about biological systems using MD simulations remains enormously challenging. Among the most difficult problems is the characterization of large conformational transitions occurring over long-time scales. Issues of force field accuracy, the neglect of induced polarization, in particular, are also a constant concern. Transition path theory offers a powerful paradigm for mapping the conformational landscape of biomolecular systems is to combine free energy methods, string method, transition pathway techniques, and stochastic Markov State Model based massively distributed simulations.[1-5] These concepts will be illustrated with a few recent computational studies of biomolecular systems.
References
[1] Pan, A. C., Sezer, D. & Roux, B. Finding transition pathways using the string method with swarms of trajectories. J. Phys. Chem. B 112, 3432-3440, (2008).
[2] Pan, A. C. & Roux, B. Building Markov state models along pathways to determine free energies and rates of transitions. J. Chem. Phys. 129, 064107, (2008).
[3] Roux, B. String Method with Swarms-of-Trajectories, Mean Drifts, Lag Time, and Committor. J. Phys. Chem. A 125, 7558-7571, (2021).
[4] Roux, B. Transition rate theory, spectral analysis, and reactive paths. J. Chem. Phys. 156, 134111, (2022).
[5] He, Z., Chipot, C. & Roux, B. Committor-Consistent Variational String Method. J. Phys. Chem. Lett. 13, 9263−9271, (2022).
Physical Principles of Protein Phase Separation in Biomolecular Condensates Explored by Theory and Computation
1/25/23 4:00 pm EST at MIT building 32 room 124
Hue Sun Chan
Compartmentalization at the cellular and sub-cellular levels is essential for biological functions. Some of the intra-organismic compartmentalized bodies are devoid of a lipid membrane (hence sometimes called “membrane-less organelles”) and possess material properties similar to those of mesoscopic liquid droplets. Referred to collectively as “biomolecular condensates”, their assembly is underpinned to a significant degree by liquid-liquid phase separation (LLPS) of intrinsically disordered proteins (IDPs), intrinsically disordered regions (IDRs) of proteins, globular protein domains, and nucleic acids—though other physicochemical processes also contribute [1]. To gain physical insights, our group has developed analytical theories [2]—including Flory-Huggins formulations, random phase approximation [3], Kuhn-length renormalization [4], and new formulations of field-theoretic simulation that account for both short- and long-spatial-range interactions [5,6]—as well as coarse-grained explicit-chain molecular dynamics models for sequence-specific LLPS of IDPs/IDRs [7]. This effort has elucidated the effect of sequence charge pattern [2-7], π-related interactions [7], pH, salt [4], and osmolytes [8] on biomolecular LLPS. Moreover, our results point to a “fuzzy” mode of molecular recognition by charge pattern matching [9] modulated by excluded-volume effects [10], which is relevant to deciphering how different IDP species may de-mix upon LLPS to achieve functional sub-compartmentalization. A first step has also been taken toward rationalizing the temperature and pressure dependence of LLPS by empirical and atomic models of solvent-mediated hydrophobic interactions [11] and the interplay between stoichiometric and less-specific multivalent interactions in the assembly of biomolecular condensates [12]. Biological ramifications of our findings will be discussed, including how the pressure sensitivity of an in vitro model of postsynaptic densities might offer biophysical insights into pressure-related neurological disorders in terrestrial vertebrates [13].
References
[1] Lyson, Peeple, Rosen (2021) Nat Rev Mol Cell Biol 22:215-235.
[2] Lin, Forman-Kay, Chan (2018) Biochemistry 57:2499-2508.
[3] Lin, Forman-Kay, Chan (2016) Phys Rev Lett 117: 178101.
[4] Lin et al. (2020) J Chem Phys 152:045102.
[5] Lin et al. (2023) In: Methods in Molecular Biology (Springer-Nature), Vol. 2563, Ch. 3, pp.51-94.
[6] Wessén et al. (2022) J Phys Chem B 126:9222-9245.
[7] Das et al. (2020) Proc Natl Acad Sci USA 117:28795-28805.
[8] Cinar et al. (2019) J Am Chem Soc 141:7347-7354.
[9] Lin et al. (2017) New J Phys 19:115003.
[10] Pal et al. (2021) Phys Rev E 103:042406.
[11] Cinar et al. (2019) Chem Eur J 25:13049-13069.
[12] Lin et al. (2022) Biophys J 121:157-171.
[13] Cinar et al. (2020) Chem Eur J 26:11024-11031.
What can vibrational structure theory learn (or steal) from electronic structure theory?
3/1/23 4:30 pm EST at MIT building 32 room 144
Tim Berkelbach
Approximate vibrational structure methods are successful for systems whose dynamics are nearly harmonic, but they struggle in the presence of strong anharmonicities, which correspond to interactions between normal-mode vibrations. The same situation occurs in electronic structure theory, where state-of-the-art methods now allow the simulation of the electronic structure of large molecules and solids, even with strong Coulomb interactions. Can similar methods be applied to vibrations? In this talk, I’ll describe similarities and differences between vibrational structure and electronic structure, culminating in the identification of two electronic structure methods that are especially promising for vibrational structure applications: heat-bath configuration interaction (HCI) and dynamical mean-field theory (DMFT). I’ll present the theory behind vibrational HCI and vibrational DMFT, as well as applications to the vibrational structure of large molecules and solids.
Simulating excitons and multiexcitons under confinement
3/15/23 4:30 pm EST at MIT building 32 room 144
Eran Rabani
The description of carrier dynamics in spatially confined semiconductor nanocrystals (NCs), which have enhanced electron-hole and exciton-phonon interactions, is a great challenge for modern computational science. These NCs typically contain thousands of atoms and tens of thousands of valence electrons with discrete spectra at low excitation energies, similar to atoms and molecules, that converge to the continuum bulk limit at higher energies. Computational methods developed for molecules are limited to very small nanoclusters, and methods for bulk systems with periodic boundary conditions are not suitable due to the lack of translational symmetry in NCs.
In this talk I will review on our recent efforts in developing a unified atomistic model based on the semiempirical pseudopotential approach, parametrized by first-principle calculations and validated against experimental measurements, to describe two of the main nonradiative relaxation processes of quantum confined excitons: exciton cooling and Auger recombination. I will focus on the description of both electron-hole and exciton-phonon interactions and discuss the role of size, shape, and interfacing on the electronic properties and dynamics for II-VI and III-V semiconductor NCs.
Ab-initio solid state chemistry as a new frontier of theory
3/22/23 4:30 pm EST at MIT building 32 room 144
Dominika Zgid
The search for new materials is at the core of the technological advancement of our society. While many newly synthesized materials can be analyzed by current quantum chemical techniques, mostly based on the density functional theory (DFT), there is a large number of materials that cannot be treated successfully by existing methodologies. This is mostly due to the presence of strong electron correlation, relativistic effects, and disorder. These materials require a post-DFT description that explicitly includes electron-electron interactions.
[In my talk, I will discuss current theoretical challenges in the study of solid state materials and I will describe my group’s contributions to the development of post-DFT methods. In the first part, I will present the newest relativistic methodologies for solids. In the second part, I will talk about the treatment of strongly correlated electrons residing in d- and f-orbitals of crystals with transition metals. Finally, I will sketch future directions for computational ab-initio solid state chemistry.
Part1: Excited-state dynamics simulations of complex systems
Part2: Pushing (QM/MM) boundaries: Nonadiabatic dynamics simulations of solvent-supported electronic states
4/5/23 4:30 pm EST at MIT building 32 room 144
William J. Glover
Three pillars of theoretical chemistry are quantum mechanics (governing the physics of light particles), statistical mechanics (connecting microscopic interactions to macroscopic observables), and chemical dynamics (describing the motions of atoms on potential energy surfaces). As theoretical tools have developed, simultaneously with exponential increases in computational power, the boundaries between these three fields have blurred. This is evident especially for studies of photon-induced molecular processes, where it is now possible to simulate the real-time dynamics of chromophores and their environment, including a large quantum mechanical region of hundreds of atoms. In the first part of my talk, I will briefly review the theories and tools for first-principles simulations of excited-state dynamics, with a particular focus on multiscale models of complex systems. In the second part of my talk, I will present our group’s developments in this field, with applications to the simulation of solvent-supported electronic states.
A central problem in chemistry is to understand how solvents affect the properties and reactivity of chemical systems. From an electronic perspective, solute-solvent interactions can have a range of effects, from solvatochromic shifts of excitation energies to the emergence of entirely new solvent-supported states. Examples of the latter are solvated electrons, corresponding to unpaired electrons supported by the solvent but not bound to any single molecule. The aqueous solvated electron e–(aq) is of particular interest since low-energy electrons induce strand breaks in hydrated DNA. Of relevance is the excited-state dynamics of e–(aq) in a pre-solvated state since the ground state is unreactive to DNA. Interestingly, current theories (based on one-electron models) are unable to rationalize the ultrafast (~50 fs) excited-state decay of e–(aq) measured in recent ultrafast time-resolved photoelectron experiments. There is thus considerable room for improvement in the theoretical treatment of solvent-supported states. I will present our recent developments in this area that allow for an efficient quantum mechanics/molecular mechanics (QM/MM) embedding description of a solvated system’s dynamics, even when solvent molecules are included in the QM region. Combining our approach with nonadiabatic dynamics simulations, we reconcile theory with experiment and rationalize the ultrafast excited-state lifetime of e–(aq) from an electron-transfer perspective. The new technologies open the door to studying a wide range of solution-phase chemistry where a QM description of the solvent is necessary.
Four Generations of Neural Network Potentials
5/3/23 11:00 am EST on zoom
Jörg Behler
A lot of progress has been made in recent years in the development of machine learning potentials (MLP) for atomistic simulations [1]. Neural network potentials (NNPs), which have been introduced more than two decades ago [2], are an important class of MLPs. While the first generation of NNPs has been restricted to small molecules with only a few degrees of freedom, the second generation extended the applicability of MLPs to high-dimensional systems containing thousands of atoms by constructing the total energy as a sum of environment-dependent atomic energies [3]. Long-range electrostatic interactions can be included in third-generation NNPs employing environment-dependent charges [4], but only recently limitations of this locality approximation could be overcome by the introduction of fourth-generation NNPs [5,6], which are able to describe non-local charge transfer using a global charge equilibration step. In this talk an overview about the evolution of high-dimensional neural network potentials will be given along with typical applications in large-scale atomistic simulations.