Annual Review of Physical Chemistry - Volume 71, 2020

We review recent advances in the characterization of electronic forms of energy transport in emerging semiconductors. The approaches described all temporally and spatially resolve the evolution of initially localized populations of photogenerated excitons or charge carriers. We first provide a comprehensive background for describing the physical origin and nature of electronic energy transport both microscopically and from the perspective of the observer. We introduce the new family of far-field, time-resolved optical microscopies developed to directly resolve not only the extent of this transport but also its potentially temporally and spatially dependent rate. We review a representation of examples from the recent literature, including investigation of energy flow in colloidal quantum dot solids, organic semiconductors, organic-inorganic metal halide perovskites, and 2D transition metal dichalcogenides. These examples illustrate how traditional parameters like diffusivity are applicable only within limited spatiotemporal ranges and how the techniques at the core of this review,especially when taken together, are revealing a more complete picture of the spatiotemporal evolution of energy transport in complex semiconductors, even as a function of their structural heterogeneities.

Various organic reactions, including important synthetic reactions involving C–C, C–N, and C–O bond formation as well as reactions of biomolecules, are accelerated when the reagents are present in sprayed or levitated microdroplets or in thin films. The reaction rates increase by orders of magnitude with decreasing droplet size or film thickness. The effect is associated with reactions at the solution–air interface. A key factor is partial solvation of the reagents at the interface, which reduces the critical energy for reaction. This phenomenon is of intrinsic interest and potentially of practical value as a simple, rapid method of performing small-scale synthesis.

Biological phase separation is known to be important for cellular organization, which has recently been extended to a new class of biomolecules that form liquid-like droplets coexisting with the surrounding cellular or extracellular environment. These droplets are termed membraneless organelles, as they lack a dividing lipid membrane, and are formed through liquid-liquid phase separation (LLPS). Elucidating the molecular determinants of phase separation is a critical challenge for the field, as we are still at the early stages of understanding how cells may promote and regulate functions that are driven by LLPS. In this review, we discuss the role that disorder, perturbations to molecular interactions resulting from sequence, posttranslational modifications, and various regulatory stimuli play on protein LLPS, with a particular focus on insights that may be obtained from simulation and theory. We finally discuss how these molecular driving forces alter multicomponent phase separation and selectivity.

Roaming reactions were first clearly identified in photodissociation of formaldehyde 15 years ago, and roaming dynamics are now recognized as a universal aspect of chemical reactivity. These reactions typically involve frustrated near-dissociation of a quasibound system to radical fragments, followed by reorientation at long range and intramolecular abstraction. The consequences can be unexpected formation of molecular products, depletion of the radical pool in chemical systems, and formation of products with unusual internal state distributions. In this review, I examine some current aspects of roaming reactions with an emphasis on experimental results, focusing on possible quantum effects in roaming and roaming dynamics in bimolecular systems. These considerations lead to a more inclusive definition of roaming reactions as those for which key dynamics take place at long range.

Chromatosomes are fundamental units of chromatin structure that are formed when a linker histone protein binds to a nucleosome. The positioning of the linker histone on the nucleosome influences the packing of chromatin. Recent simulations and experiments have shown that chromatosomes adopt an ensemble of structures that differ in the geometry of the linker histone–nucleosome interaction. In this article we review the application of Brownian, Monte Carlo, and molecular dynamics simulations to predict the structure of linker histone–nucleosome complexes, to study the binding mechanisms involved, and to predict how this binding affects chromatin fiber structure. These simulations have revealed the sensitivityof the chromatosome structure to variations in DNA and linker histone sequence, as well as to posttranslational modifications, thereby explaining the structural variability observed in experiments. We propose that a concerted application of experimental and computational approaches will reveal the determinants of chromatosome structural variability and how it impacts chromatin packing.

Modern computational chemistry has reached a stage at which massive exploration into chemical reaction space with unprecedented resolution with respect to the number of potentially relevant molecular structures has become possible. Various algorithmic advances have shown that such structural screenings must and can be automated and routinely carried out. This will replace the standard approach of manually studying a selected and restricted number of molecular structures for a chemical mechanism. The complexity of the task has led to many different approaches. However, all of them address the same general target, namely to produce a complete atomistic picture of the kinetics of a chemical process. It is the purpose of this overview to categorize the problems that should be targeted and to identify the principal components and challenges of automated exploration machines so that the various existing approaches and future developments can be compared based on well-defined conceptual principles.

Dynamic nuclear polarization (DNP) is one of the most prominent methods of sensitivity enhancement in nuclear magnetic resonance (NMR). Even though solid-state DNP under magic-angle spinning (MAS) has left the proof-of-concept phase and has become an important tool for structural investigations of biomolecules as well as materials, it is still far from mainstream applicability because of the potentially overwhelming combination of unique instrumentation, complex sample preparation, and a multitude of different mechanisms and methods available. In this review, I introduce the diverse field and history of DNP, combining aspects of NMR and electron paramagnetic resonance. I then explain the general concepts and detailed mechanisms relevant at high magnetic field, including solution-state methods based on Overhauser DNP but with a greater focus on the more established MAS DNP methods. Finally, I review practical considerations and fields of application and discuss future developments.

Gram-negative bacteria are protected by a multicompartmental molecular architecture known as the cell envelope that contains two membranes and a thin cell wall. As the cell envelope controls influx and efflux of molecular species, in recent years both experimental and computational studies of such architectures have seen a resurgence due to the implications for antibiotic development. In this article we review recent progress in molecular simulations of bacterial membranes. We show that enormous progress has been made in terms of the lipidic and protein compositions of bacterial systems. The simulations have moved away from the traditional setup of one protein surrounded by a large patch of the same lipid type toward a more bio-logically representative viewpoint. Simulations with multiple cell envelope components are also emerging. We review some of the key method developments that have facilitated recent progress, discuss some current limitations, and offer a perspective on future directions.

The gas-phase ground-state dissociation energy D0(S0) of an isolated and cold bimolecular complex is a fundamental measure of the intermolecular interaction strength between its constituents. Accurate D0 values are important for the understanding of intermolecular bonding, for benchmarking high-level theoretical calculations, and for the parameterization of dispersion-corrected density functionals or force-field models that are used in fields ranging from crystallography to biochemistry. We review experimental measurements of the gas-phase D0(S0) and D0(S1) values of 55 different M⋅S complexes, where M is a (hetero)aromatic molecule and S is a closed-shell solvent atom or molecule. The experiments employ the triply resonant SEP-R2PI laser method, which involves M-centered (S0 → S1) electronic excitation, followed by S1 → S0 stimulated emission spanning a range of S0 state vibrational levels. At sufficiently high vibrational energy, vibrational predissociation of the M⋅S complex occurs. A total of 49 dissociation energies were bracketed to within ≤1.0 kJ/mol, providing a large experimental database of accurate noncovalent interactions.

Ever since Clausius in 1865 and Boltzmann in 1877, the concepts of entropy and of its maximization have been the foundations for predicting how material equilibria derive from microscopic properties. But, despite much work, there has been no equally satisfactory general variational principle for nonequilibrium situations. However, in 1980, a new avenue was opened by E.T. Jaynes and by Shore and Johnson. We review here maximum caliber, which is a maximum-entropy-like principle that can infer distributions of flows over pathways, given dynamical constraints. This approach is providing new insights, particularly into few-particle complex systems, such as gene circuits, protein conformational reaction coordinates, network traffic, bird flocking, cell motility, and neuronal firing.

The structure–function relationships of biomolecules have captured the interest and imagination of the scientific community and general public since the field of structural biology emerged to enable the molecular understanding of life processes. Proteins that play numerous functional roles in cellular processes have remained in the forefront of research, inspiring new characterization techniques. In this review, we present key theoretical concepts and recent experimental strategies using femtosecond stimulated Raman spectroscopy (FSRS) to map the structural dynamics of proteins, highlighting the flexible chromophores on ultrafast timescales. In particular, wavelength-tunable FSRS exploits dynamic resonance conditions to track transient-species-dependent vibrational motions, enabling rational design to alter functions. Various ways of capturing excited-state chromophore structural snapshots in the time and/or frequency domains are discussed. Continuous development of experimental methodologies, synergistic correlation with theoretical modeling, and the expansion to other nonequilibrium, photoswitchable, and controllable protein systems will greatly advance the chemical, physical, and biological sciences.

This review examines low-frequency vibrational modes of proteins and their coupling to enzyme catalytic sites. That protein motions are critical to enzyme function is clear, but the kinds of motions present in proteins and how they are involved in function remain unclear. Several models of enzyme-catalyzed reaction suggest that protein dynamics may be involved in the chemical step of the catalyzed reaction, but the evidence in support of such models is indirect. Spectroscopic studies of low-frequency protein vibrations consistently show that there are underdamped modes of the protein with frequencies in the tens of wavenumbers where overdamped behavior would be expected. Recent studies even show that such underdamped vibrations modulate enzyme active sites. These observations suggest that increasingly sophisticated spectroscopic methods will be able to unravel the link between low-frequency protein vibrations and enzyme function.

Nonstatistical dynamics is important for many chemical reactions. The Rice-Ramsperger-Kassel-Marcus (RRKM) theory of unimolecular kinetics assumes a reactant molecule maintains a statistical microcanonical ensemble of vibrational states during its dissociation so that its unimolecular dynamics are time independent. Such dynamics results when the reactant's atomic motion is chaotic or irregular. Intrinsic non-RRKM dynamics occurs when part of the reactant's phase space consists of quasiperiodic/regular motion and a bottleneck exists, so that the unimolecular rate constant is time dependent. Nonrandom excitation of a molecule may result in short-time apparent non-RRKM dynamics. For rotational activation, the 2J + 1 K levels for a particular J may be highly mixed, making K an active degree of freedom, or K may be a good quantum number and an adiabatic degree of freedom. Nonstatistical dynamics is often important for bimolecular reactions and their intermediates and for product-energy partitioning of bimolecular and unimolecular reactions. Post–transition state dynamics is often highly complex and nonstatistical.

Intriguing properties of photoemission from free, unsupported particles and droplets were predicted nearly 50 years ago, though experiments were a technical challenge. The last few decades have seen a surge of research in the field, due to advances in aerosol technology (generation, characterization, and transfer into vacuum), the development of photoelectron imaging spectrometers, and advances in vacuum ultraviolet and ultrafast light sources. Particles and droplets offer several advantages for photoemission studies. For example, photoemission spectra are dependent on the particle's size, shape, and composition, providing a wealth of information that allows for the retrieval of genuine electronic properties of condensed phase. In this review, with a focus on submicrometer-sized, dielectric particles and droplets, we explain the utility of photoemission from such systems, summarize several applications from the literature, and present some thoughts on future research directions.

This review summarizes progress in understanding electron transfer from photoexcited nanocrystals to redox enzymes. The combination of the light-harvesting properties of nanocrystals and the catalytic properties of redox enzymes has emerged as a versatile platform to drive a variety of enzyme-catalyzed reactions with light. Transfer of a photoexcited charge from a nanocrystal to an enzyme is a critical first step for these reactions. This process has been studied in depth in systems that combine Cd-chalcogenide nanocrystals with hydrogenases. The two components can be assembled in close proximity to enable direct interfacial electron transfer or integrated with redox mediators to transport charges. Time-resolved spectroscopy and kinetic modeling have been used to measure the rates and efficiencies of the electron transfer. Electron transfer has been described within the framework of Marcus theory, providing insights into the factors that can be used to control the photochemical activity of these biohybrid systems. The range of potential applications and reactions that can be achieved using nanocrystal–enzyme systems is expanding, and numerous fundamental and practical questions remain to be addressed.

Machine learning (ML) is transforming all areas of science. The complex and time-consuming calculations in molecular simulations are particularly suitable for an ML revolution and have already been profoundly affected by the application of existing ML methods. Here we review recent ML methods for molecular simulation, with particular focus on (deep) neural networks for the prediction of quantum-mechanical energies and forces, on coarse-grained molecular dynamics, on the extraction of free energy surfaces and kinetics, and on generative network approaches to sample molecular equilibrium structures and compute thermodynamics. To explain these methods and illustrate open methodological problems, we review some important principles of molecular physics and describe how they can be incorporated into ML structures. Finally, we identify and describe a list of open challenges for the interface between ML and molecular simulation.

Intrinsically disordered proteins (IDPs) are now widely recognized as playing critical roles in a broad range of cellular functions as well as being implicated in diverse diseases. Their lack of stable secondary structure and tertiary interactions, coupled with their sensitivity to measurement conditions, stymies many traditional structural biology approaches. Single-molecule Förster resonance energy transfer (smFRET) is now widely used to characterize the physicochemical properties of these proteins in isolation and is being increasingly applied to more complex assemblies and experimental environments. This review provides an overview of confocal diffusion-based smFRET as an experimental tool, including descriptions of instrumentation, data analysis, and protein labeling. Recent papers are discussed that illustrate the unique capability of smFRET to provide insight into aggregation-prone IDPs, protein–protein interactions involving IDPs, and IDPs in complex experimental milieus.

At the intersection of spectroscopy and microscopy lie techniques that are capable of providing subnanometer imaging of excited states of individual molecules or nanoparticles. Such approaches are particularly important for imaging macromolecules or nanoparticles large enough to have a high probability of containing a defect. These inevitable defects often control properties and function despite an otherwise ideal structure. We discuss real-space imaging techniques such as using scanning tunneling microscopy tips to enhance optical measurements and electron energy-loss spectroscopy in a scanning transmission electron microscope, which is based on focused electron beams to obtain high-resolution spatial information on excited states. The outlook for these methods is bright, as they will provide critical information for the characterization and improvement of energy-switching, electron-switching, and energy-harvesting materials.

Bose–Einstein condensation describes the macroscopic occupation of a single-particle mode: the condensate. This state can in principle be realized for any particles obeying Bose–Einstein statistics; this includes hybrid light-matter excitations known as polaritons. Some of the unique optoelectronic properties of organic molecules make them especially well suited for the realization of polariton condensates. Exciton-polaritons form in optical cavities when electronic excitations couple collectively to the optical mode supported by the cavity. These polaritons obey bosonic statistics at moderate densities, are stable at room temperature, and have been observed to form a condensed or lasing state. Understanding the optimal conditions for polariton condensation requires careful modeling of the complex photophysics of organic molecules. In this article, we introduce the basic physics of exciton-polaritons and condensation and review experiments demonstrating polariton condensation in molecular materials.

Ions transiting biomembranes might pass readily from water through ion-specific membrane proteins if these protein channels provide environments similar to the aqueous solution hydration environment. Indeed, bulk aqueous solution is an important reference condition for the ion permeation process. Assessment of this hydration mimicry concept depends on understanding the hydration structure and free energies of metal ions in water in order to provide a comparison for the membrane channel environment. To refine these considerations, we review local hydration structures of ions in bulk water and the molecular quasi-chemical theory that provides hydration free energies. In doing so, we note some current views of ion binding to membrane channels and suggest new physical chemical calculations and experiments that might further clarify the hydration mimicry concept.