2021 JCP Emerging Investigator Special Collection

The Theoretical Methods and Algorithms section is well represented in this Collection by a total of 16 papers. These span the entire breadth of theoretical and computational chemical physics, ranging from fundamental aspects of electronic structure theory to the use of machine learning in molecular simulation.

Two of the papers on electronic structure theory apply numerical methods to fundamental systems. The Communication by Burton investigates the properties of the Hartree–Fock (HF) ground state of a two-electron atom in the restricted and unrestricted HF formalisms.⁴ Using numerical techniques, the author reveals the critical nuclear charge for spin-symmetry breaking in the HF wave function and, thus, provides the first estimate of a spin-symmetry breaking threshold in the complete-basis-set HF limit. The path integral Monte Carlo (PIMC) method provides direct access to the imaginary time correlation functions that arise in the context of response properties. The paper by Dornheim et al. presents hitherto unexplored expressions for quadratic and cubic response functions and illustrates their utility with the facile extraction of numerical data from PIMC simulations of the uniform electron gas.⁵ Their work introduces a powerful new approach to the calculation of nonlinear electronic structure observables in condensed phase systems.

Several further papers develop methods that extend the applicability of modern electronic structure techniques to larger molecules. Abraham and Mayhall develop a many-body expansion of the electron correlation energy from a tensor product state reference wave function in which the active space is partitioned into small subsets (or clusters) of orbitals.⁶ This is shown to improve upon previous approaches based on a restricted HF reference, as the authors illustrate with calculations on a variety of challenging systems. Banerjee and Sokolov present an efficient implementation of second- and third-order algebraic diagrammatic construction theory for electron attachment and ionization energies and spectra.⁷ This allows for the computation of photoelectron spectra, vertical and adiabatic electron affinities, and ionization potentials in larger systems than previously accessible.

Blaschke and Stopkowicz consider the calculation of molecular properties in strong magnetic fields, where the reduced permutational symmetry and the need to use complex basis functions significantly increases the effort required to evaluate electron repulsion integrals.⁸ The authors apply a Cholesky decomposition technique to compress the integrals, leading to efficient algorithms that are applicable to larger molecules. Xu and Yang use constrained nuclear–electronic orbital density functional theory to estimate the Hessian of a molecular or condensed-phase system in a way that takes into account both nuclear and electronic corrections to harmonic vibrational frequencies.⁹ Their method serves as an alternative to second-order vibrational perturbation theory that is competitive in terms of accuracy and computational efficiency, as they demonstrate with a comprehensive set of benchmark calculations.

The remaining electronic structure paper goes beyond the usual ab initio treatment of liquids at the level of density functional theory. Weng and Vlček adapt and refine a powerful stochastic approach to the calculation of the electronic structure of solvated molecules at the GW level of theory, for supercells containing 1000 or more electrons.¹⁰ They are able to use this approach to accurately determine the contribution of the solvent to electronic spectral shifts, as they illustrate by comparison with high-level calculations and experimental data for a variety of organic molecules in aqueous solution.

The next set of papers we shall discuss are on various aspects of electronically non-adiabatic dynamics. Ab initio multiple spawning (AIMS) is a method for photo-dynamics problems in which one can, in principle, converge on the exact quantum mechanical result by increasing the number of spawned Gaussian basis functions. However, in its standard implementation, the number of basis functions required for an accurate solution can often become impractical. In their Communication, Lassmann and Curchod describe an improvement to the stochastic selection variant of AIMS that aims to overcome this bottleneck and illustrate its utility with example applications to the photo-physics of several organic molecules.¹¹ Surface hopping is a cheaper and more approximate alternative to AIMS that is widely used in non-adiabatic dynamics simulations. Krotz and Tempelaar extend the applicability of this method by introducing a novel surface hopping algorithm in reciprocal space that is compatible with electronic band structure calculations.¹² This new method has potential applications in simulating the phenomena that arise from strong electron–phonon coupling in band-like materials.

Another Communication on non-adiabatic dynamics by Tiwari and co-workers uses a model excitonic trimer to analyze vibronic resonance effects in excitation energy transfer.¹³ Their analysis involves an effective-mode construction that sheds light on the difference between active and spectator vibrational modes in bridge-mediated energy transfer. Sun and co-workers propose new mapping strategies to construct multi-state harmonic models with globally shared bath modes from energy gap time correlation functions obtained from molecular dynamics simulations.¹⁴ These strategies could prove to be useful in the study of various condensed phase non-adiabatic dynamics problems.

The remaining paper on electronically non-adiabatic dynamics by Chowdhury and Huo develops an approximation to non-adiabatic quantum dynamics in which the electronic degrees of freedom are treated using mapping variables and the nuclear motion is treated using Matsubara dynamics.¹⁵ Their paper is largely formal but it may eventually provide a framework for the development of more practical methods for calculating non-adiabatic real-time correlation functions with the inclusion of nuclear quantum effects.

Turning now to papers on other topics, a second paper by Huo and co-workers proposes a theoretical model for the recent experimental observation that chemical reactions can be modified by coupling to vibrational polaritons.¹⁶ The authors use their model to demonstrate that the reaction rate constant can be suppressed by coupling molecular vibrations with an optical cavity, as a result of both a collective coupling effect and cavity frequency modification of the rate constant. Hsu and co-workers investigate the effect of the electromagnetic environment factor (EEF) on the power spectrum of a molecular emitter in a plasmonic structure.¹⁷ They show that propagation of the emitted photon in the complex dielectric environment has an important impact on the observed spectrum, which cannot, therefore, simply be interpreted as arising from the quantum dynamics of the molecular emitter.

Finally, there are two papers that use machine learning techniques to solve complex problems in chemical physics. The first by Wang and Tiwary introduces a state predictive information bottleneck approach to learning reaction coordinates from high dimensional molecular simulation trajectories, showing how it can be linked to traditional statistical physics concepts such as the committor.¹⁸ The second by Chennakesavalu and Rotskoff considers how the assembly of matter into well-defined target states can be controlled.¹⁹ Their paper synthesizes developments in machine learning, optimal transport theory, and non-equilibrium statistical mechanics in an approach to directed self-assembly that focuses on how external control guides the system to a target distribution.

This section of the journal is represented by a single paper in the Collection by Foroozandeh and co-workers, who demonstrate the implementation of CHORUS (CHirped, ORdered pulses for Ultra-broadband Spectroscopy), an ultra-broadband excitation scheme for ESR spectroscopy.²⁰ This technique allows them to bypass the traditional shortcomings of multi-pulse ESR, including limited microwave power, and therefore, the limited excitation bandwidth, to achieve excitation pulses with constant amplitude and phase that produce signals over a wide spectral width, and that are robust to instrumental imperfections.

Two of the papers in this section of the Collection are in the traditional area of gas phase spectroscopy and dynamics, and the remaining two are in the emerging area of polariton chemistry.

Gao and co-workers report the first high-resolution spectroscopic study of the ¹⁴N¹⁵N isotopomer of N₂ in the VUV energy range between 109 000 and 117 500 cm⁻¹.²¹ The resulting spectroscopic parameters are useful for understanding the dissociation dynamics of N₂ and may find application in modeling the properties of planetary atmospheres. Hansen and co-workers perform extensive experimental studies of the C–C bond fission in gas phase CF₃CHO molecules following excitation to the S₁ state and use their results to shed light on the competition between the ground (S₀) state and triplet (T₁) state reaction pathways.²² 

The first paper on polariton chemistry by Phuc studies how the collective behavior of molecular polaritons can lead to enhanced reactions under strong light–matter interaction conditions.²³ The second by Xiong and Wiesehan notes that, while perturbation of reaction rates and pathways by coupling vibrational modes of reactants and intermediates to an optical cavity to form polaritons has the potential to create a new means of control in chemical synthesis, these effects must be reproducible to be useful.²⁴ Using a demonstration of cavity-coupled catalytic ester hydrolysis of para-nitrophenyl acetate with vibrational strong coupling, these authors present a cautionary tale that highlights the sensitivity of strong coupling-perturbed reaction rates to experimental conditions.

This section of the Collection contains three articles on aspects of molecular simulation and a Communication on the experimental control of crystalline self-assembly and the photo-physical properties of the resulting crystal.

He and co-workers present an automated fragmentation QM/MM approach for the efficient parallel calculation of chemical shifts in molecular crystals, along with a software implementation.²⁵ The authors also systematically investigate the accuracy of the computed shifts as a function of different density functionals and basis sets.

Budroni and colleagues revisit the mechanism underlying chemo-hydrodynamic oscillations, that is, oscillations in composition resulting from competing chemical processes and chemistry-driven hydrodynamic flows.²⁶ From their numerical simulations, the authors were able to identify the mechanism underlying the oscillations. This work opens new perspectives for controlling the formation of chemical patterns on the basis of simple chemistry.

In the paper of Moultos and co-workers, the vapor pressures of deep eutectic solvents are determined by using computer simulations with state-of-the-art force fields.²⁷ The vapor pressure is very low, which makes its determination in experiments quite difficult. However, it can be estimated once the free energy is computed from molecular simulations. It is found that the vapor phase contains mostly the component of the mixture that acts as the hydrogen bond donor (HBD).

In their Communication, Campillo-Alvarado and co-workers control the crystalline assembly and optical and electronic properties of an organic semiconductor through small chemical changes to its molecular structure.²⁸ 

This section of the Collection contains six papers, two of which describe experimental studies and the remaining four theoretical modeling. Both of the Emerging Investigator Prize winning articles were published in this section of JCP.

In their prize winning transient absorption/reflection spectroscopy study of exciton polaritons, Musser and co-workers uncover interesting spectral and temporal features caused by complex optical properties of polaritonic systems and suggest the need for high-time-resolution measurements on high-quality microcavities to uncover their intrinsic polariton dynamics.²⁹ 

In a second experimental paper, Nienhaus and co-workers take a further step toward scalable photon upconversion materials with a demonstration of chemical control of sensitized triplet–triplet annihilation in lead halide perovskite-rubrene systems with varied bromide content.³⁰ They find a trade-off between the yield of interfacial charge transfer and near-IR absorption cross-section that puts the optimized bromide content of the perovskite at 5% for this application.

Turning now to the theoretical papers, the possibility of the absence of electroneutrality in confined systems has been recently proposed to explain the peculiar behavior of the space-charge density within charged nanopores. Yoav Green carefully examines the conditions for such a phenomenon in his prize winning manuscript.³¹ He suggests, supported by theoretical arguments and numerical investigations, that the breakdown within the pore results from the highly sensitive response of the electric field to the boundary condition at the dielectric edge.

Sambur and co-workers make information about the spatial relationships between emissive defects and molecular adsorbates on semiconductor nanocrystals more accessible by demonstrating that stochastic modeling of the rates of energy transfer from defect donors to molecular acceptors can be translated into average distances between the two; their co-localization indicates that the relevant defects are primarily on the surfaces of the particles.³² 

The contribution by Van Lehn and co-workers addresses a very challenging problem in chemical physics, namely the non-additive behavior that arises when molecules adsorbed at a surface self-assemble in non-random patterns.³³ This paper uses a combination of data-driven and physics-based modeling techniques, producing both new insights into this topic and a dataset of self-assembled structures that will be useful to support future studies.

Heterogeneous interfaces formed between two-photon absorbers have attracted great attention due to their importance in optoelectronic applications. The paper by Adeniran and Liu employs two approaches recently developed in their group, namely substrate screening GW and dielectric embedding GW, to systematically characterize a series of phthalocyanine:TMD interfaces and thereby develop useful quantitative insight into the structure–property relationships of the interfaces.³⁴ 

Both of the papers in this section of the Collection are purely theoretical and describe state-of-the-art modeling of phenomena in polymer and soft matter systems.

The paper by Sing and co-workers focuses on bottlebrush polymers, highly branched macromolecules.³⁵ Because of their large molecular weight, coarse-grained strategies are needed to explore their collective behavior and the possibility of their application in self-assembled photonic materials and tunable elastomers. By developing a potential of mean force between brush chains, the authors show that the branches can be efficiently coarse-grained, resulting in a much simpler and faster potential that can be effectively used to model this important class of materials.

The motility-induced phase separation (MIPS) constitutes the active matter counterpart of the thermodynamic phase separation. In MIPS, the self-correlating driven motion induces the formation of regions of different particle concentration. In their Communication, Ma and Ni explore how MIPS evolves when an active torque acts on the particles (transforming them into dynamically chiral constituents).³⁶ They show, using a combination of numerical and theoretical methods, that the circulating current opposes the phase separation process, leaving the system in a state of dynamic clustering. Interestingly, the authors note that the active torque can be self-generated by an asymmetric shape, by an inhomogeneous mass distribution, or by catalysis coating, making their observation particularly relevant to interpreting a variety of active matter systems.

This final section of the Collection is represented by a single paper by Wei and co-workers.³⁷ These authors report how they have achieved exquisite control over the photophysical behavior and electron-vibration coupling of molecular dyes to enable their function as probes in electronic pre-resonance stimulated Raman scattering, a powerful tool for live-cell optical imaging.

The 2022 Emerging Investigator Collection is currently publishing and the 2023 collection is open for submissions. The deadline for acceptance into the 2023 collection is December 31, 2023. Papers accepted after that date will be included in the 2024 collection and considered for the 2024 awards. For more information, please see the journal website: https://aip.scitation.org/jcp/info/awards. To go directly to how to submit and eligibility requirements, please see the call for papers page: https://publishing.aip.org/publications/journals/special-topics/jcp/2023-jcp-emerging-investigators-special-collection/.

We look forward to many more annual collections recognizing early career researchers. If you have any questions or comments, please contact us at jcp-journalmanager@aip.org.