Person:

Olivares-Amaya, Roberto

Excess charge stabilization of molecules in metallic environments is of particular importance for fields such as molecular electronics and surface chemistry. We study the energetics of benzene and its anion between two metallic plates. We observe that orientational effects are important at small inter-plate separation. This leads to benzene oriented perpendicular to the gates being more stable than the parallel case due to induced dipole effects. We find that the benzene anion, known for being unstable in the gas-phase, is stabilized by the plates at zero bias and an inter-plate distance of 21 Å. We also observe the effect of benzene under a voltage bias generated by the plates; under a negative bias, the anion becomes destabilized. We use the electron localization function to analyze the changes in electron density due to the bias. These findings suggest that image effects such as those present in nanoscale devices, are able to stabilize excess charge and should be important to consider when modeling molecular transport junctions and charge-transfer effects.

Two new tools for the acceleration of computational chemistry codes using graphical processing units (GPUs) are presented. First, we propose a general black-box approach for the efficient GPU acceleration of matrix−matrix multiplications where the matrix size is too large for the whole computation to be held in the GPU’s onboard memory. Second, we show how to improve the accuracy of matrix multiplications when using only single-precision GPU devices by proposing a heterogeneous computing model, whereby single- and double-precision operations are evaluated in a mixed fashion on the GPU and central processing unit, respectively. The utility of the library is illustrated for quantum chemistry with application to the acceleration of resolution-of-the-identity second-order Møller−Plesset perturbation theory calculations for molecules, which we were previously unable to treat. In particular, for the 168-atom valinomycin molecule in a cc-pVDZ basis set, we observed speedups of 13.8, 7.8, and 10.1 times for single-, double- and mixed-precision general matrix multiply (SGEMM, DGEMM, and MGEMM), respectively. The corresponding errors in the correlation energy were reduced from −10.0 to −1.2 kcal mol(^{-1}) for SGEMM and MGEMM, respectively, while higher accuracy can be easily achieved with a different choice of cutoff parameter.

Time-dependent density functional theory (TDDFT) has recently been extended to describe many-body open quantum systems evolving under nonunitary dynamics according to a quantum master equation. In the master equation approach, electronic excitation spectra are broadened and shifted due to relaxation and dephasing of the electronic degrees of freedom by the surrounding environment. In this paper, we develop a formulation of TDDFT linear-response theory (LR-TDDFT) for many-body electronic systems evolving under a master equation, yielding broadened excitation spectra. This is done by mapping an interacting open quantum system onto a noninteracting open Kohn–Sham system yielding the correct nonequilibrium density evolution. A pseudoeigenvalue equation analogous to the Casida equations of the usual LR-TDDFT is derived for the Redfield master equation, yielding complex energies and Lamb shifts. As a simple demonstration, we calculate the spectrum of a (C^{2 +}) atom including natural linewidths, by treating the electromagnetic field vacuum as a photon bath. The performance of an adiabatic exchange-correlation kernel is analyzed and a first-order frequency-dependent correction to the bare Kohn–Sham linewidth based on the Görling–Levy perturbation theory is calculated.

The modification of a general purpose code for quantum mechanical calculations of molecular properties (Q-Chem) to use a graphical processing unit (GPU) is reported. A 4.3x speedup of the resolution-of-the-identity second-order Møller−Plesset perturbation theory (RI-MP2) execution time is observed in single point energy calculations of linear alkanes. The code modification is accomplished using the compute unified basic linear algebra subprograms (CUBLAS) library for an NVIDIA Quadro FX 5600 graphics card. Furthermore, speedups of other matrix algebra based electronic structure calculations are anticipated as a result of using a similar approach.

Graphical processing units are now being used with dramatic effect to accelerate quantum chemistry calculations. However, early work exposed challenges involving memory bottlenecks and insufficient numerical precision. This research effort addresses those issues, proposing two new tools for accelerating matrix multiplications of arbitrary size where single-precision accuracy is not enough.

In this perspective we explore the use of strategies from drug discovery, pattern recognition, and machine learning in the context of computational materials science. We focus our discussion on the development of donor materials for organic photovoltaics by means of a cheminformatics approach. These methods enable the development of models based on molecular descriptors that can be correlated to the important characteristics of the materials. Particularly, we formulate empirical models, parametrized using a training set of donor polymers with available experimental data, for the important current–voltage and efficiency characteristics of candidate molecules. The descriptors are readily computed which allows us to rapidly assess key quantities related to the performance of organic photovoltaics for many candidate molecules. As part of the Harvard Clean Energy Project, we use this approach to quickly obtain an initial ranking of its molecular library with 2.6 million candidate compounds. Our method reveals molecular motifs of particular interest, such as the benzothiadiazole and thienopyrrole moieties, which are present in the most promising set of molecules.

Modeling of electronic structure of molecules in electrostatic environments is of considerable relevance for surface-enhanced spectroscopy and molecular electronics. We have developed and implemented a novel approach to the molecular electronic structure in arbitrary electrostatic environments that is compatible with standard quantum chemical methods and can be applied to medium-sized and large molecules. The scheme denoted CheESE (chemistry in electrostatic environments) is based on the description of molecular electronic structure subject to a boundary condition on the system/environment interface. Thus, it is particularly suited to study molecules on metallic surfaces. The proposed model is capable of describing both electrostatic effects near nanostructured metallic surfaces and image-charge effects. We present an implementation of the CheESE model as a library module and show example applications to neutral and negatively charged molecules.

We study the effects of chemical bonding on Raman scattering from benzenethiol chemisorbed on silver clusters using time-dependent density functional theory (TDDFT). Raman scattering cross sections are computed using a formalism that employs analytical derivatives of frequency-dependent electronic polarizabilities, which treats both off-resonant and resonant enhancement within the same scheme. In the off-resonant regime, Raman scattering into molecular vibrational modes is enhanced by one order of magnitude and shows pronounced dependence on the orientation and the local symmetry of the molecule. Additional strong enhancement of the order of (10^2) arises from resonant transitions to mixed metal–molecular electronic states. The Raman enhancement is analyzed using Raman excitation profiles (REPs) for the range of excitation energies 1.6–3.0 eV, in which isolated benzenethiol does not have electronic transitions. The computed vibrational frequency shifts and relative Raman scattering cross sections of the metal–molecular complexes are in good agreement with experimental data on surface enhanced Raman scattering (SERS) from benzenethiol adsorbed on silver surfaces. Characterization and understanding of these effects, associated with chemical enhancement mechanism, may be used to improve the detection sensitivity in molecular Raman scattering.

This Perspective introduces the Harvard Clean Energy Project (CEP), a theory-driven search for the next generation of organic solar cell materials. We give a broad overview of its setup and infrastructure, present first results, and outline upcoming developments. CEP has established an automated, high-throughput, in silico framework to study potential candidate structures for organic photovoltaics. The current project phase is concerned with the characterization of millions of molecular motifs using first-principles quantum chemistry. The scale of this study requires a correspondingly large computational resource, which is provided by distributed volunteer computing on IBM’s World Community Grid. The results are compiled and analyzed in a reference database and will be made available for public use. In addition to finding specific candidates with certain properties, it is the goal of CEP to illuminate and understand the structure–property relations in the domain of organic electronics. Such insights can open the door to a rational and systematic design of future high-performance materials. The computational work in CEP is tightly embedded in a collaboration with experimentalists, who provide valuable input and feedback to the project.

We explore the chemical contribution to surface-enhanced Raman scattering (SERS) in mixed-metal substrates, both experimentally and by computer simulation. These substrates are composed of a chemically active, transition-metal overlayer deposited on an effective SERS substrate. We report improved analytical enhancement factors obtained by using a small surface coverage of palladium or platinum over nanostructured silver substrates. Theoretical predictions of the chemical contribution to the surface enhancement using density functional theory support the experimental results. In addition, these approaches show that the increased enhancement is due not only to an increase in surface coverage of the analyte but also to a higher Raman scattering cross section per molecule. The additional chemical enhancement in mixed-metal SERS substrates correlates with the binding energy of the analyte on the surface and includes both static and dynamical effects. SERS using mixed-metal substrates has the potential to improve sensing for a large group of analyte molecules and to aid the development of chemically specific SERS-based sensors.