Edoardo Aprà

This work presents recent developments in the COSMO solvation model implementation in NWChem. A new cavity construction approach, based on the solvent-excluding surface (SES) and utilizing the well-established GEPOL algorithm, has been introduced. Additionally, a straightforward procedure to merge surface segments that are too close─often a source of numerical artifacts─has been implemented. The available methods for correcting outlying surface charges have also been reviewed and improved. To validate the new implementation, we computed dielectric solvation energies for a chemically diverse set of approximately 100 molecules, including neutral species, small ions, and common ionic liquid components. Results were compared to those from GAMESS using the double-cavity method as a reference. Although the double-cavity approach can be regarded as more accurate, the simpler correction schemes available in NWChem─based on scaling factors or Lagrange multipliers─can achieve excellent agreement if the potential is also properly corrected, with mean unsigned deviations of around 0.15 kcal/mol. Predictions of typical vapor-liquid and liquid-liquid equilibria using a COSMO-SAC variant based on NWChem also showed very promising results.

The transformative impact of modern computational paradigms and technologies, such as high-performance computing (HPC), quantum computing, and cloud computing, has opened up profound new opportunities for scientific simulations. Scalable computational chemistry is one beneficiary of this technological progress. The main focus of this paper is on the performance of various quantum chemical formulations, ranging from low-order methods to high-accuracy approaches, implemented in different computational chemistry packages and libraries, such as NWChem, NWChemEx, Scalable Predictive Methods for Excitations and Correlated Phenomena, ExaChem, and Fermi–Löwdin orbital self-interaction correction on Azure Quantum Elements, Microsoft's cloud services platform for scientific discovery. We pay particular attention to the intricate workflows for performing complex chemistry simulations, associated data curation, and mechanisms for accuracy assessment, which is demonstrated with the Arrows automated workflow for high throughput simulations. Finally, we provide a perspective on the role of cloud computing in supporting the mission of leadership computational facilities.

We report computations of the vertical ionization potentials within the GW approximation of the near-complete series of first-row transition metal (V-Cu) aqua ions in their most common oxidation states, i.e., V³⁺, Cr³⁺, Cr²⁺, Mn²⁺, Fe³⁺, Fe²⁺, Co²⁺, Ni²⁺, and Cu²⁺. The d-orbital occupancy of these systems spans a broad range from d² to d⁹. All of the structures were first optimized at the density functional theory level using a large cluster of explicit water molecules that are embedded in a continuum solvation model. Vertical ionization potentials were computed with the one-shot G₀W₀ approach on a range of transition metal ion clusters (6, 18, 40, and 60 explicit water molecules), wherein the convergence with respect to the basis set size was evaluated using the systems with 40 water molecules. We assess the results using three different density functional approximations as starting points for the vertical ionization potential calculations, namely, G₀W₀@PBE, G₀W₀@PBE0, and G₀W₀@r²SCAN. While the predicted ground-state structures are similar to all three exchange-correlation functionals, the vertical ionization potentials were in closer agreement with experiment when using the G₀W₀@PBE0 and G₀W₀@r²SCAN approaches, with the r²SCAN-based calculations being significantly less expensive. Computed bond distances and vertical ionization potentials for all structures are in good agreement with available experimental data.

Navier’s objectives were two evaluate the use of emerging technologies, especially dataflow accelerators, for high-performance computing (HPC) applications, specifically in the domain of chemistry, and to develop a prototype software stack to support such applications. Navier builds on capabilities previously developed by synergistic projects, such as PNNL Data Model Convergence (DMC) LDRD Hardware Advanced Workflows (HAW) and DuOMO, as well as DOE ARIAA. Throughout its 18 months, the Navier team developed new capabilities and artifacts at all levels of the HW/SW stack, provided a seamless way to integrate novel computing architectures (Sambanova SN10 and Xilinx Versal AI) into an existing software stack, developed chemistry workflows, data analytics tools, and HPC molecular dynamics workflows that leverage the developed stack and PNNL institutional investments in emerging architectures. Navier also explored the use of active learning to accelerate a computational chemistry workflow for organic molecules on PNNL Junction cluster (in collaboration with AMD/Xilinx). Navier developed tools, methodologies, and studies for hardware software co-design and (sparse) dataflow accelerators that are composable and can be used together or separately. These methodologies are now used in other projects, such as DOE AMAIS and HPDA. This report describes Navier’s achievement, the developed tools and methodologies, and themore » research findings and conclusions. « less

This report presents the work performed under the GRaman project, sponsored by the PCSD LDRD Seed program. The project aimed at accelerating ab initio molecular dynamics simulation using Graph Networks. The Graph Network framework is a ML framework that has been successfully employed to simulate the dynamics of several physical systems: including water splashing in a container and flags moving with the wind. In this effort, we performed a data collection campaign for 3 different molecules of interest. We have built tools for preprocessing the trajectories obtained by simulating Raman Spectroscopy with NWChem and translating them into a suitable format for training. We have developed a training algorithm to train the Graph Network based simulators based on our data and developed a simulator that produces trajectories in the same NWChem format. While the tool has improved with each iteration of development and subsequent experiments, the current state of the tool does not allow to directly incorporate the technology within the NWChem framework because the trajectories produced by the tool are not yet accurate enough. However, the technology has proved to have good potential and it is certainly worth further research and development.

The Molecular Dynamics technique Collective variable hyperdynamics (CVHD) interfaced with density functional tight-binding at the GFN1-xTB and GFN2-xTB levels of electronic structure theory has been applied to the H+GAG positively charged peptide system to explore the fragmentation of this simple tripeptide under relatively low temperature/energy conditions. Unlike conventional chemical dynamics simulations which can only be performed up to hundreds of picoseconds when coupled with semi-empirical Hamiltonians, CVHD is able to capture the long-time dynamics of multiple proton hopping, the formation of reversible ring structures, and ester rearrangement of the H+GAG system prior to fragmentation. The CVHD method applied to H+GAG is also able to uncover alternative fragmentation pathways not considered previously such as glycine fragmentation on the N-terminal side of the charged peptide and new cyclic cationic species. These findings in the H+GAG case have implications for other peptide systems.

The complex and heterogeneous nature of the lignin macromolecule has presented a lasting barrier to its utilization. To achieve high lignin yield, the technical lignin extraction process usually severely modifies and condenses the native structure of lignin, which is a critical drawback for its utilization in conversion processes. In addition, there is no method capable of separating lignin from plant biomass with controlled structural properties. Here, we developed an N-heterocycle-based deep eutectic solvent formed between lactic acid and pyrazole (La–Py DES) with a binary hydrogen bonding functionality resulting in a high affinity toward lignin. Up to 93.7% of lignin was extracted from wheat straw biomass at varying conditions from 90 °C to 145 °C. Through careful selection of treatment conditions as well as lactic acid to pyrazole ratios, lignin with controlled levels of ether linkage content, hydroxyl group content, and average molecular weight can be generated. Under mild extraction conditions (90 °C to 120 °C), light-colored native-like lignin can be produced with up to 80% yield, whereas ether linkage–free lignin with low polydispersity can be obtained at 145 °C. Overall, this study offers a new strategy for native lignin extraction and generating lignin with controlled structural properties.

We have established CCSD(T)/CBS (Complete Basis Set) limits for 3 stationary points on the benzene dimer potential energy surface, corresponding to the π⋯π (parallel displaced or PD(C_2h), minimum) and CH⋯π (T-shaped or T(C_2v), transition state) and tilted T-shaped (or TT(C_s), minimum) bonding scenarios considering both the structure and binding energy. The CCSD(T)/CBS binding energies are −2.65 ± 0.02 (PD), −2.74 ± 0.03 (T), and −2.83 ± 0.01 kcal mol⁻¹ (TT). To this end, the CH⋯π is ∼0.2 kcal mol⁻¹ stronger than the π⋯π interaction, whereas the tilting of the CH donating benzene molecule with respect to the other benzene is worth 0.1 kcal mol⁻¹. As previously discussed in the literature, the MP2 level of theory does not provide a close match for either the energy or structure, yet the SCS-MP2 yields structures in excellent agreement with respect to the CCSD(T) result. It is found that the SCS-MI-MP2 also gives optimized structures very close to SCS-MP2 (within ∼0.01 Å of the benchmark). Despite the closer match in structure, the spin-biased MP2 methods (SCS-, SCS-MI-, and SOS-MP2) incorrectly predict the relative stabilities of the isomers. That said, none of the spin biased MP2 methods offers a good compromise between energy and structure for the systems examined. Finally, the CCSD(T)/CBS benchmarks were used to assess the performance of 13 DFT functionals selected from different rungs of Jacob's ladder. Several functionals such as TPSS-D3, B3LYP-D3, B97-D, B97-D3, and B2PLYP-D3 provided a good description of the binding energies for both CH⋯π and π⋯π interactions, yielding values within 6% of the CCSD(T)/CBS benchmark values. Unlike the MP2 methods, these functionals correctly predict the relative stability of the PD(C_2h) and T(C_2v) dimers. Further, we find that there is no systematic improvement as Jacob's ladder is ascended (increased complexity of functional). The best functionals that result in a good compromise between structure and energy accuracy are B97-D3 and B2PLYP-D3 for both the CH⋯π and π⋯π interaction. Despite the impressive performance of these functionals, a challenge that remains is ensuring the transferability of these density functionals in accurately describing the interaction between dimers of larger aromatic molecules, the latter requiring high-level benchmarks for these systems.