ACS Publications最新文献

We describe the implementation details of highly efficient ab initio electrostatic potential (ESP) calculations on graphics processing units (GPUs) and introduce a novel scheme for partial charges that are robust against molecular orientation. Performance analyses are discussed, and we highlight that in our new implementation, a single data center GPU can outperform 128 corresponding data center CPU cores in time to solution. This implementation in the open-source Quantum Interaction Computational Kernel code (QUICK) enables ESP computations on highly dense grids that surpass what is reported in the literature, on the order of N_grid points ∼ 20000 points/atom. We demonstrate that, in this dense-grid limit, ESP charges become independent of molecular orientation. We denote such ESP charges as being robust against molecular orientation and validate this desirable attribute against standard charge schemes. Our proposed charge scheme, called reweighted RESP (rwRESP), is designed to significantly overcome the sensitivity to N_grid points that limits the reliability of canonical RESP charges. By effectively amending this N_grid points-sensitivity, we demonstrate that rwRESP charges also achieve robustness against molecular orientation. Ultradense-grid ESP computations and rwRESP fits can be readily performed via the seamless integration of QUICK with AmberTools, enabling highly efficient and reliable parametrization of the general AMBER force field (GAFF) for nonstandard residues. In this spirit, we believe that our fully fledged GPU protocol for obtaining robust molecular charges will facilitate a wide range of applications, such as high-throughput parametrization of molecular interaction potentials, while also serving as a foundational step toward GPU-accelerated on-the-fly polarizable QM/MM simulations with QUICK.

We report how gases impact the hydrogen concentration in the palladium metal lattice during electrochemical hydrogen loading. We built a unique in situ X-ray diffraction cell, where one surface of a palladium membrane is electrochemically loaded with hydrogen and the other surface faces a gas flow. Under N₂ and CO₂ gas, rapid phase transformation from α-Pd to β-PdH occurred with moderate H/Pd ratios of 0.63 ± 0.02 and 0.64 ± 0.01, respectively. Under CO gas, the α → β phase transformation was also fast, but the H/Pd ratio increased to 0.752 ± 0.001. In contrast, the O₂ gas induced a more gradual α → β phase transformation, achieving the maximum H/Pd ratio of 0.66 ± 0.03, followed by the reverse β → α phase transformation. Gas chromatography confirmed that the increased H/Pd ratio under CO originates from the suppressed recombination of hydrogen atoms into H₂ gas. Additionally, we found that O₂ reacts with hydrogen on the Pd surface to form water and hydrogen peroxide, which together promote hydrogen removal. These findings demonstrate that electrochemical hydrogen loading of Pd is governed not only by the applied electrochemical potential but also by gas-surface interactions.

Organic electrochemical transistors (OECTs) and electrolyte-gated organic field-effect transistors (EGOFETs) represent promising technologies for neuromorphic computing. Yet, conventional dual-mode devices suffer from fundamental performance trade-offs, where optimization for one mode compromises the other. The primary challenge stems from incompatible interfacial requirements: EGOFETs require stable polarization layers preventing ion penetration, while OECTs demand efficient ionic transport for volumetric doping. Here, we present a novel materials engineering strategy employing zwitterionic-modified poly(methyl methacrylate) (PMMA-ZI) as an interlayer to address this fundamental incompatibility. The amphiphilic zwitterionic moieties simultaneously enhance dipolar polarization for EGOFET operation and facilitate balanced ion transport for OECT functionality through controlled electrostatic interactions. PMMA-ZI devices demonstrate remarkable performance enhancements, with a 13.57-fold improvement in the volumetric capacitance-mobility product to 57.25 F cm^-1 V^-1 s^-1 in OECT mode, compared to conventional PMMA interlayers. The devices exhibit exceptional synaptic plasticity with 6.12 times improved memory retention and successful implementation of 4-bit reservoir computing for pattern recognition. This work establishes a new paradigm for dual-mode organic transistors, enabling uncompromised multifunctional operation essential for next-generation neuromorphic computing and bioelectronics applications.

Shape memory polymers (SMPs) with remote activation capabilities have garnered significant attention in biomedical applications. However, traditional activation methods exhibit limitations in tissue penetration depth and spatial-temporal control precision. Here, we report a multifunctional SMP system that synergistically integrates Fe₃O₄ nanoparticles and neodymium-iron-boron (NdFeB) microparticles to achieve simultaneous magnetic navigation and dual-mode thermal activation. The NdFeB microparticles, magnetized under external magnetic fields, provide superior magnetic moments for precise navigation control, while Fe₃O₄ nanoparticles enable both near-infrared II (NIR-II) photothermal conversion and magnetothermal heating under alternating magnetic fields (AMF). Additionally, the resulting magnetic polyurethane (MPU) exhibits excellent shape memory performance with rapid activation kinetics and high recovery ratios while maintaining superior mechanical properties and biocompatibility. Cell viability studies demonstrate minimal cytotoxicity, and animal experiments confirm successful magnetic navigation, precise shape recovery, and the absence of inflammatory responses in physiological environments. The findings demonstrate that this integrated MPU platform has significant potential for applications in minimally invasive surgical instruments, smart tissue scaffolds, and targeted therapeutic delivery systems.

Conventional amine-based electrolytes exhibit superior compatibility with Mg metal anodes, but their practical application is fundamentally constrained by both a restricted electrochemical stability window (<2.0 V on Al foils) and non-negligible high cation desolvation energy barriers. Herein, we first focus on enhancing cation-solvent interaction through a rational high-concentration Li/Mg dual-salt strategy, effectively suppressing free amine solvent molecules and thereby expanding the electrochemical window to exceed 3.0 V. This widened electrochemical window ensures direct compatibility with the industrial Li-ion cathode. Furthermore, to address the high cation desolvation energy barriers and low conductivity in this electrolyte, the ether solvent with a lower coordination ability is introduced into the electrolyte, where part of the ether can participate in the Li⁺ solvation structure to alleviate overly strong amine coordination, while the rest can serve as a pseudo-diluent, promoting a reduced cathode desolvation energy barrier and enhanced ion transport. Finally, the Mg//LiFePO₄ battery delivers a stable plateau of 2.7 V and a high energy density at the electrode level. This work proposes an efficient electrolyte design paradigm that simultaneously balances Mg anode reversibility, high-voltage cathode compatibility, and a facile preparation method in Mg batteries for the first time, revealing a comprehensive exploration process for high-voltage Mg batteries.

The UV-B (broadband 280-320 nm, peak λ = 313 nm) and UV-C (λ = 254 nm) initiated photolytic oxidation of 1,1-difluoroacetone (CF₂HC(O)CH₃) and 1,1,1-trifluoroacetone (CF₃C(O)CH₃) was studied as a function of total pressure using smog chamber techniques. The UV spectrum of CF₂HC(O)CH₃ and CF₃C(O)CH₃ are reported; the former for the first time. The UV-B and UV-C photolysis rates were measured relative to that of CH₃C(O)CH₃. The approximate UV-B (313 nm) quantum yield for CF₂HC(O)CH₃ and CF₃C(O)CH₃ were determined as 0.03 and 0.007. At 254 nm, the overall quantum yields for CF₂HC(O)CH₃ and CF₃C(O)CH₃ were determined as (1.11 ± 0.13) and (0.69 ± 0.08), respectively, at 700 Torr, (298 ± 1) K, independent of diluent gas. This is the first chamber study of the photolysis of CF₂HC(O)CH₃ and CF₃C(O)CH₃. The measured yields of HCOF, COF₂ (and CO) suggest that photolysis of CF₂HC(O)CH₃ and CF₃C(O)CH₃ produces CF₂H and CF₃ radicals, respectively, both in yields of unity. Additional products identified include CH₃OH and HCHO. Pressure-dependent decomposition pathways were identified in the UV-C photolysis and overall photolysis mechanisms are proposed. Finally, the atmospheric photolysis-lifetimes of CF₂HC(O)CH₃ and CF₃C(O)CH₃ were estimated based on a Tropospheric Ultraviolet Visible (TUV) model calculation.

Molecular graph generation is a key task in drug discovery, aiming to efficiently identify novel compounds with desired properties. While variational autoencoders (VAEs) excel at latent space modeling and discrete flow matching (DFM) enables efficient continuous-time sampling, existing approaches still face critical limitations. VAE decoders struggle with permutation invariance and suffer from one-shot generation bottlenecks, whereas DFM models often rely on a fixed prior initialization that lacks adaptability to specific molecular structures. To address these issues, we propose VFMol, a novel framework that synergistically integrates personalized VAE latent space modeling with the efficient stepwise sampling mechanism of DFM in the discrete space. Specifically, the encoder learns a posterior distribution tailored to each input graph as the generation starting point, thereby enhancing both structural fidelity and diversity. Moreover, we introduce a lightweight property-guided framework based on KAN and classifier-free guidance, enabling conditional generation without auxiliary property predictors. Experiments on two widely used molecular data sets demonstrate that VFMol achieves state-of-the-art performance in terms of molecular structural quality and property controllability, verifying its generality and effectiveness.

Predicting how chemical modifications affect drug binding is central to rational drug design. Free energy perturbation (FEP) calculations provide accurate estimates of these binding affinity changes, but existing methods often require substantial computational resources and expert knowledge. Here, we present QligFEP v2.1.0, a flexible open-source workflow based on a graphical and command-line interface for calculating relative binding free energies using spherical boundary conditions, which dramatically reduces simulation system size by confining simulations to a focused region around the binding site. QligFEP features a configurable restraint algorithm that automatically handles diverse chemical transformations, streamlined setup procedures, and enhanced analysis tools. We validated the method using industry benchmarks comprising 16 protein targets and 639 ligand transformations. Statistical analysis demonstrates that QligFEP achieves comparable accuracy to established commercial and open-source alternatives while requiring only a fraction of the computational resources. The perturbation protocol simulates ∼6250 atoms per perturbation leg and completes transformation replicates in under 2 h on standard computational clusters. Unlike full-system simulations, QligFEP's modest computational requirements make FEP accessible for less than $1 on current AWS spot instances. The combination of accuracy, flexibility, and computational efficiency positions QligFEP as a practical solution for accelerating compound optimization in drug discovery, making rigorous binding affinity predictions accessible for large scale applications and to research groups with limited computational infrastructure.

AI-driven molecular generation encounters a "generation-synthesis gap": most computationally designed molecules cannot be synthesized in laboratories, limiting AI-assisted drug design (AIDD) applications. Current approaches to assess synthetic accessibility (SA) include computer-aided synthesis planning (CASP) tools that perform retrosynthetic searches and machine learning-based SA prediction models that provide rapid scoring. CASP tools are computationally expensive for high-throughput screening, while existing SA prediction models may lack chemical synthesis logic or exhibit variable performance across different chemical spaces. We developed SynFrag, an SA prediction model using fragment assembly autoregressive generation to learn stepwise molecular construction patterns. Self-supervised pretraining on millions of unlabeled molecules enables the learning of dynamic fragment assembly patterns beyond fragment occurrence statistics or reaction step annotations. This approach captures connectivity relationships relevant to synthesis difficulty cliffs, where minor structural changes substantially alter SA. Evaluation across public benchmarks, clinical drugs with intermediates, and AI-generated molecules shows consistent performance across diverse chemical spaces. The model produces subsecond predictions with attention mechanisms corresponding to key reactive sites. SynFrag provides computational efficiency suitable for large-scale screening while maintaining interpretability for detailed SA assessment in drug discovery workflows. Online platform: https://synfrag.simm.ac.cn. Code and data available: https://github.com/simmzx/SynFrag.

Rydberg excited states of molecules pose a challenge for electronic structure calculations because of their highly diffuse electron distribution. Even large and elaborate atomic basis sets tend to underrepresent the long-range tail, overly confining the Rydberg state. An approach is presented here where the molecular orbitals are variationally optimized for the excited state using a plane wave basis set in a Hartree-Fock calculation, followed by a configuration interaction calculation. The use of excited state optimized orbitals greatly enhances the convergence of the many-body calculation, as illustrated by a full configuration interaction calculation of the 2s Rydberg state of H₂. A neural-network-based selective configuration interaction approach is then applied to calculations of 3s and 3p states of H₂O and NH₃. The obtained values of excitation energy are in close agreement with experimental measurements as well as previous many-body calculations where sufficiently diffuse atomic basis sets were used. Calculations using atomic basis sets lacking extra diffuse functions, such as aug-cc-pVTZ, give significantly higher estimates due to confinement of the Rydberg states.