To address the challenge of limited experimental materials data,extensive physical property databases are being developed based on high-throughput computational experiments,such as molecular dynamics simulations.Previ...To address the challenge of limited experimental materials data,extensive physical property databases are being developed based on high-throughput computational experiments,such as molecular dynamics simulations.Previous studies have shown that fine-tuning a predictor pretrained on a computational database to a real system can result in models with outstanding generalization capabilities compared to learning from scratch.This study demonstrates the scaling law of simulationto-real(Sim2Real)transfer learning for several machine learning tasks in materials science.Case studies of three prediction tasks for polymers and inorganic materials reveal that the prediction error on real systems decreases according to a power-law as the size of the computational data increases.Observing the scaling behavior offers various insights for database development,such as determining the sample size necessary to achieve a desired performance,identifying equivalent sample sizes for physical and computational experiments,and guiding the design of data production protocols for downstream real-world tasks.展开更多
Palladium(Pd) has exceptional H_(2) adsorption capacity and has been used as an adsorptive filler in mixed matrix membranes(MMMs) to enhance H_(2) separation performance.However,a high Pd loading(20 wt%-60 wt%) is imp...Palladium(Pd) has exceptional H_(2) adsorption capacity and has been used as an adsorptive filler in mixed matrix membranes(MMMs) to enhance H_(2) separation performance.However,a high Pd loading(20 wt%-60 wt%) is impractical due to cost.In this study,highly dispersed Pd nanoclusters are confined within the channels of mesoporous silica nanoparticles(MSNs),largely improving Pd atom utilization for facilitating H_(2) transport while greatly reducing Pd loading content.MMMs prepared by mixing Pd@MSN with polybenzimidazole matrix,corresponding to a very low Pd loading of 0.6 wt%-3.0 wt%,exhibit much improved H_(2)/CO_(2) separation performance.Specifically,an MMM containing only 2.5 wt% Pd shows mixed-gas separation performance of302.6 barrer of H_(2) permeability and 16.3 of H_(2)/CO_(2) selectivity at 120 ℃,largely surpassing the latest 150 ℃ upper bound.Our work demonstrates the enormous potential for applying Pd-based MMMs in gas separation by reducing noble metal loading by nearly two orders of magnitude.展开更多
A quantum chemistry study of the first singlet(S_(1))and triplet(T_(1))excited states of phenylsulfonyl-carbazole compounds,proposed as useful thermally activated delayed fluorescence(TADF)emitters for organic light e...A quantum chemistry study of the first singlet(S_(1))and triplet(T_(1))excited states of phenylsulfonyl-carbazole compounds,proposed as useful thermally activated delayed fluorescence(TADF)emitters for organic light emitting diode(OLED)applications,was performed with the quantum Equation-Of-Motion Variational Quantum Eigensolver(qEOM-VQE)and Variational Quantum Deflation(VQD)algorithms on quantum simulators and devices.These quantum simulations were performed with double zeta quality basis sets on an active space comprising the highest occupied and lowest unoccupied molecular orbitals(HOMO,LUMO)of the TADF molecules.The differences in energy separations between S_(1) and T_(1)(ΔEST)predicted by calculations on quantum simulators were found to be in excellent agreement with experimental data.Differences of 17 and 88 mHa with respect to exact energies were found for excited states by using the qEOM-VQE and VQD algorithms,respectively,to perform simulations on quantum devices without error mitigation.By utilizing state tomography to purify the quantum states and correct energy values,the large errors found for unmitigated results could be improved to differences of,at most,4 mHa with respect to exact values.Consequently,excellent agreement could be found between values ofΔEST predicted by quantum simulations and those found in experiments.展开更多
The ground and excited state calculations at key geometries, such as the Frank–Condon (FC) and the conical intersection (CI)geometries, are essential for understanding photophysical properties. To compute these geome...The ground and excited state calculations at key geometries, such as the Frank–Condon (FC) and the conical intersection (CI)geometries, are essential for understanding photophysical properties. To compute these geometries on noisy intermediate-scalequantum devices, we proposed a strategy that combined a chemistry-inspired spin-restricted ansatz and a new excited statecalculation method called the variational quantum eigensolver under automatically-adjusted constraints (VQE/AC). Unlike theconventional excited state calculation method, called the variational quantum deflation, the VQE/AC does not require the pre-determination of constraint weights and has the potential to describe smooth potential energy surfaces. To validate this strategy,we performed the excited state calculations at the FC and CI geometries of ethylene and phenol blue at the complete active spaceself-consistent field (CASSCF) level of theory, and found that the energy errors were at most 2 kcal mol−1 even on the ibm_kawasakidevice.展开更多
基金support from MEXT as“Program for Promoting Researches on the Supercomputer Fugaku”(project ID:hp210264)JST CREST(Grant Numbers JPMJCR19I3,JPMJCR22O3,JPMJCR2332)+5 种基金MEXT/JSPS KAKENHI Grant-in-Aid for Scientific Research on Innovative Areas(19H05820)Grant-in-Aid for Scientific Research(A)(19H01132)Grant-in-Aid for Research Activity Start-up(23K19980)Grant-in-Aid for Scientific Research(C)(22K11949)Computational resources were provided by Fugaku at the RIKEN Center for Computational Science,Kobe,Japan(hp210264)the supercomputer at the Research Center for Computational Science,Okazaki,Japan(project:23-IMS-C113,24-IMS-C107).
文摘To address the challenge of limited experimental materials data,extensive physical property databases are being developed based on high-throughput computational experiments,such as molecular dynamics simulations.Previous studies have shown that fine-tuning a predictor pretrained on a computational database to a real system can result in models with outstanding generalization capabilities compared to learning from scratch.This study demonstrates the scaling law of simulationto-real(Sim2Real)transfer learning for several machine learning tasks in materials science.Case studies of three prediction tasks for polymers and inorganic materials reveal that the prediction error on real systems decreases according to a power-law as the size of the computational data increases.Observing the scaling behavior offers various insights for database development,such as determining the sample size necessary to achieve a desired performance,identifying equivalent sample sizes for physical and computational experiments,and guiding the design of data production protocols for downstream real-world tasks.
基金supported by the National Natural Science Foundation of China (22378282,22472113,21988102,U23A20116)the Gusu Innovation and Entrepreneurship Leading Talent Plan (ZXL2023189)+2 种基金the Natural Science Foundation of the Jiangsu Higher Education Institutions of China (23KJB150029)the Key Development Project of Jiangsu Province (BE2022056)the State Key Laboratory of Engines (SKLE),Tianjin University,for support。
文摘Palladium(Pd) has exceptional H_(2) adsorption capacity and has been used as an adsorptive filler in mixed matrix membranes(MMMs) to enhance H_(2) separation performance.However,a high Pd loading(20 wt%-60 wt%) is impractical due to cost.In this study,highly dispersed Pd nanoclusters are confined within the channels of mesoporous silica nanoparticles(MSNs),largely improving Pd atom utilization for facilitating H_(2) transport while greatly reducing Pd loading content.MMMs prepared by mixing Pd@MSN with polybenzimidazole matrix,corresponding to a very low Pd loading of 0.6 wt%-3.0 wt%,exhibit much improved H_(2)/CO_(2) separation performance.Specifically,an MMM containing only 2.5 wt% Pd shows mixed-gas separation performance of302.6 barrer of H_(2) permeability and 16.3 of H_(2)/CO_(2) selectivity at 120 ℃,largely surpassing the latest 150 ℃ upper bound.Our work demonstrates the enormous potential for applying Pd-based MMMs in gas separation by reducing noble metal loading by nearly two orders of magnitude.
基金Q.G.,M.S.,H.C.W.,E.W.,Y.O.,H.N.and N.Y.acknowledge support from MEXT Quantum Leap Flagship Program Grant Number JP-MXS0118067285 and JP-MXS0120319794。
文摘A quantum chemistry study of the first singlet(S_(1))and triplet(T_(1))excited states of phenylsulfonyl-carbazole compounds,proposed as useful thermally activated delayed fluorescence(TADF)emitters for organic light emitting diode(OLED)applications,was performed with the quantum Equation-Of-Motion Variational Quantum Eigensolver(qEOM-VQE)and Variational Quantum Deflation(VQD)algorithms on quantum simulators and devices.These quantum simulations were performed with double zeta quality basis sets on an active space comprising the highest occupied and lowest unoccupied molecular orbitals(HOMO,LUMO)of the TADF molecules.The differences in energy separations between S_(1) and T_(1)(ΔEST)predicted by calculations on quantum simulators were found to be in excellent agreement with experimental data.Differences of 17 and 88 mHa with respect to exact energies were found for excited states by using the qEOM-VQE and VQD algorithms,respectively,to perform simulations on quantum devices without error mitigation.By utilizing state tomography to purify the quantum states and correct energy values,the large errors found for unmitigated results could be improved to differences of,at most,4 mHa with respect to exact values.Consequently,excellent agreement could be found between values ofΔEST predicted by quantum simulations and those found in experiments.
基金This work was supported by JSPS KAKENHI Grant no.JP17H06445,20K05438,and JST Gannt no.JPMJPF2221.We also acknowledge the computer resources provided by the Academic Center for Computing and Media Studies(ACCMS)at Kyoto University and by the Research Center of Computer Science(RCCS)at the Institute for Molecular Science.
文摘The ground and excited state calculations at key geometries, such as the Frank–Condon (FC) and the conical intersection (CI)geometries, are essential for understanding photophysical properties. To compute these geometries on noisy intermediate-scalequantum devices, we proposed a strategy that combined a chemistry-inspired spin-restricted ansatz and a new excited statecalculation method called the variational quantum eigensolver under automatically-adjusted constraints (VQE/AC). Unlike theconventional excited state calculation method, called the variational quantum deflation, the VQE/AC does not require the pre-determination of constraint weights and has the potential to describe smooth potential energy surfaces. To validate this strategy,we performed the excited state calculations at the FC and CI geometries of ethylene and phenol blue at the complete active spaceself-consistent field (CASSCF) level of theory, and found that the energy errors were at most 2 kcal mol−1 even on the ibm_kawasakidevice.
