Search

In this work, we utilize our recently developed machine learning (ML)-corrected ab initio dispersion (aiD) potential, known as D3-ML, which is based on the comprehensive SAPT10K dataset and relies solely on Cartesian coordinates as input, to address the dispersion deficiencies in second-order Møller−Plesset perturbation theory (MP2) by replacing its problematic dispersion and exchange-dispersion terms with D3-ML. This leads to the development of a new dispersion-corrected MP2 method, MP2+aiD(CCD), which outperforms other spin-component-scaled and dispersion-corrected MP2 methods as well as popular ML models for predicting noncovalent interactions across various datasets, including S66 × 8, NAP6 (containing 6 naphthalene dimers), L7, S12L, DNA−ellipticine, the C60 dimer, and C60[6]CPPA. In addition, MP2+aiD(CCD) exhibits comparable or even superior performance compared to the contemporary ωB97M-V functional. The limited performance of pure ML models for systems outside the training set or larger than those in the training set highlights their instability and unpredictability. Conversely, the outstanding performance and transferability of the hybrid MP2+aiD(CCD) method can be attributed to the fusion of the physical electronic structure method and a data-driven ML model, combining the strengths of both sides. This investigation firmly establishes MP2+aiD(CCD) as one of the most accurate and reliable fifth-order scaling correlated wave function methods currently available for modeling noncovalent interactions, even for large complexes. MP2+aiD(CCD) is expected to be reliably applicable in investigating real-life complexes at the hundred-atom scale. [ABSTRACT FROM AUTHOR]

In this study, we introduce SAPT10K, a comprehensive dataset comprising 9982 noncovalent interaction energies and their binding energy components (electrostatics, exchange, induction, and dispersion) for diverse intermolecular complexes of 944 unique dimers. These complexes cover significant portions of the intermolecular potential energy surface and were computed using higher-order symmetry-adapted perturbation theory, SAPT2+(3)(CCD), with a large aug-cc-pVTZ basis set. The dispersion energy values in SAPT10K serve as crucial inputs for refining the ab initio dispersion potentials based on Grimme's D3 and many-body dispersion (MBD) models. Additionally, Δ machine learning (ML) models based on newly developed intermolecular features, which are derived from intermolecular histograms of distances for element/substructure pairs to simultaneously account for local environments as well as long-range correlations, are also developed to address deficiencies of the D3/MBD models, including the inflexibility of their functional forms, the absence of MBD contributions in D3, and the standard Hirshfeld partitioning scheme used in MBD. The developed dispersion models can be applied to complexes involving a wide range of elements and charged monomers, surpassing other popular ML models, which are limited to systems with only neutral monomers and specific elements. The efficient D3-ML model, with Cartesian coordinates as the sole input, demonstrates promising results on a testing set comprising 6714 dimers, outperforming another popular ML model, component-based machine-learned intermolecular force field (CLIFF), by 1.5 times. These refined D3/MBD-ML models have the capability to replace the time-consuming dispersion components in symmetry-adapted perturbation theory-based calculations and can promptly illustrate the dispersion contribution in noncovalent complexes for supramolecular assembly and chemical reactions. [ABSTRACT FROM AUTHOR]

In this work, we develop an accurate and efficient XGBoost machine learning model for predicting the global-density-dependent range-separation parameter, ωGDD, for long-range corrected functional (LRC)-ωPBE. This ω GDD ML model has been built using a wide range of systems (11 466 complexes, ten different elements, and up to 139 heavy atoms) with fingerprints for the local atomic environment and histograms of distances for the long-range atomic correlation for mapping the quantum mechanical range-separation values. The promising performance on the testing set with 7046 complexes shows a mean absolute error of 0.001 117 a 0 − 1 and only five systems (0.07%) with an absolute error larger than 0.01 a 0 − 1 , which indicates the good transferability of our ω GDD ML model. In addition, the only required input to obtain ω GDD ML is the Cartesian coordinates without electronic structure calculations, thereby enabling rapid predictions. LRC-ωPBE ( ω GDD ML ) is used to predict polarizabilities for a series of oligomers, where polarizabilities are sensitive to the asymptotic density decay and are crucial in a variety of applications, including the calculations of dispersion corrections and refractive index, and surpasses the performance of all other popular density functionals except for the non-tuned LRC-ωPBE. Finally, LRC-ωPBE ( ω GDD ML ) combined with (extended) symmetry-adapted perturbation theory is used in calculating noncovalent interactions to further show that the traditional ab initio system-specific tuning procedure can be bypassed. The present study not only provides an accurate and efficient way to determine the range-separation parameter for LRC-ωPBE but also shows the synergistic benefits of fusing the power of physically inspired density functional LRC-ωPBE and the data-driven ω GDD ML model. [ABSTRACT FROM AUTHOR]

While short-range noncovalent interactions (NCIs) are proving to be of importance in many chemical and biological systems, these atypical bindings happen within the so-called van der Waals envelope and pose an enormous challenge for current computational methods. We introduce SNCIAA, a database of 723 benchmark interaction energies of short-range noncovalent interactions between neutral/charged amino acids originated from protein x-ray crystal structures at the "gold standard" coupled-cluster with singles, doubles, and perturbative triples/complete basis set [CCSD(T)/CBS] level of theory with a mean absolute binding uncertainty less than 0.1 kcal/mol. Subsequently, a systematic assessment of commonly used computational methods, such as the second-order Møller−Plesset theory (MP2), density functional theory (DFT), symmetry-adapted perturbation theory (SAPT), composite electronic-structure methods, semiempirical approaches, and the physical-based potentials with machine learning (IPML) on SNCIAA is carried out. It is shown that the inclusion of dispersion corrections is essential even though these dimers are dominated by electrostatics, such as hydrogen bondings and salt bridges. Overall, MP2, ωB97M-V, and B3LYP+D4 turned out to be the most reliable methods for the description of short-range NCIs even in strongly attractive/repulsive complexes. SAPT is also recommended in describing short-range NCIs only if the δMP2 correction has been included. The good performance of IPML for dimers at close-equilibrium and long-range conditions is not transferable to the short-range. We expect that SNCIAA will assist the development/improvement/validation of computational methods, such as DFT, force-fields, and ML models, in describing NCIs across entire potential energy surfaces (short-, intermediate-, and long-range NCIs) on the same footing. [ABSTRACT FROM AUTHOR]

Although reliable methods exist to predict the apparent sound reduction index of heavy, homogeneous isotopic building constructions, these methods are not appropriate for use with lightweight building constructions which typically have critical frequencies in or above the frequency range of interest. Three main methods have been proposed for extending the prediction of flanking sound transmission to frequencies below the critical frequency. The first method is the direct prediction which draws on a database of measurements of the flanking transmission of individual flanking paths. The second method would be a modification of the method in existing standards. This method requires the calculation of the resonant sound transmission factors. However, most of the approaches proposed to calculate the resonant sound transmission factor work only for the case of single leaf homogeneous isotropic building elements and therefore are not readily applicable to complex building elements. The third method is the measurement or prediction of the resonant radiation efficiency and the airborne diffuse field excited radiation efficiency which includes both the resonant and the non-resonant radiation efficiencies. The third method can currently deal with complex building elements if the radiation efficiencies can be measured or predicted. This paper examines these prediction methods. [ABSTRACT FROM AUTHOR]

The present paper provides a modification of near-field acoustical holography (NAH) enabling reconstruction of sound fields in a room in order to study plane structures radiating in enclosed spaces; the new technique is called phonoscopy. A description of the measurement laboratory is given; a single microphone scanner is used to measure the pressure on the hologram plane. Three examples of measurement are presented: a point source located on a rigid wall, a homogeneous wall, and a window mounted in a wall. Maps of the velocity of the structure and of the acoustic intensity radiated are given. The accuracy of phonoscopy is demonstrated by comparison with the accelerometer methods (for measuring vibration velocities) and the two-microphone technique (for measuring acoustic intensities). Some K-space spectra are also presented and analyzed, giving more information about the physics of the vibrator. [ABSTRACT FROM AUTHOR]

Measurements were performed, under semireverberant conditions, to examine uncertainties in the two-microphone cross-spectral method for determining acoustic intensity. Calculations of the pressure gradient error, the error associated with the correction for phase mismatch, and the cross-spectrum random error are discussed. The results of preliminary tests performed under free-field and plane-wave conditions are also presented. The influence of semireverberant conditions on the accuracy of the intensity measurement was studied through the use of a standing-wave tube as a reference sound source. With the tube source placed in a 112-m3 room having reverberation times of about 0.5 s, the radiated power was determined both from the intensity measured inside the tube and from the intensity integrated over a spherical surface enclosing the source. For a measurement system configured to limit the errors listed above to 0.5 dB under free-field conditions, phase correction and random errors of several decibels occurred for measurements in the semireverberant room, when determined for the individual frequency components of the Fourier analysis spectrum. However, the calculated 1/3-octave-band sound power levels agreed within ±0.5 dB for the bands from 160 to 630 Hz. Reductions of the errors by selecting a probe configuration to increase the cross-spectrum phase, increasing the number of time records sampled, and, for measurements within a tube, increasing the tube radius are discussed. [ABSTRACT FROM AUTHOR]