Many-body Contributions Research Articles

LDOS of electron pair and the role of the Pauli exclusion principle

The local density of states (LDOS) for a pair of non-relativistic electrons, influenced by repulsive Coulomb forces, is expressed in term of one-dimensional integrals over Whittaker functions. The computation of the electron pair's LDOS relies on a two-particle Green's function (GF), a generalization of the one-particle GF applicable to a charged particle in an attractive Coulomb potential. By incorporating electron spins and considering the Pauli exclusion principle, the resulting LDOS consists of two components: one originating from an exchange-even two-particle GF and the other from an exchange-odd two-particle GF. The calculated LDOS reveals its dependence on both inter-electron distance and energy. The pseudo-LDOS, derived from the two-body contribution to the LDOS, is examined. This term ensures complete LDOS suppression atr = 0, exhibiting a limited spatial extent, and the reasons for its emergence are elucidated. It is shown that for energies exceeding the effective Hartree energy and inter-electron distances beyond the effective Bohr radius, the impact of many-body contributions to the LDOS can be disregarded. The induced LDOS for an electron pair subjected to an attractive contact potential in two dimensions is evaluated. At small distancesafrom the potential center, a predicted relative difference in LDOS between even and odd state pair reaches approximately 8%. The calculated LDOS is compared with available experimental findings from a two-dimensional electron gas (2DEG). Both exhibit similar oscillation periods; however, the LDOS of the electron pair decays as1/a3, significantly faster than the1/adecay observed for free electrons in a 2DEG.

Protein-Ligand Interaction Energies from Quantum-Chemical Fragmentation Methods: Upgrading the MFCC-Scheme with Many-Body Contributions.

Quantum-chemical fragmentation methods offer an attractive approach for the accurate calculation of protein-ligand interaction energies. While the molecular fractionation with conjugate caps (MFCC) scheme offers a rather straightforward approach for this purpose, its accuracy is often not sufficient. Here, we upgrade the MFCC scheme for the calculation of protein-ligand interactions by including many-body contributions. The resulting fragmentation scheme is an extension of our previously developed MFCC-MBE(2) scheme [J. Comput. Chem. 2023, 44, 1634-1644]. For a diverse test set of protein-ligand complexes, we demonstrate that by upgrading the MFCC scheme with many-body contributions, the error in protein-ligand interaction energies can be reduced significantly, and one generally achieves errors below 20 kJ/mol. Our scheme allows for systematically reducing these errors by including higher-order many-body contributions. As it combines the use of single amino acid fragments with high accuracy, our scheme provides an ideal starting point for the parametrization of accurate machine learning potentials for proteins and protein-ligand interactions.

Depletion potential, correlation functions and demixing transition in model colloid-polymer mixtures

We describe a theoretical framework to calculate depletion potential between colloid particles induced by non-adsorbing ideal polymer chains (s-species) and correlation functions in a coarse-grained one-component system of colloids (c-species). A Padé approximant is used to express the depletion potential as a pair potential with many-body contributions subsumed in it. The depletion potential and correlation functions of c-species are calculated using a self-consistent procedure. Results for several values of size ratio q=σsσc (σs and σc are, respectively diameters of the polymer chain and a colloid particle) and packing fractions of s- and c-species are reported. The spinodal curve and critical point of demixing transition are determined for several values of q. Calculated values are compared with values found from other theories and simulations.

The Near-Sightedness of Many-Body Interactions in Anharmonic Vibrational Couplings.

Couplings between vibrational motions are driven by electronic interactions, and these couplings carry special significance in vibrational energy transfer, multidimensional spectroscopy experiments, and simulations of vibrational spectra. In this investigation, the many-body contributions to these couplings are analyzed computationally in the context of clathrate-like alkali metal cation hydrates, including Cs+(H2O)20, Rb+(H2O)20, and K+(H2O)20, using both analytic and quantum-chemistry potential energy surfaces. Although the harmonic spectra and one-dimensional anharmonic spectra depend strongly on these many-body interactions, the mode-pair couplings were, perhaps surprisingly, found to be dominated by one-body effects, even in cases of couplings to low-frequency modes that involved the motion of multiple water molecules. The origin of this effect was traced mainly to geometric distortion within water monomers and cancellation of many-body effects in differential couplings, and the effect was also shown to be agnostic to the identity of the ion. These outcomes provide new understanding of vibrational couplings and suggest the possibility of improved computational methods for the simulation of infrared and Raman spectra.

CO2 inside sI clathrate-like cages: Automated construction of neural network/machine learned guest-host potential and quantum spectra computations.

We present new results on the underlying guest-host interactions and spectral characterization of a CO2 molecule confined in the cages of the sI clathrate hydrate. Such types of porous solids raise computational challenges, as they are of practical interest as gas storage/capture materials. Accordingly, we have directed our efforts toward addressing their modeling in a proper manner, ensuring the quality of the input data and the efficiency of the computational approaches. The computational procedure for spectral simulations, within the multi-configurational time-dependent Hartree framework, involves the development of a fully coupled Hamiltonian, including an exact kinetic energy operator and a many-body representation of the potential, along with dipole moment surfaces, both obtained through neural network machine learning techniques. The resulting models were automatically trained and tested on extensive datasets generated by PW86PBE-XDM calculations, following the outcome of previous benchmark studies. Our simulations enable us to explore various aspects of the quantized dynamics upon confinement of CO2@D/T, such as constrained rotational-translational quantum motions and the averaged position/orientation of the CO2 guest in comparison to the experimental data available. Particularly notable are the distinct energy patterns observed in the computed spectra for the confined CO2 in the D and T cages, with a considerably high rotational-translational coupling in the CO2@T case. Leveraging reliable computations has proved instrumental, highlighting the sensitivity of the spectral features to the shape and strength of the potential interactions, with the explicit description of many-body contributions being significant.

Systematic improvement of empirical energy functions in the era of machine learning.

The impact of targeted replacement of individual terms in empirical force fields is quantitatively assessed for pure water, dichloromethane (CH Cl ), and solvated K and Cl ions. For the electrostatic interactions, point charges (PCs) and machine learning (ML)-based minimally distributed charges (MDCM) fitted to the molecular electrostatic potential are evaluated together with electrostatics based on the Coulomb integral. The impact of explicitly including second-order terms is investigated by adding a fragment molecular orbital (FMO)-derived polarization energy to an existing force field, in this case CHARMM. It is demonstrated that anisotropic electrostatics reduce the RMSE for water (by 1.4 kcal/mol), CH Cl (by 0.8 kcal/mol) and for solvated Cl clusters (by 0.4 kcal/mol). An additional polarization term can be neglected for CH Cl but further improves the models for pure water (by 1.0 kcal/mol) and hydrated Cl (by 0.4 kcal/mol), and is key for solvated K , reducing the RMSE by 2.3 kcal/mol. A 12-6 Lennard-Jones functional form performs satisfactorily with PC and MDCM electrostatics, but is not appropriate for descriptions that account for the electrostatic penetration energy. The importance of many-body contributions is assessed by comparing a strictly 2-body approach with self-consistent reference data. Two-body interactions suffice for CH Cl whereas water and solvated K and Cl ions require explicit many-body corrections. Finally, a many-body-corrected dimer potential energy surface exceeds the accuracy attained using a conventional empirical force field, potentially reaching that of an FMO calculation. The present work systematically quantifies which terms improve the performance of an existing force field and what reference data to use for parametrizing these terms in a tractable fashion for ML fitting of pure and heterogeneous systems.

Entropy approximations for simple fluids.

Liquid state entropy formulas based on configurational probability distributions are examined for Lennard-Jones fluids across a range temperatures and densities. These formulas are based on expansions of the entropy in a series of n-body distribution functions. We focus on two special cases. One, which we term the "perfect gas" series, starts with the entropy of an ideal gas; the other, which we term the "dense liquid" series, removes a many-body contribution from the ideal gas entropy and reallocates it among the subsequent n-body terms. We show that the perfect gas series is most accurate at low density, while the dense liquid series is most accurate at high density. We propose empirical interpolation methods that are capable of connecting the two series and giving consistent predictions in most situations.

Global Neural Network Potential with Explicit Many-Body Functions for Improved Descriptions of Complex Potential Energy Surface.

The high dimensional machine learning potential (MLP) that has developed rapidly in the past decade represents a giant step forward in large-scale atomic simulation for complex systems. The long-range interaction and the poor description of chemical reactions are typical problems of high dimensional MLP, which are mainly caused by the poor structure discrimination of the atom-centered ML model. Herein, we propose a low-cost neural-network-based MLP architecture for fitting global potential energy surface data, namely, G-MBNN, that can offer improved energy and force resolution on a complex potential energy surface. In G-MBNN, a set of many-body energy terms based on the local atomic environment are explicitly included in computing the total energy─the total energy of the system is written as the sum of atomic energy and many-body energy contributions. These extra many-body energy terms are computationally low-cost and, importantly, can provide easy access to delicate energy terms in complex systems such as very short repulsion, long-range attractions, and sensitive angular-dependent covalent interactions. We implement G-MBNN in the LASP code and demonstrate the improved accuracy of the new framework in representative systems, including ternary-element energy materials LiCoOx, TiO2 with defects, and a series of organic reactions.

Beyond the Status Quo: Density Functional Tight Binding and Neural Network Potentials as a Versatile Simulation Strategy to Characterize Host-Guest Interactions in Metal- and Covalent Organic Frameworks.

In recent years, research focused on synthesis, characterization, and application of metal-organic frameworks (MOFs) has attracted increased interest, from both an experimental as well as a theoretical perspective. Self-consistent charge density functional tight binding (SCC DFTB) in conjunction with a suitable constrained molecular dynamics (MD) simulation protocol provides a versatile and flexible platform for the study of pristine MOFs as well as guest@MOF systems. Although being a semi-empirical quantum mechanical method, SCC DFTB inherently accounts for polarization and many-body contributions, which may become a limiting factor in purely force field-based simulation studies. A number of examples such as CO2, indigo, and drug molecules embedded in various MOF hosts are discussed to highlight the capabilities of the presented simulation approach. Furthermore, a promising extension of the outlined simulation strategy toward the treatment of covalent organic frameworks utilizing state-of-the-art neural network potentials providing a description at DFT accuracy and force field cost is outlined.

Deep neural network finds lack of transferability when modeling supercooled water

Decomposition analyses of the molecular interactions show DNN potentials were unable to correctly represent fundamental individual many-body contributions.

Many-Body Contributions in Water Nanoclusters.

Many-body interactions in water are known to be important but difficult to treat in atomistic models and often are included only as a correction. Polarizable models treat them explicitly, with long-range many-body potentials, within their classical approximation. However, their calculation is computationally expensive. Here, we evaluate how relevant the contributions to the many-body interaction associated with different coordination shells are. We calculate the global energy minimum, and the corresponding configuration, for nanoclusters of up to 20 water molecules. We find that including the first coordination shell, i.e., the five-body term of the central molecule, is enough to approximate within 5% the global energy minimum and its structure. We show that this result is valid for three different polarizable models, the Dang-Chang, the MB-pol, and the Kozack-Jordan potentials. This result suggests a strategy to develop many-body potentials for water that are reliable and, at the same time, computationally efficient.

A Formulation of the Many-Body Expansion (MBE) for Periodic Systems: Application to Several Ice Phases

We introduce a new formulation of the many-body expansion (MBE) for periodic systems and apply it to 7 ice polymorphs (Ih, II, VIII, IX, XIII, XIV, and XV). This new formulation is built via a hierarchical procedure that connects gas-phase clusters that mimic unit cells over finite supercells to infinite solids. For periodic systems, the method is validated by showing that the lattice energies computed up to the 4-body in the MBE reproduce the lattice energies obtained using periodic boundary conditions with an Ewald summation for the 7 ice polymorphs. This development makes it possible to quantify, for the first time, the many-body contributions to the lattice energy of various ice polymorphs, which vary significantly among the 7 ice phases, amounting to between 7 and 24% of the total lattice energies. This development opens the door for obtaining insights into solid-state properties, while leveraging the computational benefits of the MBE.

Polarizable Force Field for Acetonitrile Based on the Single-Center Multipole Expansion.

A polarizable potential function describing the interaction between acetonitrile molecules is introduced. The molecules are described as rigid and linear, with three mass sites corresponding to the CH3 group (methyl, Me), the central carbon atom (C), and the nitrogen atom (N). The electrostatic interaction is represented using a single-center multipole expansion as has been done previously for H2O [Wikfeldt et al., Phys. Chem. Chem. Phys. 15, 16542 (2013)], by including multipole moments from dipole up to and including hexadecapole, as well as anisotropic dipole-dipole, dipole-quadrupole, and quadrupole-quadrupole polarizability tensors. The model is free of point charges. The non-electrostatic part is described in a pair-wise fashion by a Born-Mayer repulsion and damped dispersion attraction. The potential function is parameterized to fit the interaction energy of small (CH3CN)n, n = 2-6, clusters calculated using the PBE0 hybrid functional with an additional atomic many-body dispersion contribution. The parameterized potential function is found to compare well with results of the electronic structure calculations of dissociation curves for different dimer orientations and cohesive properties (the equilibrium volume, cohesive energy, and the bulk modulus) of the α-phase of acetonitrile crystal. The average value of the molecular dipole moment obtained in the α-phase is 5.53 D, corresponding to ca. 40% increase as compared to the dipole moment of an isolated acetonitrile molecule, 3.92 D. The calculated densities of solid and liquid acetonitrile turn out to be 8-10% higher than experimental values. This appears to be caused by an overestimate of the atomic many-body dispersion interaction in the density functional calculations used as input in the parametrization of the potential function.

Improving the accuracy of GIPAW chemical shielding calculations with cluster and fragment corrections

Ab initio methods for predicting NMR parameters in the solid state are an essential tool for assigning experimental spectra and play an increasingly important role in structural characterizations. Recently, a molecular correction (MC) technique has been developed which combines the strengths of plane-wave methods (GIPAW) with single molecule calculations employing Gaussian basis sets. The GIPAW + MC method relies on a periodic calculation performed at a lower level of theory to model the crystalline environment. The GIPAW result is then corrected using a single molecule calculation performed at a higher level of theory. The success of the GIPAW + MC method in predicting a range of NMR parameters is a result of the highly local character of the tensors underlying the NMR observable. However, in applications involving strong intermolecular interactions we find that expanding the region treated at the higher level of theory more accurately captures local many-body contributions to the N15 NMR chemical shielding (CS) tensor. We propose alternative corrections to GIPAW which capture interactions between adjacent molecules at a higher level of theory using either fragment or cluster-based calculations. Benchmark calculations performed on N15 and C13 data sets show that these advanced GIPAW-corrected calculations improve the accuracy of chemical shielding tensor predictions relative to existing methods. Specifically, cluster-based N15 corrections show a 24% and 17% reduction in RMS error relative to GIPAW and GIPAW + MC calculations, respectively. Comparing the benchmark data sets using multiple computational models demonstrates that N15 CS tensor calculations are significantly more sensitive to intermolecular interactions relative to C13. However, fragment and cluster-based corrections that include direct hydrogen bond partners are sufficient for optimizing the accuracy of GIPAW-corrected methods. Finally, GIPAW-corrected methods are applied to the particularly challenging NMR spectral assignment of guanosine dihydrate which contains two guanosine molecules in the asymmetric unit.

Coarse-grained Hamiltonian and effective one component theory of colloidal suspensions

We develop a theory to trace out the solvent degrees of freedom from the grand partition function of colloid-solvent mixtures. Our approach to coarse-graining is based on density functional formalism of density profile and the grand thermodynamic potential of solvent. The solvent-induced interaction which is many-body in character is expressed in terms of two functionals; one that couples the solvent to the colloidal density distribution and the second represents the density–density correlation function of the solvent. The nature, strength, and range of the potential depend on these functionals and therefore on the thermodynamic state of the solvent. The solvent-induced contribution to free energy functional is also derived. A self-consistent procedure is developed to calculate the effective potential between colloidal particles, colloid-solvent, and colloid-colloid correlation functions. The theory is used to investigate both additive and nonadditive binary hard-sphere mixtures. Results are reported for the two systems for several values of packing fractions ηb and ηs and particles diameter ratio q=σsσb where symbols b and s refer to colloid and solvent, respectively. Several interesting features are found: The short-range attractive part of the potential shows non-monotonic dependence on ηb; when ηb is increased from zero, initially the potential becomes more attractive but beyond a certain value of ηb that depends on q, the attraction starts weakening. The repulsive peaks formed at R∼(1+12nq) where R is a distance between centers of colloidal particles expressed in units of σb and n is an integer, become stronger on increasing ηb. These results show that many-body contribution to the effective potential depends in a subtle way on packing fractions ηb,ηs, size ratio q, and on nature of the interaction model and makes a non-negligible contribution to the coarse-grained Hamiltonian.

Stochastic exciton-scattering theory of optical line shapes: Renormalized many-body contributions.

Spectral line shapes provide a window into the local environment coupled to a quantum transition in the condensed phase. In this paper, we build upon a stochastic model to account for non-stationary background processes produced by broad-band pulsed laser stimulation, as distinguished from those for stationary phonon bath. In particular, we consider the contribution of pair-fluctuations arising from the full bosonic many-body Hamiltonian within a mean-field approximation, treating the coupling to the system as a stochastic noise term. Using the Itô transformation, we consider two limiting cases for our model, which lead to a connection between the observed spectral fluctuations and the spectral density of the environment. In the first case, we consider a Brownian environment and show that this produces spectral dynamics that relax to form dressed excitonic states and recover an Anderson-Kubo-like form for the spectral correlations. In the second case, we assume that the spectrum is Anderson-Kubo like and invert to determine the corresponding background. Using the Jensen inequality, we obtain an upper limit for the spectral density for the background. The results presented here provide the technical tools for applying the stochastic model to a broad range of problems.

Evaluation of net interactions for liquid methane based on coarse-grained simulation

The procedure for generating the coarse-grained potential based on the information of liquid state simulations is adopted for liquid methane. The spherically projected potential (SPP) approach consistently generates the effective pair potential for liquid methane, similar to previous liquid water research. The SPP procedure includes a specific feature to extract the net interactions between molecules. For liquid methane, the naïve attractive interaction is incorporated into the SPP, and the potential profile is transferred to the nearly repulsive interaction as a hard-sphere system. A comparison study using two energy profiles, the SPP and the direct interaction contribution of the SPP, gives a similar liquid structure, and the difference between these profiles reflects the existence of the structure-invariant energy profile. The features of the profiles are important for recognizing the characteristics of coarse-grained potential. In the simulation, the potential of mean force (PMF) can be generally evaluated based on the calculated liquid structure. The difference between the PMF and SPP leads to fluctuations in local number density (LND), and it expresses the many-body contribution to liquid structure on the SPP surface. The specific characteristic of the liquid water structure at an ambient condition is explained from the viewpoint of coarse-grained potential by comparing the relation between the SPP and fluctuations in LND for liquid methane with those for liquid water.

The many-body electronic interactions of Fe(II)-porphyrin.

Fe(II)-porphyrin complexes exhibit a diverse range of electronic interactions between the metal and macrocycle. Herein, the incremental full configuration interaction method is applied to the entire space of valence orbitals of a Fe(II)-porphyrin model using a modest basis set. A novel visualization framework is proposed to analyze individual many-body contributions to the correlation energy, providing detailed maps of this complex's highly correlated electronic structure. This technique is used to parse the numerous interactions of two low-lying triplet states (3A2g and 3Eg) and to show that strong metal d-d and macrocycle π-π orbital interactions preferentially stabilize the 3A2g state. d-π interactions, on the other hand, preferentially stabilize the 3Eg state and primarily appear when correlating six electrons at a time. Ultimately, the Fe(II)-porphyrin model's full set of 88 valence electrons are correlated in 275 orbitals, showing the interactions up to the 4-body level, which covers the great majority of correlations in this system.

Resonance-Assisted Hydrogen Bond-Revisiting the Original Concept in the Context of Its Criticism in the Literature.

In the presented research, we address the original concept of resonance-assisted hydrogen bonding (RAHB) by means of the many-body interaction approach and electron density delocalization analysis. The investigated molecular patterns of RAHBs are open chains consisting of two to six molecules in which the intermolecular hydrogen bond stabilizes the complex. Non-RAHB counterparts are considered to be reference systems. The results show the influence of the neighbour monomers on the unsaturated chains in terms of the many-body interaction energy contribution. Exploring the relation between the energy parameters and the growing number of molecules in the chain, we give an explicit extrapolation of the interaction energy and its components in the infinite chain. Electron delocalization within chain motifs has been analysed from three different points of view: three-body delocalization between C=C-C, two-body hydrogen bond delocalization indices and also between fragments (monomers). A many-body contribution to the interaction energy as well as electron density helps to establish the assistance of resonance in the strength of hydrogen bonds upon the formation of the present molecular chains. The direct relation between interaction energy and delocalization supports the original concept, and refutes some of the criticisms of the RAHB idea.

Machine learning many-body potentials for colloidal systems.

Simulations of colloidal suspensions consisting of mesoscopic particles and smaller species such as ions or depletants are computationally challenging as different length and time scales are involved. Here, we introduce a machine learning (ML) approach in which the degrees of freedom of the microscopic species are integrated out and the mesoscopic particles interact with effective many-body potentials, which we fit as a function of all colloid coordinates with a set of symmetry functions. We apply this approach to a colloid-polymer mixture. Remarkably, the ML potentials can be assumed to be effectively state-independent and can be used in direct-coexistence simulations. We show that our ML method reduces the computational cost by several orders of magnitude compared to a numerical evaluation and accurately describes the phase behavior and structure, even for state points where the effective potential is largely determined by many-body contributions.