Edward Hohenstein
Member, Stanford PULSE Institute
All Publications

TeraChem protocol buffers (TCPB): Accelerating QM and QM/MM simulations with a client–server model
THE JOURNAL OF CHEMICAL PHYSICS
2023; 158 (044801)
View details for DOI 10.1063/5.0130886

TeraChem: A graphical processing unitacceleratedelectronic structure package forlargescaleab initio molecular dynamics
WILEY INTERDISCIPLINARY REVIEWSCOMPUTATIONAL MOLECULAR SCIENCE
2020
View details for DOI 10.1002/wcms.1494
View details for Web of Science ID 000552279200001

Strong, Nonresonant Radiation Enhances CisTrans Photoisomerization of Stilbene in Solution.
The journal of physical chemistry. A
2020
Abstract
Previously, it has been demonstrated that external electric fields may be used to exert control over chemical reactivity. In this study, the impact of a strong, nonresonant IR field (1064 nm) on the photoisomerization of cisstilbene is investigated in cyclohexane solution. The design of a suitable reaction vessel for characterization of this effect is presented. The electric field supplied by the pulsed, nearIR radiation (epsilonl = 4.5 * 107 V/cm) enhances the cis trans photoisomerization yield at the red edge of the absorption spectrum (wavelengths between 337 and 340 nm). Within the microliter focal volume, up to 75% of all cisstilbene molecules undergo isomerization to transstilbene in the strong electricfield environment, indicating a significant increase relative to the 35% yield of transstilbene under fieldfree conditions. This result correlates with a 13% enhancement in the transstilbene concentration throughout the bulk solution. Theoretical analysis suggests that the observed change is the result of dynamic Stark shifting of the ground and first excited states, leading to a significant redshift in cisstilbene's absorption spectrum. The predicted increase in the absorption cross section in this range of excitation wavelengths is qualitatively consistent with the experimental increase in transstilbene production.
View details for DOI 10.1021/acs.jpca.0c02732
View details for PubMedID 32585098

Performance of CoupledCluster Singles and Doubles on Modern Stream Processing Architectures.
Journal of chemical theory and computation
2020
Abstract
We develop a new implementation of coupledcluster singles and doubles (CCSD) optimized for the most recent graphical processing unit (GPU) hardware. We find that a single node with 8 NVIDIA V100 GPUs is capable of performing CCSD computations on roughly 100 atoms and 1300 basis functions in less than 1 day. Comparisons against massively parallel implementations of CCSD suggest that more than 64 CPUbased nodes (each with 16 cores) are required to match this performance.
View details for DOI 10.1021/acs.jctc.0c00336
View details for PubMedID 32567305

Psi4 1.4: Opensource software for highthroughput quantum chemistry.
The Journal of chemical physics
2020; 152 (18): 184108
Abstract
PSI4 is a free and opensource ab initio electronic structure program providing implementations of HartreeFock, density functional theory, manybody perturbation theory, configuration interaction, density cumulant theory, symmetryadapted perturbation theory, and coupledcluster theory. Most of the methods are quite efficient, thanks to density fitting and multicore parallelism. The program is a hybrid of C++ and Python, and calculations may be run with very simple text files or using the Python API, facilitating postprocessing and complex workflows; method developers also have access to most of PSI4's core functionalities via Python. Job specification may be passed using The Molecular Sciences Software Institute (MolSSI) QCSCHEMA data format, facilitating interoperability. A rewrite of our toplevel computation driver, and concomitant adoption of the MolSSI QCARCHIVE INFRASTRUCTURE project, makes the latest version of PSI4 well suited to distributed computation of large numbers of independent tasks. The project has fostered the development of independent software components that may be reused in other quantum chemistry programs.
View details for DOI 10.1063/5.0006002
View details for PubMedID 32414239

TeraChem: Accelerating electronic structure and ab initio molecular dynamics with graphical processing units.
The Journal of chemical physics
2020; 152 (22): 224110
Abstract
Developed over the past decade, TeraChem is an electronic structure and ab initio molecular dynamics software package designed from the ground up to leverage graphics processing units (GPUs) to perform largescale ground and excited state quantum chemistry calculations in the gas and the condensed phase. TeraChem's speed stems from the reformulation of conventional electronic structure theories in terms of a set of individually optimized highperformance electronic structure operations (e.g., Coulomb and exchange matrix builds, one and twoparticle density matrix builds) and rankreduction techniques (e.g., tensor hypercontraction). Recent efforts have encapsulated these core operations and provided languageagnostic interfaces. This greatly increases the accessibility and flexibility of TeraChem as a platform to develop new electronic structure methods on GPUs and provides clear optimization targets for emerging parallel computing architectures.
View details for DOI 10.1063/5.0007615
View details for PubMedID 32534542

Nonadiabatic Dynamics of Photoexcited cisStilbene Using Ab Initio Multiple Spawning.
The journal of physical chemistry. B
2020
Abstract
The photochemistry of cisstilbene proceeds through two pathways: cistrans isomerization and ring closure to 4a,4bdihydrophenanthrene (DHP). Despite serving for many decades as a model system for photoisomerization, the photodynamics of cisstilbene is still not fully understood. We use ab initio multiple spawning on a SA2CASSCF(2,2) potential energy surface to simulate the nonadiabatic dynamics of isolated cisstilbene. We find the cyclization (to DHP and cisstilbene) and isomerization (to trans and cisstilbene) reaction coordinates to be orthogonal; branching between the two pathways is determined on the S1 excited state within 150 fs of photoexcitation. Trajectory basis functions (TBFs) undergoing cyclization decay rapidly to the ground state in 250 fs, while TBFs moving along the isomerization coordinate remain on the excited state longer, with the majority decaying between 300 and 500 fs. We observe three avoided crossing regions in the dynamics: two along the isomerization coordinate (displaying pyramidalization and migration of an ethylenic hydrogen or phenyl group), and one DHPlike conical intersection along the cyclization coordinate. The isomeric form of the vibrationally hot photoproducts (as determined by measurement 2 ps after photoexcitation) is determined within less than 50 fs of decay to the ground state mediated by passage through a conical intersection. Excess vibrational energy of ground state cis and transstilbene is channelled into phenyl torsions (with mostly opposing directionality). Our simulations are validated by direct comparison to experiment for the absorption spectrum, branching ratio of the three photoproducts (44:52:4 cisstilbene:transstilbene:DHP), and excited state lifetime (520 ± 40 fs).
View details for DOI 10.1021/acs.jpcb.0c03344
View details for PubMedID 32428407

Holehole TammDancoffapproximated density functional theory: A highly efficient electronic structure method incorporating dynamic and static correlation.
The Journal of chemical physics
2020; 153 (2): 024110
Abstract
The study of photochemical reaction dynamics requires accurate as well as computationally efficient electronic structure methods for the ground and excited states. While timedependent density functional theory (TDDFT) is not able to capture static correlation, complete active space selfconsistent field methods neglect much of the dynamic correlation. Hence, inexpensive methods that encompass both static and dynamic electron correlation effects are of high interest. Here, we revisit holehole TammDancoff approximated (hhTDA) density functional theory for this purpose. The hhTDA method is the holehole counterpart to the more established particleparticle TDA (ppTDA) method, both of which are derived from the particleparticle random phase approximation (ppRPA). In hhTDA, the Nelectron electronic states are obtained through double annihilations starting from a doubly anionic (N+2 electron) reference state. In this way, hhTDA treats ground and excited states on equal footing, thus allowing for conical intersections to be correctly described. The treatment of dynamic correlation is introduced through the use of commonly employed density functional approximations to the exchangecorrelation potential. We show that hhTDA is a promising candidate to efficiently treat the photochemistry of organic and biochemical systems that involve several lowlying excited statesparticularly those with both lowlying ππ* and nπ* states where inclusion of dynamic correlation is essential to describe the relative energetics. In contrast to the existing literature on ppTDA and ppRPA, we employ a functionaldependent choice for the response kernel in pp and hhTDA, which closely resembles the response kernels occurring in linear response and collinear spinflip TDDFT.
View details for DOI 10.1063/5.0003985
View details for PubMedID 32668944

Rank reduced coupled cluster theory. II. Equationofmotion coupledcluster singles and doubles.
The Journal of chemical physics
2019; 151 (16): 164121
Abstract
Equationofmotion coupledcluster singles and doubles (EOMCCSD) is a reliable and popular approach to the determination of electronic excitation energies. Recently, we have developed a rankreduced CCSD (RRCCSD) method that allows the groundstate coupledcluster energy to be determined with lowrank cluster amplitudes. Here, we extend this approach to excitedstate energies through a RREOMCCSD method. We start from the EOMCCSD energy functional and insert lowrank approximations to the doubles amplitudes. The result is an approximate EOMCCSD method with only a quadratic number (in the molecular size) of free parameters in the wavefunction. Importantly, our formulation of RREOMCCSD preserves the size intensivity of the excitation energy and size extensivity of the total energy. Numerical tests of the method suggest that accuracy on the order of 0.050.01 eV in the excitation energy is possible with 1% or less of the original number of wavefunction coefficients; accuracy of better than 0.01 eV can be achieved with about 4% or less of the free parameters. The amount of compression at a given accuracy level is expected to increase with the size of the molecule. The RREOMCCSD method is a new path toward the efficient determination of accurate electronic excitation energies.
View details for DOI 10.1063/1.5121867
View details for PubMedID 31675873

Quantum Computation of Electronic Transitions Using a Variational Quantum Eigensolver.
Physical review letters
2019; 122 (23): 230401
Abstract
We develop an extension of the variational quantum eigensolver (VQE) algorithmmultistate contracted VQE (MCVQE)that allows for the efficient computation of the transition energies between the ground state and several lowlying excited states of a molecule, as well as the oscillator strengths associated with these transitions. We numerically simulate MCVQE by computing the absorption spectrum of an ab initio exciton model of an 18chromophore lightharvesting complex from purple photosynthetic bacteria.
View details for DOI 10.1103/PhysRevLett.122.230401
View details for PubMedID 31298869

Stereoisomer specific reaction of hexabromocyclododecane with reduced sulfur species in aqueous solutions.
Chemosphere
2019; 226: 238245
Abstract
The individual degradation rates of the three dominant stereoisomers (α, β, γ) of hexabromocyclododecane (HBCDD) with bisulfide and polysulfides were investigated at pH 9 in methanol/water solutions at two different temperatures (25 °C and 40 °C). Under all conditions investigated, αHBCDD reacts 10 to 20 times slower with bisulfide than βHBCDD and γHBCDD. The difference in reactivity of HBCDD isomers can be explained by the different populations of stable conformers with large dihedral angle between the vicinal bromine atoms. It was also observed that the reaction of HBCDD with polysulfides is about six times faster than with bisulfide. The experiments performed in solvent mixtures with increased water content at 40 °C indicated that the reaction of HBCDD with bisulfide is faster with higher percentage of water. The much slower abiotic reaction of αHBCDD compared to βHBCDD and γHBCDD could potentially contribute to the fact that αHBCDD is more persistent in the environment than γHBCDD. Only one isomer of tetrabromocyclododecene (TBCDe5) was identified as a degradation product of the reaction of HBCDD with reduced sulfur species. TBCDe5 itself reacts about ten times slower with bisulfide and twenty times slower with polysulfide than HBCDD. The study demonstrates that polysulfides and bisulfides can reduce HBCDD sufficiently in natural anoxic environments and the dominant pathway for the degradation of HBCDD by reduced sulfur species is very likely to be the reductive debromination of vicinal dibromides via concerted antielimination.
View details for DOI 10.1016/j.chemosphere.2019.03.134
View details for PubMedID 30928716

Rank reduced coupled cluster theory. I. Ground state energies and wavefunctions.
The Journal of chemical physics
2019; 150 (16): 164118
Abstract
We propose a compression of the oppositespin coupled cluster doubles amplitudes of the form τijab≡UiaVTVWUjbW, where UiaV are the nVhighest magnitude eigenvectors of the MP2 or MP3 doubles amplitudes. Together with a corresponding parameterization of the oppositespin coupled cluster Lagrange multipliers of the form λabij≡UiaVLVWUjbW, this yields a fully selfconsistent parameterization of reducedrank coupled cluster equations in terms of the Lagrangian L0TVW,LVW. Making this Lagrangian stationary with respect to the LVW parameters yields a perfectly determined set of equations for the TVW equations and coupled cluster energy. These equations can be solved using a Lyapunov equation for the firstorder amplitude updates. We test this "rankreduced coupled cluster" method for coupled cluster singles and doubles in medium sized molecules and find that substantial compression of the T^2 amplitudes is possible with acceptable accuracy.
View details for PubMedID 31042891

Effect of Nonplanarity on ExcitedState Proton Transfer and Internal Conversion in Salicylideneaniline
JOURNAL OF PHYSICAL CHEMISTRY A
2018; 122 (25): 5555–62
Abstract
Salicylideneaniline (SA) is a prototype for excitedstate intramolecular proton transfer (ESIPT) reactions in nonplanar molecules. It is generally understood that the dominant photochemical pathway in this molecule is ESIPT followed by nonradiative decay due to twisting about its phenolic bond. However, the presence of a secondary internal conversion pathway resulting from frustrated proton transfer remains a matter of contention. We perform a detailed nonadiabatic dynamics simulation of SA and definitively identify the existence of both reaction pathways, thereby showing the presence of a secondary photochemical pathway and providing insight into the nature of ESIPT dynamics in molecules with nonplanar groundstate geometries.
View details for PubMedID 29851483

Nonadiabatic Ab Initio Molecular Dynamics with the Floating Occupation Molecular OrbitalComplete Active Space Configuration Interaction Method
JOURNAL OF CHEMICAL THEORY AND COMPUTATION
2018; 14 (1): 339–50
Abstract
We show that the floating occupation molecular orbital complete active space configuration interaction (FOMOCASCI) method is a promising alternative to the widely used complete active space selfconsistent field (CASSCF) method in direct nonadiabatic dynamics simulations. We have simulated photodynamics of three archetypal molecules in photodynamics: ethylene, methaniminium cation, and malonaldehyde. We compared the time evolution of electronic populations and reaction mechanisms as revealed by the FOMOCASCI and CASSCF approaches. Generally, the two approaches provide similar results. Some dynamical differences are observed, but these can be traced back to energetically minor differences in the potential energy surfaces. We suggest that the FOMOCASCI method represents, due to its efficiency and stability, a promising approach for direct ab initio dynamics in the excited state.
View details for PubMedID 29207238

Complete active space configuration interaction from stateaveraged configuration interaction singles natural orbitals: Analytic first derivatives and derivative coupling vectors
JOURNAL OF CHEMICAL PHYSICS
2017; 147 (9): 094104
Abstract
A new complete active space configuration interaction (CASCI) method was recently introduced that uses stateaveraged natural orbitals from the configuration interaction singles method (configuration interaction singles natural orbital CASCI, CISNOCASCI). This method has been shown to perform as well or better than stateaveraged complete active space selfconsistent field for a variety of systems. However, further development and testing of this method have been limited by the lack of available analytic first derivatives of the CISNOCASCI energy as well as the derivative coupling between electronic states. In the present work, we present a Lagrangianbased formulation of these derivatives as well as a highly efficient implementation of the resulting equations accelerated with graphical processing units. We demonstrate that the CISNOCASCI method is practical for dynamical simulations of photochemical processes in molecular systems containing hundreds of atoms.
View details for PubMedID 28886625

Robust and Efficient Spin Purification for Determinantal Configuration Interaction
JOURNAL OF CHEMICAL THEORY AND COMPUTATION
2017; 13 (9): 4162–72
Abstract
The limited precision of floating point arithmetic can lead to the qualitative and even catastrophic failure of quantum chemical algorithms, especially when high accuracy solutions are sought. For example, numerical errors accumulated while solving for determinantal configuration interaction wave functions via Davidson diagonalization may lead to spin contamination in the trial subspace. This spin contamination may cause the procedure to converge to roots with undesired ⟨Ŝ2⟩, wasting computer time in the best case and leading to incorrect conclusions in the worst. In hopes of finding a suitable remedy, we investigate five purification schemes for ensuring that the eigenvectors have the desired ⟨Ŝ2⟩. These schemes are based on projection, penalty, and iterative approaches. All of these schemes rely on a direct, graphics processing unitaccelerated algorithm for calculating the S2c matrixvector product. We assess the computational cost and convergence behavior of these methods by application to several benchmark systems and find that the firstorder spin penalty method is the optimal choice, though firstorder and Löwdin projection approaches also provide fast convergence to the desired spin state. Finally, to demonstrate the utility of these approaches, we computed the lowest several excited states of an openshell silver cluster (Ag19) using the stateaveraged complete active space selfconsistent field method, where spin purification was required to ensure spin stability of the CI vector coefficients. Several lowlying states with significant multiply excited character are predicted, suggesting the value of a multireference approach for modeling plasmonic nanomaterials.
View details for PubMedID 28772070

ExcitedState Dynamics of a Benzotriazole Photostabilizer: 2(2'Hydroxy5'methylphenyl)benzotriazole
JOURNAL OF PHYSICAL CHEMISTRY A
2017; 121 (34): 6377–87
Abstract
A large number of common photostabilizers are based on the 2(2'hydroxyphenyl)benzotriazole structure. One common example is 2(2'hydroxy5'methylphenyl)benzotriazole, or TINUVINP. The excitedstate dynamics of this molecule have been extensively characterized by ultrafast spectroscopies. These experiments have established that upon photoexcitation TINUVINP exhibits excitedstate proton transfer followed by a remarkably fast internal conversion. We simulate the excitedstate dynamics using ab initio multiple spawning (AIMS) and a complete active space configuration interaction (CASCI) wave function with a correction from density functional theory (DFT) to generate the potential energy surfaces. We predict ultrafast proton transfer on the order of 20 fs followed by simultaneous twisting and pyramidalization until a seam of conical intersection is reached. Near the intersection seam population transfer to the ground state is highly efficient. The process is best described as ballistic wavepacket motion from the FranckCondon point along a barrierless coordinate leading to the seam of intersection. Internal conversion is primarily mediated by a minimumenergy conical intersection (MECI) with a high degree of pyramidalization. We posit that the presence of a nitrogen atom in the bond linking the phenyl to the benzotriazole allows for the rapid pyramidalization and the short excitedstate lifetime.
View details for PubMedID 28783946

PSI4 1.1: An OpenSource Electronic Structure Program Emphasizing Automation, Advanced Libraries, and Interoperability
JOURNAL OF CHEMICAL THEORY AND COMPUTATION
2017; 13 (7): 3185–97
Abstract
Psi4 is an ab initio electronic structure program providing methods such as HartreeFock, density functional theory, configuration interaction, and coupledcluster theory. The 1.1 release represents a major update meant to automate complex tasks, such as geometry optimization using completebasisset extrapolation or focalpoint methods. Conversion of the toplevel code to a Python module means that Psi4 can now be used in complex workflows alongside other Python tools. Several new features have been added with the aid of libraries providing easy access to techniques such as density fitting, Cholesky decomposition, and Laplace denominators. The build system has been completely rewritten to simplify interoperability with independent, reusable software components for quantum chemistry. Finally, a wide range of new theoretical methods and analyses have been added to the code base, including functionalgroup and openshell symmetry adapted perturbation theory, densityfitted coupled cluster with frozen natural orbitals, orbitaloptimized perturbation and coupledcluster methods (e.g., OOMP2 and OOLCCD), densityfitted multiconfigurational selfconsistent field, density cumulant functional theory, algebraicdiagrammatic construction excited states, improvements to the geometry optimizer, and the "X2C" approach to relativistic corrections, among many other improvements.
View details for PubMedID 28489372

ExcitedState Dynamics of 2(2 'Hydroxyphenyl)benzothiazole: Ultrafast Proton Transfer and Internal Conversion
JOURNAL OF PHYSICAL CHEMISTRY A
2017; 121 (24): 4595–4605
Abstract
One of the most widely studied model systems for excitedstate proton transfer (ESPT) is the 2(2'hydroxyphenyl)benzothiazole (HBT) molecule. This compound undergoes ultrafast ESPT followed by internal conversion to return to the ground state. In the present work, we simulate the nonadiabatic photochemistry of HBT using ab initio multiple spawning (AIMS) nuclear dynamics and a complete active space configuration interaction (CASCI) method in conjunction with wave functioninDFT embedding to obtain ground and excitedstate potential surfaces onthefly. Our simulation predicts ultrafast ESPT with a time constant of 4854 fs and an excitedstate lifetime of 1.71.8 ps. Following proton transfer, HBT becomes trapped in a metastable keto structure on the S1 state. Eventually, the molecule begins to twist and proceeds toward a seam of intersection with the ground state where internal conversion is highly efficient.
View details for PubMedID 28558210

A directcompatible formulation of the coupled perturbed complete active space selfconsistent field equations on graphical processing units
JOURNAL OF CHEMICAL PHYSICS
2017; 146 (17)
Abstract
We recently developed an algorithm to compute response properties for the stateaveraged complete active space selfconsistent field method (SACASSCF) that capitalized on sparsity in the atomic orbital basis. Our original algorithm was limited to treating small to moderate sized active spaces, but the recent development of graphical processing unit (GPU) based directconfiguration interaction algorithms provides an opportunity to extend this to large active spaces. We present here a directcompatible version of the coupled perturbed equations, enabling us to compute response properties for systems treated with arbitrary active spaces (subject to available memory and computation time). This work demonstrates that the computationally demanding portions of the SACASSCF method can be formulated in terms of seven fundamental operations, including Coulomb and exchange matrix builds and their derivatives, as well as, generalized one and twoparticle density matrix and σ vector constructions. As in our previous work, this algorithm exhibits low computational scaling and is accelerated by the use of GPUs, making possible optimizations and nonadiabatic dynamics on systems with O(1000) basis functions and O(100) atoms, respectively.
View details for DOI 10.1063/1.4979844
View details for Web of Science ID 000400625800014
View details for PubMedID 28477593

Atomistic nonadiabatic dynamics of the LH2 complex with a GPUaccelerated ab initio exciton model.
Physical chemistry chemical physics : PCCP
2017
Abstract
We recently outlined an efficient multitiered parallel ab initio excitonic framework that utilizes time dependent density functional theory (TDDFT) to calculate ground and excited state energies and gradients of large supramolecular complexes in atomistic detail  enabling us to undertake nonadiabatic simulations which explicitly account for the coupled anharmonic vibrational motion of all the constituent atoms in a supramolecular system. Here we apply that framework to the 27 coupled bacteriochlorophylla chromophores which make up the LH2 complex, using it to compute an onthefly nonadiabatic surfacehopping (SH) trajectory of electronically excited LH2. Part one of this article is focussed on calibrating our ab initio exciton Hamiltonian using two key parameters: a shift δ, which corrects for the error in TDDFT vertical excitation energies; and an effective dielectric constant ε, which describes the average screening of the transitiondipole coupling between chromophores. Using snapshots obtained from equilibrium molecular dynamics simulations (MD) of LH2, we tune the values of both δ and ε through fitting to the thermally broadened experimental absorption spectrum, giving a linear absorption spectrum that agrees reasonably well with experiment. In part two of this article, we construct a timeresolved picture of the coupled vibrational and excitation energy transfer (EET) dynamics in the subpicosecond regime following photoexcitation. Assuming FranckCondon excitation of a narrow eigenstate band centred at 800 nm, we use surface hopping to follow a single nonadiabatic dynamics trajectory within the full eigenstate manifold. Consistent with experimental data, this trajectory gives timescales for B800→B850 population transfer (τB800→B850) between 6501050 fs, and B800 population decay (τ800→) between 1050 fs. The dynamical picture that emerges is one of rapidly fluctuating LH2 eigenstates that are delocalized over multiple chromophores and undergo frequent crossing on a femtosecond timescale as a result of the atomic vibrations of the constituent chromophores. The eigenstate fluctuations arise from disorder that is driven by vibrational dynamics with multiple characteristic timescales. The scalability of our ab initio excitonic computational framework across massively parallel architectures opens up the possibility of addressing a wide range of questions, including how specific dynamical motions impact both the pathways and efficiency of electronic energytransfer within large supramolecular systems.
View details for DOI 10.1039/c7cp00492c
View details for PubMedID 28430270

Improved Complete Active Space Configuration Interaction Energies with a Simple Correction from Density Functional Theory
JOURNAL OF CHEMICAL THEORY AND COMPUTATION
2017; 13 (3): 1130–46
Abstract
Recent algorithmic advances have extended the applicability of complete active space configuration interaction (CASCI) methods to molecular systems with hundreds of atoms. While this enables simulation of photochemical dynamics in the condensed phase, the underlying CASCI method has some wellknown problems resulting from a severe neglect of dynamic electron correlation. Vertical excitation energies, vibrational frequencies, and reaction barriers are systematically overestimated; these errors limit the applicability of CASCI. We develop a correction for the CASCI energy using density functional theory (DFT). The DFT correction incorporates the effect of dynamic electron correlation among the core electrons into the CASCI Hamiltonian. We show that the resulting DFTcorrected CASCI approach is applicable in situations where the usual singlereference DFT methods fail, such as the description of systems with biradicaloid electronic structure and conical intersections between ground and excited electronic states. Finally, we apply this DFTcorrected CASCI approach to ultrafast excitedstate proton transfer dynamics. Without the DFT correction, CASCI predicts spurious reaction barriers to these processes, and, as a result, a qualitatively correct description of the dynamics is not possible. With the DFTcorrected CASCI method, we demonstrate qualitative and quantitative agreement with both theory and experiment for two model systems for excitedstate intramolecular proton transfer. Finally, we apply the DFTcorrected CASCI method to excitedstate proton transfer dynamics in a system with more than 150 atoms.
View details for PubMedID 28157312

Analytic formulation of derivative coupling vectors for complete active space configuration interaction wavefunctions with floating occupation molecular orbitals
JOURNAL OF CHEMICAL PHYSICS
2016; 145 (17): 174110
Abstract
The floating occupation molecular orbital complete active space configuration interaction (FOMOCASCI) method is quite promising for the study of nonadiabatic processes. Use of this method directly in nonadiabatic dynamics simulations has been limited by the lack of available firstorder nonadiabatic coupling vectors. Here, an analytic formulation of these derivative coupling vectors is presented for FOMOCASCI wavefunctions using a simple Lagrangianbased approach. The derivative coupling vectors are applied in the optimization of minimum energy conical intersections of an aqueously solvated model compound for the chromophore of the green fluorescent protein (including 100 water molecules). The computational cost of the FOMOCASCI derivative coupling vector is shown to scale quadratically, O(N2), with system size and is applied to systems with up to 1000 atoms.
View details for PubMedID 27825203

Comment on "Positive semidefinite tensor factorizations of the twoelectron integral matrix for lowscaling ab initio electronic structure" [J. Chem. Phys. 143, 064103 (2015)].
journal of chemical physics
2016; 145 (2): 027101?
View details for DOI 10.1063/1.4955316
View details for PubMedID 27421428

"Balancing" the Block DavidsonLiu Algorithm
JOURNAL OF CHEMICAL THEORY AND COMPUTATION
2016; 12 (7): 30033007
Abstract
We describe a simple modification ("balancing") of the block DavidsonLiu eigenvalue algorithm which allows the norms of the Krylov search directions to decrease naturally as convergence is approached. In the context of integraldirect configuration interaction singles and timedependent density functional theory, this provides for efficient utilization of densitybased screening. Tests within the TeraChem GPGPU code exhibit speedups of ∼2× on systems with up to 1500 atoms, with negligible loss in accuracy.
View details for DOI 10.1021/acs.jctc.6b00459
View details for Web of Science ID 000379703800001
View details for PubMedID 27253494

Mechanism for the Enhanced ExcitedState Lewis Acidity of Methyl Viologen
JOURNAL OF THE AMERICAN CHEMICAL SOCIETY
2016; 138 (6): 1868–76
Abstract
Aqueous solutions of methyl viologen (MV(2+)) exhibit anomalous fluorescence behavior. Although it has long fluorescence lifetimes in polar solvents such as acetonitrile, MV(2+) has a short fluorescence lifetime in water. Recent experiments by Kohler and coworkers (Henrich et al. J. Phys. Chem. B 2015, 119, 27372748) have implicated an excitedstate acid/base reaction as the source of the nonradiative decay pathway. While many chemical species exhibit enhanced Brønsted acidity in their excited state, MV(2+) is the first example of a species with enhanced Lewis acidity. Using a complete active space configuration interaction (CASCI) approach, excitedstate molecular dynamics simulations of aqueous MV(2+) are performed in order to test the hypothesis that MV(2+) acts as a Lewis photoacid and to elucidate a mechanism for this behavior. These simulations show that the Lewis acidity of MV(2+) is indeed enhanced by photoexcitation. On its S1 excited state, MV(2+) reacts with water to generate a hydronium ion approximately 1.5 ps after excitation. After the hydronium ion is produced, the corresponding hydroxide ion adds to MV(2+) to form a covalently bound photoproduct and, subsequently, evolves toward a conical intersection.
View details for PubMedID 26725644

Competition Between pipi and CH/pi Interactions: A Comparison of the Structural and Electronic Properties of AlkoxySubstituted 1,8Bis((propyloxyphenyl)ethynyl)naphthalenes
CHEMISTRYA EUROPEAN JOURNAL
2015; 21 (52): 19168–75
Abstract
The structural and electronic consequences of ππ and CH/π interactions in two alkoxysubstituted 1,8bis ((propyloxyphenyl)ethynyl)naphthalenes are explored by using Xray crystallography and electronic structure computations. The crystal structure of analogue 4, bearing an alkoxy side chain in the 4position of each of the phenyl rings, adopts a πstacked geometry, whereas analogue 8, bearing alkoxy groups at both the 2 and the 5positions of each ring, has a geometry in which the rings are splayed away from a πstacked arrangement. Symmetryadapted perturbation theory analysis was performed on the two analogues to evaluate the interactions between the phenylethynyl arms in each molecule in terms of electrostatic, steric, polarization, and London dispersion components. The computations support the expectation that the πstacked geometry of the alkoxyphenyl units in 4 is simply a consequence of maximizing ππ interactions. However, the splayed geometry of 8 results from a more subtle competition between different noncovalent interactions: this geometry provides a favorable antialignment of CO bond dipoles, and two CH/π interactions in which hydrogen atoms of the alkyl side chains interact favorably with the π electrons of the other phenyl ring. These favorable interactions overcome competing ππ interactions to give rise to a geometry in which the phenylethynyl substituents are in an offset, unstacked arrangement.
View details for DOI 10.1002/chem.201502363
View details for Web of Science ID 000368282100041
View details for PubMedID 26568396

An atomic orbitalbased formulation of analytical gradients and nonadiabatic coupling vector elements for the stateaveraged complete active space selfconsistent field method on graphical processing units
JOURNAL OF CHEMICAL PHYSICS
2015; 143 (15)
Abstract
We recently presented an algorithm for stateaveraged complete active space selfconsistent field (SACASSCF) orbital optimization that capitalizes on sparsity in the atomic orbital basis set to reduce the scaling of computational effort with respect to molecular size. Here, we extend those algorithms to calculate the analytic gradient and nonadiabatic coupling vectors for SACASSCF. Combining the low computational scaling with acceleration from graphical processing units allows us to perform SACASSCF geometry optimizations for molecules with more than 1000 atoms. The new approach will make minimal energy conical intersection searches and nonadiabatic dynamics routine for molecular systems with O(10(2)) atoms.
View details for DOI 10.1063/1.4932613
View details for Web of Science ID 000363418400008
View details for PubMedID 26493897

Tensor Hypercontraction SecondOrder MøllerPlesset Perturbation Theory: Grid Optimization and Reaction Energies.
Journal of chemical theory and computation
2015; 11 (7): 304252
Abstract
We have recently introduced the tensor hypercontraction (THC) method for electronic structure, including MP2. Here, we present an algorithm for THCMP2 that lowers the memory requirements as well as the prefactor while maintaining the formal quartic scaling that we demonstrated previously. We also describe a procedure to optimize quadrature grids used in gridbased leastsquares (LS) THCMP2. We apply this algorithm to generate grids for firstrow atoms with less than 100 points/atom while incurring negligible errors in the computed energies. We benchmark the LSTHCMP2 method using optimized grids for a wide variety of tests sets including conformational energies and reaction barriers in both the ccpVDZ and ccpVTZ basis sets. These tests demonstrate that the THC methodology is not limited to small basis sets and that it incurs negligible errors in both absolute and relative energies.
View details for DOI 10.1021/acs.jctc.5b00272
View details for PubMedID 26575741

Analytic first derivatives of floating occupation molecular orbitalcomplete active space configuration interaction on graphical processing units.
journal of chemical physics
2015; 143 (1): 014111?
Abstract
The floating occupation molecular orbitalcomplete active space configuration interaction (FOMOCASCI) method is a promising alternative to the stateaveraged complete active space selfconsistent field (SACASSCF) method. We have formulated the analytic first derivative of FOMOCASCI in a manner that is wellsuited for a highly efficient implementation using graphical processing units (GPUs). Using this implementation, we demonstrate that FOMOCASCI gradients are of similar computational expense to configuration interaction singles (CIS) or timedependent density functional theory (TDDFT). In contrast to CIS and TDDFT, FOMOCASCI can describe multireference character of the electronic wavefunction. We show that FOMOCASCI compares very favorably to SACASSCF in its ability to describe molecular geometries and potential energy surfaces around minimum energy conical intersections. Finally, we apply FOMOCASCI to the excited state hydrogen transfer reaction in methyl salicylate.
View details for DOI 10.1063/1.4923259
View details for PubMedID 26156469

Analytic first derivatives of floating occupation molecular orbitalcomplete active space configuration interaction on graphical processing units.
journal of chemical physics
2015; 143 (1): 014111?
Abstract
The floating occupation molecular orbitalcomplete active space configuration interaction (FOMOCASCI) method is a promising alternative to the stateaveraged complete active space selfconsistent field (SACASSCF) method. We have formulated the analytic first derivative of FOMOCASCI in a manner that is wellsuited for a highly efficient implementation using graphical processing units (GPUs). Using this implementation, we demonstrate that FOMOCASCI gradients are of similar computational expense to configuration interaction singles (CIS) or timedependent density functional theory (TDDFT). In contrast to CIS and TDDFT, FOMOCASCI can describe multireference character of the electronic wavefunction. We show that FOMOCASCI compares very favorably to SACASSCF in its ability to describe molecular geometries and potential energy surfaces around minimum energy conical intersections. Finally, we apply FOMOCASCI to the excited state hydrogen transfer reaction in methyl salicylate.
View details for DOI 10.1063/1.4923259
View details for PubMedID 26156469

Tensor Hypercontraction SecondOrder MollerPlesset Perturbation Theory: Grid Optimization and Reaction Energies
JOURNAL OF CHEMICAL THEORY AND COMPUTATION
2015; 11 (7): 30423052
Abstract
We have recently introduced the tensor hypercontraction (THC) method for electronic structure, including MP2. Here, we present an algorithm for THCMP2 that lowers the memory requirements as well as the prefactor while maintaining the formal quartic scaling that we demonstrated previously. We also describe a procedure to optimize quadrature grids used in gridbased leastsquares (LS) THCMP2. We apply this algorithm to generate grids for firstrow atoms with less than 100 points/atom while incurring negligible errors in the computed energies. We benchmark the LSTHCMP2 method using optimized grids for a wide variety of tests sets including conformational energies and reaction barriers in both the ccpVDZ and ccpVTZ basis sets. These tests demonstrate that the THC methodology is not limited to small basis sets and that it incurs negligible errors in both absolute and relative energies.
View details for DOI 10.1021/acs.jctc.5b00272
View details for Web of Science ID 000358104800015

An atomic orbitalbased formulation of the complete active space selfconsistent field method on graphical processing units.
journal of chemical physics
2015; 142 (22): 224103?
Abstract
Despite its importance, stateoftheart algorithms for performing complete active space selfconsistent field (CASSCF) computations have lagged far behind those for single reference methods. We develop an algorithm for the CASSCF orbital optimization that uses sparsity in the atomic orbital (AO) basis set to increase the applicability of CASSCF. Our implementation of this algorithm uses graphical processing units (GPUs) and has allowed us to perform CASSCF computations on molecular systems containing more than one thousand atoms. Additionally, we have implemented analytic gradients of the CASSCF energy; the gradients also benefit from GPU acceleration as well as sparsity in the AO basis.
View details for DOI 10.1063/1.4921956
View details for PubMedID 26071697

An atomic orbitalbased formulation of the complete active space selfconsistent field method on graphical processing units
JOURNAL OF CHEMICAL PHYSICS
2015; 142 (22)
Abstract
Despite its importance, stateoftheart algorithms for performing complete active space selfconsistent field (CASSCF) computations have lagged far behind those for single reference methods. We develop an algorithm for the CASSCF orbital optimization that uses sparsity in the atomic orbital (AO) basis set to increase the applicability of CASSCF. Our implementation of this algorithm uses graphical processing units (GPUs) and has allowed us to perform CASSCF computations on molecular systems containing more than one thousand atoms. Additionally, we have implemented analytic gradients of the CASSCF energy; the gradients also benefit from GPU acceleration as well as sparsity in the AO basis.
View details for DOI 10.1063/1.4921956
View details for Web of Science ID 000356176600005
View details for PubMedID 26071697

Determination of Hydrogen Bond Structure in Water versus Aprotic Environments To Test the Relationship Between Length and Stability
JOURNAL OF THE AMERICAN CHEMICAL SOCIETY
2015; 137 (17): 57305740
Abstract
Hydrogen bonds profoundly influence the architecture and activity of biological macromolecules. Deep appreciation of hydrogen bond contributions to biomolecular function thus requires a detailed understanding of hydrogen bond structure and energetics and the relationship between these properties. Hydrogen bond formation energies (ΔGf) are enormously more favorable in aprotic solvents than in water, and two classes of contributing factors have been proposed to explain this energetic difference, focusing respectively on the isolated and hydrogenbonded species: (I) water stabilizes the dissociated donor and acceptor groups much better than aprotic solvents, thereby reducing the driving force for hydrogen bond formation; and (II) water lengthens hydrogen bonds compared to aprotic environments, thereby decreasing the potential energy within the hydrogen bond. Each model has been proposed to provide a dominant contribution to ΔGf, but incisive tests that distinguish the importance of these contributions are lacking. Here we directly test the structural basis of model II. Neutron crystallography, NMR spectroscopy, and quantum mechanical calculations demonstrate that OH···O hydrogen bonds in crystals, chloroform, acetone, and water have nearly identical lengths and very similar potential energy surfaces despite ΔGf differences >8 kcal/mol across these solvents. These results rule out a substantial contribution from solventdependent differences in hydrogen bond structure and potential energy after association (model II) and thus support the conclusion that differences in hydrogen bond ΔGf are predominantly determined by solvent interactions with the dissociated groups (model I). These findings advance our understanding of universal hydrogenbonding interactions and have important implications for biology and engineering.
View details for DOI 10.1021/ja512980h
View details for Web of Science ID 000354338500017
View details for PubMedID 25871450

Advances in molecular quantum chemistry contained in the QChem 4 program package
MOLECULAR PHYSICS
2015; 113 (2): 184215
View details for DOI 10.1080/00268976.2014.952696
View details for Web of Science ID 000346358600007

Configuration interaction singles natural orbitals: An orbital basis for an efficient and size intensive multireference description of electronic excited states
JOURNAL OF CHEMICAL PHYSICS
2015; 142 (2): 024102
Abstract
Multireference quantum chemical methods, such as the complete active space selfconsistent field (CASSCF) method, have long been the state of the art for computing regions of potential energy surfaces (PESs) where complex, multiconfigurational wavefunctions are required, such as near conical intersections. Herein, we present a computationally efficient alternative to the widely used CASSCF method based on a complete active space configuration interaction (CASCI) expansion built from the stateaveraged natural orbitals of configuration interaction singles calculations (CISNOs). This CISNOCASCI approach is shown to predict vertical excitation energies of molecules with closedshell ground states similar to those predicted by state averaged (SA)CASSCF in many cases and to provide an excellent reference for a perturbative treatment of dynamic electron correlation. Absolute energies computed at the CISNOCASCI level are found to be variationally superior, on average, to other CASCI methods. Unlike SACASSCF, CISNOCASCI provides vertical excitation energies which are both size intensive and size consistent, thus suggesting that CISNOCASCI would be preferable to SACASSCF for the study of systems with multiple excitable centers. The fact that SACASSCF and some other CASCI methods do not provide a size intensive/consistent description of excited states is attributed to changes in the orbitals that occur upon introduction of noninteracting subsystems. Finally, CISNOCASCI is found to provide a suitable description of the PES surrounding a biradicaloid conical intersection in ethylene.
View details for PubMedID 25591333

Rotational state analysis of AIH(+) by twophoton dissociation
JOURNAL OF MOLECULAR SPECTROSCOPY
2014; 300: 108111
View details for DOI 10.1016/j.jms.2014.03.023
View details for Web of Science ID 000336885800019

Communication: Acceleration of coupled cluster singles and doubles via orbitalweighted leastsquares tensor hypercontraction.
journal of chemical physics
2014; 140 (18): 181102?
Abstract
We apply orbitalweighted leastsquares tensor hypercontraction decomposition of the electron repulsion integrals to accelerate the coupled cluster singles and doubles (CCSD) method. Using accurate and flexible lowrank factorizations of the electron repulsion integral tensor, we are able to reduce the scaling of the most vexing particleparticle ladder term in CCSD from [Formula: see text] to [Formula: see text], with remarkably low error. Combined with a T1transformed Hamiltonian, this leads to substantial practical accelerations against an optimized densityfitted CCSD implementation.
View details for DOI 10.1063/1.4876016
View details for PubMedID 24832246

Tractability gains in symmetryadapted perturbation theory including coupled double excitations: CCD plus ST(CCD) dispersion with natural orbital truncations
JOURNAL OF CHEMICAL PHYSICS
2013; 139 (17)
Abstract
This work focuses on efficient and accurate treatment of the intermolecular dispersion interaction using the CCD+ST(CCD) dispersion approach formulated by Williams et al. [J. Chem. Phys. 103, 4586 (1995)]. We apply natural orbital truncation techniques to the solution of the monomer coupledcluster double (CCD) equations, yielding substantial accelerations in this computationally demanding portion of the SAPT2+(CCD), SAPT2+(3)(CCD), and SAPT2+3(CCD) analyses. It is shown that the wholly ratelimiting dimerbasis particleparticle ladder term can be computed in a reduced natural virtual space which is essentially the same size as the monomerbasis virtual space, with an error on the order of a few thousandths of 1 kcal mol(1). Coupled with our existing natural orbital techniques for the perturbative triple excitation contributions [E. G. Hohenstein and C. D. Sherrill, J. Chem. Phys. 133, 104107 (2010)], this technique provides speedups of greater than an order of magnitude for the evaluation of the complete SAPT2+3(CCD) decomposition, with a total error of a few hundredths of 1 kcal mol(1). The combined approach yields tractability gains of almost 2× in the system size, allowing for SAPT2+3(CCD)/augccpVTZ analysis to be performed for systems such as adeninethymine for the first time. Natural orbital based SAPT2+3(CCD)/augccpVTZ results are presented for stacked and hydrogenbonded configurations of uracil dimer and the adeninethymine dimer.
View details for DOI 10.1063/1.4826520
View details for Web of Science ID 000326922300005
View details for PubMedID 24206282

Tensor Hypercontraction EquationofMotion SecondOrder Approximate Coupled Cluster: Electronic Excitation Energies in O(N4) Time
JOURNAL OF PHYSICAL CHEMISTRY B
2013; 117 (42): 1297212978
Abstract
The tensor hypercontraction (THC) formalism is applied to equationofmotion secondorder approximate coupled cluster singles and doubles (EOMCC2). The resulting method, THCEOMCC2, is shown to scale as [Formula: see text], a reduction of one order from the formal [Formula: see text] scaling of conventional EOMCC2. Numerical tests for a variety of molecules show that errors of less than 0.02 eV are introduced into the excitation energies.
View details for DOI 10.1021/jp4021905
View details for Web of Science ID 000326259800032

Exact tensor hypercontraction: a universal technique for the resolution of matrix elements of local finiterange Nbody potentials in manybody quantum problems.
Physical review letters
2013; 111 (13): 132505?
Abstract
Configurationspace matrix elements of Nbody potentials arise naturally and ubiquitously in the RitzGalerkin solution of manybody quantum problems. For the common specialization of local, finiterange potentials, we develop the exact tensor hypercontraction method, which provides a quantized renormalization of the coordinatespace form of the Nbody potential, allowing for a highly separable tensor factorization of the configurationspace matrix elements. This representation allows for substantial computational savings in chemical, atomic, and nuclear physics simulations, particularly with respect to difficult "exchangelike" contractions.
View details for PubMedID 24116775

Discrete variable representation in electronic structure theory: Quadrature grids for leastsquares tensor hypercontraction
JOURNAL OF CHEMICAL PHYSICS
2013; 138 (19)
Abstract
We investigate the application of molecular quadratures obtained from either standard Becketype grids or discrete variable representation (DVR) techniques to the recently developed leastsquares tensor hypercontraction (LSTHC) representation of the electron repulsion integral (ERI) tensor. LSTHC uses leastsquares fitting to renormalize a twosided pseudospectral decomposition of the ERI, over a physicalspace quadrature grid. While this procedure is technically applicable with any choice of grid, the best efficiency is obtained when the quadrature is tuned to accurately reproduce the overlap metric for quadratic products of the primary orbital basis. Properly selected Becke DFT grids can roughly attain this property. Additionally, we provide algorithms for adopting the DVR techniques of the dynamics community to produce two different classes of grids which approximately attain this property. The simplest algorithm is radial discrete variable representation (RDVR), which diagonalizes the finite auxiliarybasis representation of the radial coordinate for each atom, and then combines LebedevLaikov spherical quadratures and Becke atomic partitioning to produce the full molecular quadrature grid. The other algorithm is full discrete variable representation (FDVR), which uses approximate simultaneous diagonalization of the finite auxiliarybasis representation of the full position operator to produce nondirectproduct quadrature grids. The qualitative features of all three grid classes are discussed, and then the relative efficiencies of these grids are compared in the context of LSTHCDFMP2. Coarse Becke grids are found to give essentially the same accuracy and efficiency as RDVR grids; however, the latter are built from explicit knowledge of the basis set and may guide future development of atomcentered grids. FDVR is found to provide reasonable accuracy with markedly fewer points than either Becke or RDVR schemes.
View details for DOI 10.1063/1.4802773
View details for Web of Science ID 000319291600009
View details for PubMedID 23697409

Quartic scaling secondorder approximate coupled cluster singles and doubles via tensor hypercontraction: THCCC2
JOURNAL OF CHEMICAL PHYSICS
2013; 138 (12)
Abstract
The secondorder approximate coupled cluster singles and doubles method (CC2) is a valuable tool in electronic structure theory. Although the density fitting approximation has been successful in extending CC2 to larger molecules, it cannot address the steep O(N(5)) scaling with the number of basis functions, N. Here, we introduce the tensor hypercontraction (THC) approximation to CC2 (THCCC2), which reduces the scaling to O(N(4)) and the storage requirements to O(N(2)). We present an algorithm to efficiently evaluate the THCCC2 correlation energy and demonstrate its quartic scaling. This implementation of THCCC2 uses a gridbased leastsquares THC (LSTHC) approximation to the densityfitted electron repulsion integrals. The accuracy of the CC2 correlation energy under these approximations is shown to be suitable for most practical applications.
View details for DOI 10.1063/1.4795514
View details for Web of Science ID 000316969500064
View details for PubMedID 23556713

QuantumMechanical Analysis of the Energetic Contributions to pi Stacking in Nucleic Acids versus Rise, Twist, and Slide
JOURNAL OF THE AMERICAN CHEMICAL SOCIETY
2013; 135 (4): 13061316
Abstract
Symmetryadapted perturbation theory (SAPT) is applied to pairs of hydrogenbonded nucleobases to obtain the energetic components of base stacking (electrostatic, exchangerepulsion, induction/polarization, and London dispersion interactions) and how they vary as a function of the helical parameters Rise, Twist, and Slide. Computed average values of Rise and Twist agree well with experimental data for Bform DNA from the Nucleic Acids Database, even though the model computations omitted the backbone atoms (suggesting that the backbone in Bform DNA is compatible with having the bases adopt their ideal stacking geometries). London dispersion forces are the most important attractive component in base stacking, followed by electrostatic interactions. At values of Rise typical of those in DNA (3.36 Å), the electrostatic contribution is nearly always attractive, providing further evidence for the importance of chargepenetration effects in ?? interactions (a term neglected in classical force fields). Comparison of the computed stacking energies with those from model complexes made of the "parent" nucleobases purine and 2pyrimidone indicates that chemical substituents in DNA and RNA account for 2040% of the basestacking energy. A lack of correspondence between the SAPT results and experiment for Slide in RNA basepair steps suggests that the backbone plays a larger role in determining stacking geometries in RNA than in Bform DNA. In comparisons of basepair steps with thymine versus uracil, the thymine methyl group tends to enhance the strength of the stacking interaction through a combination of dispersion and electrosatic interactions.
View details for DOI 10.1021/ja30633091
View details for Web of Science ID 000314492500032
View details for PubMedID 23265256

Communication: Tensor hypercontraction. III. Leastsquares tensor hypercontraction for the determination of correlated wavefunctions
JOURNAL OF CHEMICAL PHYSICS
2012; 137 (22)
Abstract
The manipulation of the rankfour tensor of double excitation amplitudes represents a challenge to the efficient implementation of many electronic structure methods. We present a proof of concept for the approximation of doubles amplitudes in the tensor hypercontraction (THC) representation. In particular, we show how THC can be used to both reduce the scaling with respect to molecular size of coupled cluster singles and doubles (CCSD) (and related methods) by two orders [from O(N(6)) to O(N(4))] and remove the memory bottleneck associated with storage of the doubles amplitudes. The accuracy of correlated methods as integral and amplitude approximations are introduced is examined. For a set of 20 small molecules, single and doubleexcitation configuration interaction (CISD), quadratic CISD (QCISD), and CCSD correlation energies could be reproduced with millihartree accuracy after the introduction of these approximations. Our approach exploits otherwise hidden factorizable tensor structure in both the electron repulsion integrals and the wavefunction coefficients and should be applicable with suitable modifications to many methods in electronic structure theory.
View details for DOI 10.1063/1.4768241
View details for Web of Science ID 000312491400055
View details for PubMedID 23248980

Tensor hypercontraction. II. Leastsquares renormalization
JOURNAL OF CHEMICAL PHYSICS
2012; 137 (22)
Abstract
The leastsquares tensor hypercontraction (LSTHC) representation for the electron repulsion integral (ERI) tensor is presented. Recently, we developed the generic tensor hypercontraction (THC) ansatz, which represents the fourthorder ERI tensor as a product of five secondorder tensors [E. G. Hohenstein, R. M. Parrish, and T. J. Martínez, J. Chem. Phys. 137, 044103 (2012)]. Our initial algorithm for the generation of the THC factors involved a twosided invocation of overlapmetric density fitting, followed by a PARAFAC decomposition, and is denoted PARAFAC tensor hypercontraction (PFTHC). LSTHC supersedes PFTHC by producing the THC factors through a leastsquares renormalization of a spatial quadrature over the otherwise singular 1∕r(12) operator. Remarkably, an analytical and simple formula for the LSTHC factors exists. Using this formula, the factors may be generated with O(N(5)) effort if exact integrals are decomposed, or O(N(4)) effort if the decomposition is applied to densityfitted integrals, using any choice of density fitting metric. The accuracy of LSTHC is explored for a range of systems using both conventional and densityfitted integrals in the context of MP2. The grid fitting error is found to be negligible even for extremely sparse spatial quadrature grids. For the case of densityfitted integrals, the additional error incurred by the grid fitting step is generally markedly smaller than the underlying Coulombmetric density fitting error. The present results, coupled with our previously published factorizations of MP2 and MP3, provide an efficient, robust O(N(4)) approach to both methods. Moreover, LSTHC is generally applicable to many other methods in quantum chemistry.
View details for DOI 10.1063/1.4768233
View details for Web of Science ID 000312491400061
View details for PubMedID 23248986

Accurate Prediction of Noncovalent Interaction Energies with the Effective Fragment Potential Method: Comparison of Energy Components to SymmetryAdapted Perturbation Theory for the S22 Test Set
JOURNAL OF CHEMICAL THEORY AND COMPUTATION
2012; 8 (8): 2835–43
Abstract
Noncovalent interactions play an important role in the stabilization of biological molecules. The effective fragment potential (EFP) is a computationally inexpensive ab initiobased method for modeling intermolecular interactions in noncovalently bound systems. The accuracy of EFP is benchmarked against the S22 and S66 data sets for noncovalent interactions [Jurečka, P.; Šponer, J.; Černý, J.; Hobza, P. Phys. Chem. Chem. Phys.2006, 8, 1985; Řezáč, J.; Riley, K. E.; Hobza, P. J. Chem. Theory Comput.2011, 7, 2427]. The mean unsigned error (MUE) of EFP interaction energies with respect to coupledcluster singles, doubles, and perturbative triples in the complete basis set limit [CCSD(T)/CBS] is 0.9 and 0.6 kcal/mol for S22 and S66, respectively, which is similar to the MUE of MP2 and SCSMP2 for the same data sets, but with a greatly reduced computational expense. Moreover, EFP outperforms classical force fields and popular DFT functionals such as B3LYP and PBE, while newer dispersioncorrected functionals provide a more accurate description of noncovalent interactions. Comparison of EFP energy components with the symmetryadapted perturbation theory (SAPT) energies for the S22 data set shows that the main source of errors in EFP comes from Coulomb and polarization terms and provides a valuable benchmark for further improvements in the accuracy of EFP and force fields in general.
View details for DOI 10.1021/ct200673a
View details for Web of Science ID 000307478800032
View details for PubMedID 26592124

Tensor hypercontraction density fitting. I. Quartic scaling second and thirdorder MollerPlesset perturbation theory
JOURNAL OF CHEMICAL PHYSICS
2012; 137 (4)
Abstract
Many approximations have been developed to help deal with the O(N(4)) growth of the electron repulsion integral (ERI) tensor, where N is the number of oneelectron basis functions used to represent the electronic wavefunction. Of these, the density fitting (DF) approximation is currently the most widely used despite the fact that it is often incapable of altering the underlying scaling of computational effort with respect to molecular size. We present a method for exploiting sparsity in threecenter overlap integrals through tensor decomposition to obtain a lowrank approximation to density fitting (tensor hypercontraction density fitting or THCDF). This new approximation reduces the 4thorder ERI tensor to a product of five matrices, simultaneously reducing the storage requirement as well as increasing the flexibility to regroup terms and reduce scaling behavior. As an example, we demonstrate such a scaling reduction for second and thirdorder perturbation theory (MP2 and MP3), showing that both can be carried out in O(N(4)) operations. This should be compared to the usual scaling behavior of O(N(5)) and O(N(6)) for MP2 and MP3, respectively. The THCDF technique can also be applied to other methods in electronic structure theory, such as coupledcluster and configuration interaction, promising significant gains in computational efficiency and storage reduction.
View details for DOI 10.1063/1.4732310
View details for Web of Science ID 000307611500004
View details for PubMedID 22852593

PSI4: an opensource ab initio electronic structure program
WILEY INTERDISCIPLINARY REVIEWSCOMPUTATIONAL MOLECULAR SCIENCE
2012; 2 (4): 556–65
View details for DOI 10.1002/wcms.93
View details for Web of Science ID 000305393700004

Wavefunction methods for noncovalent interactions
WILEY INTERDISCIPLINARY REVIEWSCOMPUTATIONAL MOLECULAR SCIENCE
2012; 2 (2): 304326
View details for DOI 10.1002/wcms.84
View details for Web of Science ID 000300668800008

Role of LongRange Intermolecular Forces in the Formation of Inorganic Nanoparticle Clusters
JOURNAL OF PHYSICAL CHEMISTRY A
2011; 115 (45): 1293312940
Abstract
An understanding of the role played by intermolecular forces in terms of the electron density distribution is fundamental to the understanding of the selfassembly of molecules in the formation of a molecular crystal. Using ab initio methods capable of describing both shortrange intramolecular interactions and longrange London dispersion interactions arising from electron correlation, analyses of inorganic dimers of As(4)S(4) and As(4)O(6) molecules cut from the structures of realgar and arsenolite, respectively, reveal that the molecules adopt a configuration that closely matches that observed for the crystal. Decomposition of the interaction energies using symmetryadapted perturbation theory reveals that both model dimers feature significant stabilization from electrostatic forces as anticipated by a Lewis acid/Lewis base picture of the interaction. London dispersion forces also contribute significantly to the interaction, although they play a greater role in the realgar structure near equilibrium than in arsenolite.
View details for DOI 10.1021/jp204044k
View details for Web of Science ID 000296685500051
View details for PubMedID 21939256

Largescale symmetryadapted perturbation theory computations via density fitting and Laplace transformation techniques: Investigating the fundamental forces of DNAintercalator interactions
JOURNAL OF CHEMICAL PHYSICS
2011; 135 (17)
Abstract
Symmetryadapted perturbation theory (SAPT) provides a means of probing the fundamental nature of intermolecular interactions. Loworders of SAPT (here, SAPT0) are especially attractive since they provide qualitative (sometimes quantitative) results while remaining tractable for large systems. The application of density fitting and Laplace transformation techniques to SAPT0 can significantly reduce the expense associated with these computations and make even larger systems accessible. We present new factorizations of the SAPT0 equations with densityfitted twoelectron integrals and the first application of Laplace transformations of energy denominators to SAPT. The improved scalability of the DFSAPT0 implementation allows it to be applied to systems with more than 200 atoms and 2800 basis functions. The Laplacetransformed energy denominators are compared to analogous partial Cholesky decompositions of the energy denominator tensor. Application of our new DFSAPT0 program to the intercalation of DNA by proflavine has allowed us to determine the nature of the proflavineDNA interaction. Overall, the proflavineDNA interaction contains important contributions from both electrostatics and dispersion. The energetics of the intercalator interaction are are dominated by the stacking interactions (twothirds of the total), but contain important contributions from the intercalatorbackbone interactions. It is hypothesized that the geometry of the complex will be determined by the interactions of the intercalator with the backbone, because by shifting toward one side of the backbone, the intercalator can form two long hydrogenbonding type interactions. The longrange interactions between the intercalator and the nextnearest base pairs appear to be negligible, justifying the use of truncated DNA models in computational studies of intercalation interaction energies.
View details for DOI 10.1063/1.3656681
View details for Web of Science ID 000296733300007
View details for PubMedID 22070292

Structures of protonated benzene dimer and intermolecular interaction decomposition via symmetryadapted perturbation theory
COMPUTATIONAL AND THEORETICAL CHEMISTRY
2011; 973 (13): 47–52
View details for DOI 10.1016/j.comptc.2011.06.027
View details for Web of Science ID 000295548300008

Accurate Interaction Energies for Problematic DispersionBound Complexes: Homogeneous Dimers of NCCN, P2, and PCCP
JOURNAL OF CHEMICAL THEORY AND COMPUTATION
2011; 7 (9): 2842–51
Abstract
All intermolecular interactions involve London dispersion forces. The accurate treatment of dispersion is essential for the computation of realistic interaction potentials. In general, the most reliable method for computing intermolecular interactions is coupledcluster singles and doubles with perturbative triples [CCSD(T)] in conjunction with a sufficiently flexible Gaussian atomic orbital basis set, a combination which is not routinely applicable due to its excessive computational demands (CPU time, memory, storage). Recently, many theoretical methods have been developed that attempt to account for dispersion in a more efficient manner. It is wellknown that dispersion interactions are more difficult to compute in some systems than others; for example, ππ dispersion is notoriously difficult, while alkanealkane dispersion is relatively simple to compute. In this work, numerous theoretical methods are tested for their ability to compute reliable interaction energies in particularly challenging systems, namely, the P2, PCCP, and NCCN dimers. Symmetryadapted perturbation theory (SAPT) is applied to these dimers to demonstrate their sensitivity to the treatment of dispersion. Due to the small size of these systems, highly accurate CCSD(T) potential energy curves could be estimated at the complete basis set limit. Numerous theoretical methods are tested against the reliable CCSD(T) benchmarks. Methods using a treatment of dispersion that relies on timedependent density functional theory (TDDFT) response functions are found to be the most reliable.
View details for DOI 10.1021/ct200374m
View details for Web of Science ID 000294790400021
View details for PubMedID 26605475

Origin of the Surprising Enhancement of Electrostatic Energies by ElectronDonating Substituents in Substituted Sandwich Benzene Dimers
JOURNAL OF THE AMERICAN CHEMICAL SOCIETY
2011; 133 (34): 1324413247
Abstract
A recent study of substituted facetoface benzene dimers by Lewis and coworkers [J. Am. Chem. Soc. 2011, 133, 38543862] indicated a surprising enhancement of electrostatic interactions for both electronwithdrawing and electrondonating substituents. Here we demonstrate that charge penetration (an attractive electrostatic interaction arising from the overlap of the electron densities on the two monomers) is the cause of this counterintuitive effect. These charge penetration effects are significant at typical ?? interaction distances, and they are not easily described by multipole models. A simple measure of a substituent's electrondonating or electronwithdrawing character, such as the Hammett parameter ?(m), is unlikely to capture subtle charge penetration effects. Indeed, correlation of the relative total energies or relative electrostatic energies with ??(m) breaks down for multiply substituted facetoface benzene dimers.
View details for DOI 10.1021/ja204294q
View details for Web of Science ID 000295551600010
View details for PubMedID 21815686

Challenges of lasercooling molecular ions
NEW JOURNAL OF PHYSICS
2011; 13
View details for DOI 10.1088/13672630/13/6/063023
View details for Web of Science ID 000292139700001

Assessment of the Performance of DFT and DFTD Methods for Describing Distance Dependence of HydrogenBonded Interactions
JOURNAL OF CHEMICAL THEORY AND COMPUTATION
2011; 7 (1): 88–96
Abstract
Noncovalent interactions such as hydrogen bonds, van der Waals forces, and ππ interactions play important roles influencing the structure, stability, and dynamic properties of biomolecules including DNA and RNA base pairs. In an effort to better understand the fundamental physics of hydrogen bonding (Hbonding), we investigate the distance dependence of interaction energies in the prototype bimolecular complexes of formic acid, formamide, and formamidine. Potential energy curves along the Hbonding dissociation coordinate are examined both by establishing reference CCSD(T) interaction energies extrapolated to the complete basis set limit and by assessing the performance of the density functional methods B3LYP, PBE, PBE0, B970, PB86, M052X, and M062X and empirical dispersion corrected methods B3LYPD3, PBED3, PBE0D3, B970D2, BP86D3, and ωB97XD, with basis sets 6311++G(3df,3pd), augccpVDZ, and augccpVTZ. Although Hbonding interactions are dominated by electrostatics, it is necessary to properly account for dispersion interactions to obtain accurate energetics. In order to quantitatively probe the nature of hydrogen bonding interactions as a function of distance, we decompose the interaction energy curves into physically meaningful components with symmetryadapted perturbation theory (SAPT). The SAPT results confirm that the contribution of dispersion and induction are significant at and near equilibrium, although electrostatics dominate. Among the DFT/DFTD techniques, the best overall results are obtained utilizing counterpoisecorrected ωB97XD with the augccpVDZ basis set.
View details for DOI 10.1021/ct100469b
View details for Web of Science ID 000285990300010
View details for PubMedID 26606221

Efficient evaluation of triple excitations in symmetryadapted perturbation theory via secondorder MollerPlesset perturbation theory natural orbitals
JOURNAL OF CHEMICAL PHYSICS
2010; 133 (10)
Abstract
An accurate description of dispersion interactions is required for reliable theoretical studies of many noncovalent complexes. This can be obtained with the wave functionbased formulation of symmetryadapted perturbation theory (SAPT) provided that the contribution of triple excitations to dispersion is included. Unfortunately, this triples dispersion correction limits the applicability of SAPT due to its O(N(7)) scaling. The efficiency of the evaluation of this correction can be greatly improved by removing virtual orbitals from the computation. The error incurred from truncating the virtual space is reduced if secondorder Mo?llerPlesset perturbation theory (MP2) natural orbitals are used in place of the canonical HartreeFock molecular orbitals that are typically used. This approximation is further improved if the triples correction to dispersion is scaled to account for the smaller virtual space. If virtual MP2 natural orbitals are removed according to their occupation numbers, in practice, roughly half of the virtual orbitals can be removed (with the augccpVDZ basis set) with negligible errors if the remaining triples dispersion contribution is scaled. This typically leads to speedups of 1520 times for the cases considered here. By combining the truncated virtual space with the frozen core approximation, the triples correction can be evaluated approximately 50 times faster than the canonical computation. These approximations cause less than 1% error (or at most 0.02?kcal?mol(1)) for the cases considered. Truncation of greater fractions of the virtual space is possible for larger basis sets (leading to speedups of over 40 times before additional speedups from the frozen core approximation).
View details for DOI 10.1063/1.3479400
View details for Web of Science ID 000282475400013
View details for PubMedID 20849164

Density fitting of intramonomer correlation effects in symmetryadapted perturbation theory
JOURNAL OF CHEMICAL PHYSICS
2010; 133 (1)
Abstract
Symmetryadapted perturbation theory (SAPT) offers insight into the nature of intermolecular interactions. In addition, accurate energies can be obtained from the wave functionbased variant of SAPT provided that intramonomer electron correlation effects are included. We apply densityfitting (DF) approximations to the intramonomer correlation corrections in SAPT. The introduction of this approximation leads to an improvement in the computational cost of SAPT by reducing the scaling of certain SAPT terms, reducing the amount of disk I/O, and avoiding the explicit computation of certain types of MO integrals. We have implemented all the intramonomer correlation corrections to SAPT through secondorder under the DF approximation. Additionally, leading thirdorder terms are also implemented. The accuracy of this truncation of SAPT is tested against the S22 test set of Hobza and coworkers [Phys. Chem. Chem. Phys. 8, 1985 (2006)]. When the intramonomer corrections to dispersion are included in SAPT, a mean absolute deviation of 0.30.4 kcal mol(1) is observed for the S22 test set when using an augccpVDZ basis. The computations on the adeninethymine complexes in the S22 test set with an augccpVDZ basis represent the largest SAPT computations to date that include this degree of intramonomer correlation. Computations of this size can now be performed routinely with our newly developed DFSAPT program.
View details for DOI 10.1063/1.3451077
View details for Web of Science ID 000279640600002
View details for PubMedID 20614953

Density fitting and Cholesky decomposition approximations in symmetryadapted perturbation theory: Implementation and application to probe the nature of pipi interactions in linear acenes
JOURNAL OF CHEMICAL PHYSICS
2010; 132 (18)
View details for DOI 10.1063/1.3426316
View details for Web of Science ID 000277756500012

Basis set consistent revision of the S22 test set of noncovalent interaction energies
JOURNAL OF CHEMICAL PHYSICS
2010; 132 (14)
Abstract
The S22 test set of interaction energies for small model complexes [Phys. Chem. Chem. Phys. 8, 1985 (2006)] has been very valuable for benchmarking new and existing methods for noncovalent interactions. However, the basis sets utilized to compute the CCSD(T) interaction energies for some of the dimers are insufficient to obtain converged results. Here we consistently extrapolate all CCSD(T)/complete basis set (CBS) interaction energies using larger basis sets for the CCSD(T) component of the computation. The revised values, which we designate S22A, represent the most accurate results to date for this set of dimers. The new values appear to be within a few hundredths of 1 kcal mol(1) of the true CCSD(T)/CBS limit at the given geometries, but the former S22 values are off by as much as 0.6 kcal mol(1) compared to the revised values. Because some of the most promising methods for noncovalent interactions are already achieving this level of agreement (or better) compared to the S22 data, more accurate benchmark values would clearly be helpful. The MP2, SCSMP2, SCSCCSD, SCS(MI)MP2, and B2PLYPD methods have been tested against the more accurate benchmark set. The B2PLYPD method outperforms all other methods tested here, with a mean average deviation of only 0.12 kcal mol(1). However, the consistent, slight underestimation of the interaction energies computed by the SCSCCSD method (an overall mean absolute deviation and mean deviation of 0.24 and 0.23 kcal mol(1), respectively) suggests that the SCSCCSD method has the potential to become even more accurate with a reoptimization of its parameters for noncovalent interactions.
View details for DOI 10.1063/1.3378024
View details for Web of Science ID 000276973500006
View details for PubMedID 20405982

Accurately Characterizing the pipi Interaction Energies of IndoleBenzene Complexes
JOURNAL OF PHYSICAL CHEMISTRY A
2010; 114 (10): 35763582
Abstract
Noncovalent interactions play a significant role in determining the structures of DNA, RNA, and proteins. Among the most prevalent are pipi interactions, which occur as favorable forces between the aromatic subunits of biochemical molecules. The aromatic side chains of amino acids tryptophan and phenylalanine are commonly modeled with indole and benzene, respectively. We have utilized the MP2 and SCSMP2 methods with the augccpVDZ basis set to compute several Tshaped interaction energies and parallel displaced (PD) threedimensional potential energy surfaces (PESs) at 3.4, 3.6, and 3.8 A vertical separations. At selected minima, CCSD(T) results extrapolated to the completebasisset (CBS) limit were obtained. The trend of the Tshaped interactions has been rationalized by considering electrostatic potential maps and symmetryadapted perturbation theory (SAPT) results. The global minimum has been verified to be the NH/pi Tshaped configuration with a CCSD(T)/CBS interaction energy of 5.62 kcal mol(1). For the PD PESs, the MP2 and SCSMP2 methods predict different minimum configurations. The CCSD(T) method favors the SCSMP2 PD configuration over the MP2 PD configuration by 0.18 kcal mol(1). Among the approximate methods considered here, the SCSCCSD method extrapolated to the CBS limit incurs only around 2% error compared to CCSD(T)/CBS results and is the most reliable for the interaction energies of the indolebenzene complex. Overall, the extension of aromaticity and the highly positive hydrogen of the NH bond, both exhibited by indole, enhance the strength of nonbonded interactions with benzene compared to those in the benzene dimer or in the pyridinebenzene complex.
View details for DOI 10.1021/jp9099495
View details for Web of Science ID 000275327600026
View details for PubMedID 20175508

Assessment of Standard Force Field Models Against HighQuality Ab Initio Potential Curves for Prototypes of pipi, CH/pi, and SH/pi Interactions
JOURNAL OF COMPUTATIONAL CHEMISTRY
2009; 30 (14): 21872193
Abstract
Several popular force fields, namely, CHARMM, AMBER, OPLSAA, and MM3, have been tested for their ability to reproduce highly accurate quantum mechanical potential energy curves for noncovalent interactions in the benzene dimer, the benzeneCH(4) complex, and the benzeneH(2)S complex. All of the force fields are semiquantitatively correct, but none of them is consistently reliable quantitatively. Reoptimization of LennardJones parameters and symmetryadapted perturbation theory analysis for the benzene dimer suggests that better agreement cannot be expected unless more flexible functional forms (particularly for the electrostatic contributions) are employed for the empirical force fields.
View details for DOI 10.1002/jcc.21226
View details for Web of Science ID 000269939600004
View details for PubMedID 19242959

An Assessment of Theoretical Methods for Nonbonded Interactions: Comparison to Complete Basis Set Limit CoupledCluster Potential Energy Curves for the Benzene Dimer, the Methane Dimer, BenzeneMethane, and BenzeneH2S
JOURNAL OF PHYSICAL CHEMISTRY A
2009; 113 (38): 1014610159
Abstract
Large, correlationconsistent basis sets have been used to very closely approximate the coupledcluster singles, doubles, and perturbative triples [CCSD(T)] complete basis set potential energy curves of several prototype nonbonded complexes, the sandwich, Tshaped, and paralleldisplaced benzene dimers, the methanebenzene complex, the H2Sbenzene complex, and the methane dimer. These benchmark potential energy curves are used to assess the performance of several methods for nonbonded interactions, including various spincomponentscaled secondorder perturbation theory (SCSMP2) methods, the spincomponentscaled coupledcluster singles and doubles method (SCSCCSD), density functional theory empirically corrected for dispersion (DFTD), and the metageneralizedgradient approximation functionals M052X and M062X. These approaches generally provide good results for the test set, with the SCS methods being somewhat more robust. M052X underbinds for the test cases considered, while the performances of DFTD and M062X are similar. Density fitting, dual basis, and local correlation approximations all introduce only small errors in the interaction energies but can speed up the computations significantly, particulary when used in combination.
View details for DOI 10.1021/jp9034375
View details for Web of Science ID 000269746200002
View details for PubMedID 19689152

Reactive desorption electrospray ionization mass spectrometry (DESIMS) of natural products of a marine alga
ANALYTICAL AND BIOANALYTICAL CHEMISTRY
2009; 394 (1): 245254
Abstract
Presented here is the optimization and development of a desorption electrospray ionization mass spectrometry (DESIMS) method for detecting natural products on tissue surfaces. Bromophycolides are algal diterpenebenzoate macrolide natural products that have been shown to inhibit growth of the marine fungal pathogen Lindra thalassiae. As such, they have been implicated in antimicrobial chemical defense. However, the defense mechanisms are not yet completely understood. Precise detection of these compounds on algal tissue surfaces under ambient conditions without any disruptive sample processing could shed more light onto the processes involved in chemical defense of marine organisms. Conventional DESIMS directly on algal tissue showed relatively low sensitivity for bromophycolide detection. Sensitivity was greatly improved by the addition of various anions including Cl(), Br(), and CF(3)COO() into the DESI spray solvent. Chloride adduction gave the highest sensitivity for all assayed anions. Density functional optimization of the bromophycolide anionic complexes produced during DESI supported this observation by showing that the chloride complex has the most favorable binding energy. Optimized DESI protocols allowed the direct and unambiguous detection of bromophycolides, including A, B, and E, from the surface of untreated algal tissue.
View details for DOI 10.1007/s0021600926743
View details for Web of Science ID 000265035700025
View details for PubMedID 19277616

Effects of Heteroatoms on Aromatic pipi Interactions: BenzenePyridine and Pyridine Dimer
JOURNAL OF PHYSICAL CHEMISTRY A
2009; 113 (5): 878886
Abstract
Heteroatoms are found in many noncovalent complexes which are of biological importance. The effect of heteroatoms on pipi interactions is assessed via highly accurate quantum chemical computations for the two simplest cases of interactions between aromatic molecules containing heteroatoms, namely, benzenepyridine and pyridine dimer. Benchmark quality estimated coupledcluster through perturbative triples [CCSD(T)] binding energies are computed near the complete basis set limit. Comparisons to the benzene dimer are made to determine the contributions from heteroatoms. The presence of a heteroatom reduces the spatial extent of the pielectron cloud and polarizability of pyridine as compared to benzene. As a result, the magnitude of the dispersion, exchange, and induction interactions in benzenepyridine and pyridine dimer is generally reduced as compared to those for the benzene dimer. Benzenepyridine and pyridine dimer bind more strongly than the benzene dimer in several configurations, and in contrast to the benzene dimer, paralleldisplaced configurations can be significantly preferred over Tshaped configurations. Hydrogens para to a heteroatom are more effective "pihydrogen bond" donors, but aromatic rings with heteroatoms are worse "pihydrogen bond" acceptors.
View details for DOI 10.1021/jp809062x
View details for Web of Science ID 000262902500014
View details for PubMedID 19132847

Assessment of the Performance of the M052X and M062X ExchangeCorrelation Functionals for Noncovalent Interactions in Biomolecules
JOURNAL OF CHEMICAL THEORY AND COMPUTATION
2008; 4 (12): 1996–2000
Abstract
The highly parametrized, empirical exchangecorrelation functionals, M052X and M062X, developed by Zhao and Truhlar have been shown to describe noncovalent interactions better than density functionals which are currently in common use. However, these methods have yet to be fully benchmarked for the types of interactions important in biomolecules. M052X and M062X are claimed to capture "mediumrange" electron correlation; however, the "longrange" electron correlation neglected by these functionals can also be important in the binding of noncovalent complexes. Here we test M052X and M062X for the nucleic acid base pairs in the JSCH2005 database. Using the CCSD(T) binding energies as a benchmark, the performance of these functionals is compared to that of a nonempirical density functional, PBE, and also to that of PBE plus Grimme's empirical dispersion correction, PBED. Due to the importance of "longrange" electron correlation in hydrogenbonded and interstrand base pairs, PBED provides more accurate interaction energies on average for the JSCH2005 database when compared to M052X or M062X. M062X does, however, perform somewhat better than PBED for interactions between stacked base pairs.
View details for DOI 10.1021/ct800308k
View details for Web of Science ID 000261613800002
View details for PubMedID 26620472

Improvement of the coupledcluster singles and doubles method via scaling same and oppositespin components of the double excitation correlation energy
JOURNAL OF CHEMICAL PHYSICS
2008; 128 (12)
Abstract
There has been much interest in costfree improvements to secondorder MøllerPlesset perturbation theory (MP2) via scaling the same and oppositespin components of the correlation energy (spincomponent scaled MP2). By scaling the same and oppositespin components of the double excitation correlation energy from the coupledcluster of single and double excitations (CCSD) method, similar improvements can be achieved. Optimized for a set of 48 reaction energies, scaling factors were determined to be 1.13 and 1.27 for the same and oppositespin components, respectively. Preliminary results suggest that the spincomponent scaled CCSD (SCSCCSD) method will outperform all MP2 type methods considered for describing intermolecular interactions. Potential energy curves computed with the SCSCCSD method for the sandwich benzene dimer and methane dimer reproduce the benchmark CCSD(T) potential curves with errors of only a few hundredths of 1 kcal mol(1) for the minima. The performance of the SCSCCSD method suggests that it is a reliable, lower cost alternative to the CCSD(T) method.
View details for DOI 10.1063/1.2883974
View details for Web of Science ID 000254537200013
View details for PubMedID 18376912

Desorption electrospray ionization reactions between host crown ethers and the influenza neuraminidase inhibitor oseltamivir for the rapid screening of Tamiflu (R)
ANALYST
2008; 133 (11): 15131522
Abstract
Competitive hostguest chemistry on a desorption electrospray ionization mass spectrometry (DESI MS) platform is presented here as the basis for a rapid and quantitative screening method for assessing the quality of Tamiflu capsules with minimal sample preparation. Oseltamivir, the active ingredient in Tamiflu, is an orally active neuraminidase inhibitor antiviral. The high cost and demand for this drug has made it a target for counterfeiters, and reports of counterfeit Tamiflu capsules have already appeared. This urges the development of rapid and sensitive tools for Tamiflu authentication. The method presented here is based on the selective recognition of oseltamivir by crown ethers added to the DESI spray solvent. Crown ethers with various ring sizes were evaluated, all being observed to form stable hostguest complexes with protonated oseltamivir. The relative gas phase stability of each of the hostguest complexes was assessed and the results compared with dispersioncorrected density functional theory. Competive experiments with various pairs of crown ethers were used to assess the relative binding selectivities for oseltamivir. The abundance ratio of the formed complexes was observed to be dependent on the amount of analyte present on the surface of the sample, and independent of DESI geometric factors. These competitive reactions were then successfully tested as a means for the rapid quantitation of oseltamivir by reactive DESI MS without the need for an internal standard.
View details for DOI 10.1039/b809471c
View details for Web of Science ID 000260198100007
View details for PubMedID 18936828