Skip to main content Skip to navigation


Friday, June 14th – Poster Session

Origins of Hydrogen-Bonding Driven X-Ray Emission Splitting in Liquid Water

Andrew Wildman, Francesco Paesani, and Xiaosong Li
University of Washington

X-Ray emission spectroscopy (XES) of liquid water exhibits a spectral feature split by the local hydrogen bonding environment. Despite an abundance of previous research on this feature, the hydrogen bonding motif that gives rise to the splitting is not yet determined. Here, a novel method for calculating XES spectra is presented, and this method is applied to calculate the split spectral feature in liquid water. An analysis of the theoretical origins of the feature splitting is performed on model water clusters, demonstrating that the dominant contributor is the acceptor hydrogen bonding environment. The performance of this method is validated on more realistic hydrogen bonding environments by applying it to clusters extracted from a path-integral molecular dynamics simulation.

The diffusion of oxygen into artificial pancreas: modeling of reaction-diffusion transport into a core-shell geometry

Clarence C. King, Amelia A. Brown, Irmak Sargin, K. M. Bratlie, and Scott P. Beckman
Washington State University
Iowa State University

Fickian diffusion into a core-shell geometry is modeled. The interior core mimics pancreatic Langerhan islets and the exterior shell acts as inert protection. The consumption of oxygen diffusing into the cells is approximated using Michaelis-Menten kinetics. The problem is transformed to dimensionless units and solved numerically. Two regimes are identified, one that is diffusion limited and the other consumption limited. A regression is fit that describes the concentration at the center of the cells as a function of the relevant physical parameters. It is determined that, in a cell culture environment, the cells will remain viable as long as the islet has a radius of around 142 um or less and the encapsulating shell has a radius of less than approximately 283 um. When the islet is on the order of 100 um it is possible for the cells to remain viable in environments with as little as 4.6*10^-2 mol/m^3 O2. The proposed encapsulation scheme is potentially viable.

Lepidocrocite Magnetic Order from Density Functional Theory and Cluster Expansion Methods

Daniel J. Pope, Aurora E. Clark, and Micah Prange
Washington State University

Lepidocrocite, FeOOH, is an important class of metal oxide relevant to a wide range of environmental and industrial applications and finds use in contaminant adsorption, waste remediation, and catalytic activity. Given its applications, understanding its fundamental chemical properties is of considerable interest. The magnetic order of this system, arising from unpaired electrons in the Fe(III) cations, have been experimentally measured to be anti-ferromagnetic, but leave some ambiguity in the precise arrangement of spins. Cluster expansion methods were employed to efficiently explore the configurational space of lepidocrocite iron spins and predict its lowest energy magnetic order. The goal of this research is to provide a more accurate theoretical description of the lepidocrocite ground state and determine the crystal properties of this configuration from first principles calculations.

Lepidocrocite configurations were calculated using DFT formalism using a plane wave basis, PAW potentials, and the GGA-PBE functional. The cluster expansion (CE) method was employed to determine effective cluster interactions (ECIs) of the metal cations. These are used to predict zero temperature, zero pressure ground state configuration whose properties were then calculated by ab initio methods.

Geometric Measure Theory approach to Characterizing Interfacial Roughness

Enrique Alvarado, Zhu Liu, Michael J. Servis, Aurora E. Clark, and Bala Krishnamoorthy
Washington State University

In an area of pure mathematics called Geometric Measure Theory the flat-norm was created to help solve the Plateau problem — that of finding the surface of smallest area which bounds a given curve. We are using the flat-norm to help us measure the interfacial roughness of liquid-liquid interfaces.

Assessment of corrections impacting the accurate computation of 27Al shielding in Al(OH)_4^-

Ernesto Martinez-Baez, Gregory Schenter, and Aurora E. Clark
Washington State University

Quantum mechanical calculation of NMR shielding tensor has proven reliable to improve, assess or interpret NMR experimental findings. Computation of highly accurate shielding constants requires taking modeling beyond the standard quantum chemical approach, incorporating relativistic effects, thermal motion, and intermolecular interactions as well as solvent effects. This project focuses on the accurate computation of 27Al nuclear shielding in Al(OH)_4^- and the experimental Al NMR reference aqueous species Al(H_2 O)_6^(3+). We present the 27Al shielding convergence behavior with respect to basis sets type and size at the HF level. Additionally, electron correlation (CCSD) and dynamical (AIMD structural ensemble) corrections, as well as relativistic and indirect and direct solvation effects, were determined.

Challenges of Cation Solvation Structure

Gregory Schenter
Pacific Northwest National Laboratory

We have established a protocol to use Extended X-ray Absorption Fine Structure (EXAFS) measurement and molecular simulation techniques to characterize local structure of water about simple ions, polyatomic, and multiply charged ions. This involves the generation of ensembles of molecular configurations coupled to the generation of the observed fine structure signal. We employ electron multiple scattering techniques, developed by John Rehr (feff code) coupled with modern molecular simulation techniques. In this presentation I will describe the challenges associated with K+, Na+, and Cs+ solvation. These systems have proven to be a challenge due to the subtle balance between ion-water and water-water interaction and its disruption of the hydrogen bonding network. We will report recent analysis employing the MB-POL parameterized potential, Density Functional Theory with dispersion corrections (revPBE) as well as the recent SCAN functional of Perdew. This work is a collaboration involving gifted Post-Docs and valued colleagues, including Santanu Roy, Mirza Galib, Tim Duignan, Francesco Paesani, Niri Govind, Marcel Baer, Jurg Hutter, Chris Mundy, and John Fulton. It builds upon developments over the years by Liem Dang, Shawn Kathmann, Vanda Glezakou, John Rehr, Maureen McCarthy, Eric Bylaska, John Weare, and Bruce Palmer. This work was supported by the U.S. Department of Energy Office of Basic Energy Sciences, Division of Chemical Sciences, Geosciences & Biosciences.

A Constraint and Configuration Interaction Scheme Implementation onto a Density Functional Tight Binding Method within DFTB+

Gunnar Carlson and Tim Kowalczyk
Western Washington University

The project’s purpose is to reduce the computation time necessary to calculate the energy gap between the ground state and charge transfer states and the calculation for the electronic coupling. This is especially relevant when calculating the electron transfer between organic materials and inorganic materials such as organic dyes adsorbed on titanium oxide in solar panels. To complete this, a seamless addition of a Constrained Density Functional Tight Binding Configuration Interaction (CDFTBCI) option within DFTB+ without hampering the functionalities within the program. This will be expectantly cutting a large amount of computation time in comparison to the Constrained Density Functional Theory Configuration Interaction (CDFTCI) implemented in other programs such as QChem and NWchem. The constraint and configuration interaction implemented in QChem and NWchem are referenced to as a template to the project’s implementation in DFTB+. A stronger foundation to our project is made by referencing Rapacioli’s method of constructing a CDFTBCI method that can calculate the energy gap between the ground state and a charge transfer state. However, Rapacioli’s method limitations prevent it from calculating the electronic coupling between an excited state and a charge transfer state and the electronic coupling between two charge transfer states, both we aim to account for by utilizing the time-independent calculations in DFTB+. The current state of the project is the construction of the constraint as defined by a user’s input file.

Understanding Reaction Mechanisms of Monoterpene Synthases

Hoshin Kim, Narayanan Srividya, Bojana Ginovska, B. Markus Lange, and Simone Raugei
Physical and Computational Sciences Directorate, Pacific Northwest National Laboratory, Richland, Washington

Monoterpene synthases (MTS) catalyze the production of a diverse array of terpenoid products, which are widely used as key ingredients for many industrial applications including pharmaceuticals or biofuels. MTS controls challenging reactions — such as hydrogen atom, hydride, and proton transfers — in a selective and stereospecific manner, using only non-covalent interactions. Despite their importance, a general understanding on how these enzymes achieve high specificities and efficiencies remain elusive. In this presentation, we report an extensive set of molecular dynamics simulations of S-limonene synthase aimed at understanding how steric and electrostatic confinement controls the reactivity of substrates and intermediates. Our simulations show that positively charged amino acids located in the vicinity of the substrate control its conformational dynamics and, at the same time, modulate the solvent accessibility. The potential impact of these residues on catalysis will be discussed.

Spin-coupled Generalized Valence Bond Description of Group 14 Species: The Carbon, Silicon and Germanium Hydrides, XHn (n = 1–4)

Jasper V. K. Thompson, Lu T. Xu, and Thom H. Dunning Jr.
University of Washington

It has long been shown that the first row elements of the p-block demonstrate anomalous chemical behavior relative to the rest of the p-block elements, this is known as the first row anomaly. A prime example of this anomaly is shown through a comparison of the group 14 hydrides, XHn (X = C, Si, Ge; n = 1-4). Using spin-coupled general valence bond (SCGVB) and coupled cluster [CCSD(T)] calculations we examine the XHn series showing the differences in bond energies, molecular structure, and ground state multiplicities. We found that the difference in the strength of the recoupled pair bonds in CH, SiH and GeH provide an explanation the differences we observed. Upon further analysis, two potential causes for the decrease in recoupled pair bond strength, going from CH to SiH to GeH, were identified. First, we observed a decrease in the overlap between the bonding orbitals involved in the recoupled pair bond and second, there was an increase in Pauli repulsion between the electrons involved in the recoupled pair bond and the lone electron that occupies the left over lobe orbital centered on the X atom.

Relativistic Time-Dependent Coupled-Cluster

Lauren N. Koulias, David B. Williams-Young,Daniel R. Nascimento, A. Eugene DePrince, and Xiaosong Li
University of Washington

We present a formalism for the implementation of a relativistic time-dependent equation-of-motion coupled-cluster method with singles and doubles (TD-EOM-CCSD). Unlike many other time-dependent methods, we choose to propagate the time correlation function of the dipole, as opposed to propagating the expectation value of the dipole. This requires the time evolution of only a single quantity, the dipole function, as opposed to the two quantities needed for propagation of the expectation value, both the left and right wavefunction. In this scheme, we variationally include both scalar relativistic effects and spin-orbit coupling. This is achieved by using the exact 2-component (X2C) wavefunction as our reference wavefunction for the coupled-cluster calculation. In order to evaluate the accuracy of X2C-TD-EOM-CCSD, the zero-field splitting in the atomic absorption spectra of open-shell systems is reproduced and compared with experimental splitting values. The results from the X2C-TD-EOM-CC calculations are also compared with TD-EOM-CC calculations using a complex generalized Hartree-Fock reference, in order to examine the effects of the X2C reference on the system.


Luke T. MacHale, Rebecca Hanscam, Eric M. Shepard, and Robert K. Szilagyi
Department of Chemistry and Biochemistry, Montana State University, Bozeman, Montana

Iron-sulfur (Fe-S) clusters are ubiquitous in biology. They are found at active sites of a wide-variety of metalloproteins that are involved in biological electron transfer processes (ferredoxin), substrate activation (hydrogenase, nitrogenase, radical S-adenosylmethionine), regulation of gene expression, and signaling. These [Fe-S] clusters have also been implicated in the chemical evolution of the building blocks of life. In geochemistry, the formally ferrous iron/sulfide ions are the dominant species (FeS)n in solution, while in biology the protein armature accomodates both ferric and ferrous ions. The smallest biological [Fe-S] motif is the rhombic [2Fe-2S] cluster that is composed of two individual [Fe-S] subunits. These subunits can further combine to form [3Fe-3S], [4Fe-4S], and larger modified clusters.

Merged wave functions: Using the Generalized Ionic Fragment Approach, the number of antiferromagnetic pairs were maximized within each cluster structure by alternating the α- and β-spin states of each iron ion to obtain the lowest energy structures. Calculated Gibbs free energy values included dispersion correction and translational entropy corrections for solution phase. Step-wise addition of a neutral [Fe-S] subunit allowed for the systematic development of all lowest energy conformations. In addition to the biologically observed full or incomplete cubane structures we also considered linear, cubic, bent, ‘egg-carton’, and T-shape arrangements for the [Fe-S] subunits up to (FeS)5 composition.

Coarse-grain model: The consideration of each [Fe-S] subunits (as a grain) allows for a simplified representation, where only the connectivity between the grains needs to be defined. Eight types of grain were defined from the various cluster compositions up to [5Fe-5S]. The coarse-grain model was evaluated using a [8Fe-8S] calculated cluster formation enthalpy. Percentage errors in predicted heat of formation are well within an acceptable range (1-2%).

The computational results explain a proposed thermodynamic pathway for cluster formation as the combination of iron and sulfide ions in solution yields neutral species according to an energetically spontaneous cluster and nanoparticle assembly.

Exploring the free energy landscape of small aqueous aluminum species and the role of pH

Maxime Pouvreau, Ernesto Martinez-Baez, Mateusz Dembowski, Carolyn Pearce, Greg Schenter, and Aurora Clark
Washington State University

Al(OH)4- is the predominant aqueous aluminum species under basic conditions, with Raman spectroscopy also supporting the presence of the dimeric aluminate species Al2O(OH)62-. No indisputable evidence exists for other dimeric or higher-order oligomeric species in these conditions, which hinders the understanding of the early stages of nucleation and crystallization of aluminum hydroxides and oxyhydroxides (e.g., gibbsite, boehmite), essential to industrial processing of nuclear waste (Hanford site, WA) and refinement of alumina (Bayer process).

In this work, we explored the free energy landscape of small aluminum species (monomers, dimers) by means of umbrella sampling and density functional tight-binding. For both types of species, two-dimensional potentials of mean force were obtained by biasing the total number of Al-O bonds and the protonation of the species. The lowest-lying species and the paths of interconversion were identified. The Al-Al distance was introduced as a third reaction coordinate to probe the dimerization. Unsurprisingly, Al(OH)4- is clearly favored in basic conditions but additional coordination by one H2O is not excluded. Several low-lying dimers are observed, however for the most stable, moderate alkalinity favors species with di-hydroxo bridging and penta-cordinated Al (Al(OH)82- or Al2(OH)7H2O). Higher pH favors the tetra-coordinated, oxo-bridged species Al2O(OH)62. However, species with di-hydroxo bridging are kinetically favored.

Molecular Dynamics Simulations reveal Insights into Human ITPA-ITP interactions

Michael G. Metro and Yao Houndonougbo
Department of Chemistry and Biochemistry, Eastern Washington University, Cheney, Washington

Protein-ligand interactions are important for biochemical functions and they provide the fundamental understanding of processes in living organisms. Inosine triphosphate pyrophosphatase (ITPA) breaks down purines inosine and xanthosine triphosphate (ITP/XTP) to their monophosphate derivatives to maintain a proper level of nonstandard nucleotides in cells. The disruption of this regulation will increase the risk of genetic disorders. The Human ITPA is a homodimer and the prime substrate ITP binds in the cleft formed by the two lobes of each monomer. We have performed Molecular Dynamics simulations of the human ITPA with ITP bound using all-atom representations of both solute and solvent. The analysis of the simulated trajectories provides useful insights on the enzyme substrate binding and catalysis. This study will contribute to the fundamental understanding of the mechanism of ITPA substrate Binding.

Heteroatom and Substitution Effects on Photophysical Properties and Excited-state Intramolecular Proton Transfers of Amino-type Hydrogen-bonding Molecules: Theoretical Insights

Narissa Kanlayakan, Nawee Kungwan, and Tim Kowalczyk
Western Washington University

Organic bifunctional molecules undergoing excited-state intramolecular proton transfer (ESIPT) have been developed as new and effective fluorescent probes due to their unique photo characteristics, such as unusually large Stokes shift which helps avoid unwanted self-reabsorption. In this work, the effects of heteroatom, electron donating and withdrawing substituents on the photophysical properties and ESIPT behaviors of amino-type hydrogen-bonding molecules and their derivatives have been systematically investigated using density functional theory (DFT) and time-dependent DFT (TD-DFT) methods. The calculated results show that the absorption and emission spectra are red-shifted ongoing from benzimidazole (BI) to benzoxazole (BO), benzothiazole (BT), and imidazo[1,2-a]pyridine (IP), respectively, and Stokes shift of parents compounds (APBI, APBT, APBO, and APIP) was found in this order: APIP > APBI > APBO > APBT. All derivatives provide larger Stokes shift than their parents especially, the APIP derivatives having Ac at R1 position and electron donating group (NMe2) and electron withdrawing group (CN) at R2 and R3 positions, respectively, give larger Stokes shift than other derivatives. Furthermore, the intramolecular hydrogen bonds (Hbonds) of the investigated compounds are strengthened in the excited-state (S1), confirmed with IR spectra and topology analysis. Particularly, those of the APIP derivatives having one H atom on the amino (R1) replaced by an Ac group and CN group at R2 and/or R3 positions are stronger than other compounds and their potential energy curves along the proton transfer (PT) process also show lower PT barrier than other compounds. Therefore, the APIP derivatives having CN at R2 and/or R3 positions, could be good candidates for the fluorescent probes due to their large Stokes shifts, low PT barriers and exothermic reactions in the S1 state.

Specific ion effects at heterogeneous electrolytic interfaces

Nitesh Kumar, Michael J. Servis, and Aurora E. Clark
Washington State University, Pullman, Washington

Ions are ubiquitous, ranging from solvent extraction, atmospheric chemistry to complex biological processes like protein folding, etc. The understanding of electrolyte behavior (especially alkali metal nitrates) at the interface and their effect on the interfacial heterogeneity is important for liquid/liquid extraction to get a deep insight into processes like sorption, complexation and ion extraction occurring at the molecular size interface. Nitrates are also of special interest because of its use as a metal complexants in many industrial processes. For example, consider the Plutonium Uranium Redox Extraction (PUREX) process, the variation in nitrate concentration changes the distribution of metal-ligand complex through a specific mechanism. Over the last two decade, various studies have been done to understand metal nitrate behavior first with non-polarizable and then with polarizable force fields for ions.

This presentation will cover molecular dynamics study of microscopic structures that govern the ion solvation and extraction mechanism at the complex heterogeneous interfaces. Topological analysis of local structures in the capillary wave is performed using a graph theoretical approach and ITIM (Identification of truly interfacial molecules) algorithm. Ion distribution, solvation, and hydrogen bonding structure are studied by converting the vertices into edges by the applications of geometric criterions. Lastly, the effect of ion adsorption on the local structure of thermally induced capillary waves is studied by quantifying their effect on interfacial heterogeneity.

Comparison of Eigenstate Thermalization and Typicality Theories for Quantum Thermodynamics

Phillip Lotshaw and Michael E. Kellman
Institute of Theoretical Science and Department of Chemistry and Biochemistry, University of Oregon

We analyze system-environment thermalization in the context of two leading theories for the foundations of quantum statistical mechanics: the “eigenstate thermalization hypothesis,” which is based on an argument about the quantum dynamics of classically chaotic systems, and “typicality,” which is based on a statistical argument about the states of many-body quantum systems. We analyze the effectiveness of these theories in describing results from simulations with a simple model. We find similarities between the theories that haven’t been emphasized in the literature, and argue that a unified theory gives a more realistic account of quantum thermodynamic behavior.

Ag7Au6 cluster as a highly active catalyst for CO oxidation: Theoretical study

Preeyaporn Poldorn, Yutthana Wongnongwa, Siriporn Jungsuttiwong, Supawadee Namuangruk, and Tim Kowalczyk
Western Washington University, Bellingham, Washington

The adsorption of CO, O2 and CO2 on Ag7Au6 alloy metal nanocluster have been investigated by density-functional theory (DFT). The adsorption energies of the most stable configuration of CO or O2 and CO2 on the Ag7Au6 catalyst were used for study CO oxidation mechanism. In addition, we also investigated reaction pathway for CO oxidation by O2 on Ag7Au6 catalyst including stepwise adsorption and coadsorption mechanisms. Calculation results showed that, coadsorption mechanism with coupling of CO and O2 molecules to form intermediate OC-OO has significantly low energy barrier on the Ag7Au6 nanocluster, therefore the dissociation of the O−O bond of OC-OO to form CO2 and O on the Ag7Au6 nanocluster can be performed easily. Moreover, the coadsorption mechanism was found to be more thermodynamically and kinetically favorable. Such a low activation barrier indicates that the Ag7Au6 catalyst is active for CO oxidation at room temperature and it could be a good candidate catalyst with lower and higher activity.

Mechanical Coupling in the Nitrogenase Complex

Qi Huang, Lewis E. Johnson, Bojana Ginovska, and Simone Raugei
Pacific Northwest National Laboratory

Nitrogenase, the bacterial catalyst for dinitrogen fixation, is a multi-subunit metalloenzyme comprising two electron carrier Fe proteins and one catalytic core MoFe protein. This complex catalyzes the reduction of dinitrogen to two ammonia molecules with electrons provided by the Fe protein upon adenosine triphosphate (ATP) hydrolysis. Association of the Fe protein with two bound ATP molecules with each half of the MoFe protein initiates a series of synchronized events, involving large-scale conformational motions that result in the transfer of an electron from the Fe protein ending on the [7Fe-9S-Mo-homocitrate] catalytic co-factor (FeMo-co) via the [8Fe-7S] P-cluster in the MoFe protein. The hydrolysis of two ATP to adenosine diphosphate (ADP) and inorganic phosphate (Pi) induces the dissociation of the Fe protein (oxidized and with two bound ADP) from the MoFe protein (reduced by 1 electron). This cascade of events must be repeated eight times for every dinitrogen molecule. Here, we conducted an extensive investigation of the conformational dynamics of the nitrogenase complex based on multi-microsecond-long molecular dynamics simulations of both the ATP- and ADP-bound complex, and graph-based analysis of the energy flow in the nitrogenase complex. The motion of the Fe protein on the surface of the MoFe protein is strongly coupled with possible electron transfer pathways as identified from semi-empirical methods. We also observed that the two halves of nitrogenase do not operate independently, but rather are mechanically linked via conformational changes that cause them to anti-cooperate, as suggested by recent kinetic measurements.


Rebecca Hanscam and Robert K. Szilagyi
Department of Chemistry and Biochemistry, Montana State University, Bozeman, Montana

An analysis toolkit was developed for evaluating molecular dynamics trajectories generated by Tinker v8.1.2 ( for aqueous models of free amino acids and short peptides containing at least two cysteine residues. Current functionalities include

  • S(Cys)…S(Cys) distance plot as a function of simulation time;
  • histogram of S(Cys)…S(Cys) distances;
  • 3D cross correlated diagram of the three pairs of S(Cys)…S(Cys) distances;
  • upper-limit for [4Fe-4S] nesting defined by non-threading orientation of the peptide backbone with respect to the S,S,S-triangle formed by the three pairs of cysteines;
  • [2Fe-2S] and [4Fe-4S] cluster nests on the basis of shape complementarity;
  • lifetime of cluster nesting sites;
  • cumulative rate of nest formation, as a measure of reaching an equilibrium state with respect to nest formation and disappearance;
  • Ramachandran plot of phi(C-N–alphaC-C) and psi(N-alphaC–C-C) as a graphical representation for the sampling of peptide conformational space;
  • cross-correlation of peptide backbone conformations between two trajectories.

Worked out examples for CIACGAC and CGGCGGC peptides, description of each algorithm, and the sensitivity of each analysis parameter with respect to the definition of cluster shapes, and minimum/maximum distances are provided. The plots generated by the programs can be directly visualized in PSI-Plot (Poly Software International, Pearl River, NY). The peptide conformational space can be reported also in wrapped and mirrored Ramachandran plots. The S…S distance analysis toolkit can recognize deprotonated cysteine, homocysteine, and thioglycine residues.


The Role of Counterions and Solvation in Determining Intermediates for H2 Evolution

Samantha I. Johnson, Geoffrey Chambers, R. Morris Bullock, and Simone Raugei
Center for Molecular Electrocatalysis, Pacific Northwest National Laboratory, Richland, Washington

Inter- and intramolecular proton transfer is crucial to many catalytic processes. Tautomerization is an important type of intramolecular proton transfer, which has been underappreciated in catalysis. Research in our Center has shown optimization of tautomerization equilibria of protonated catalytic intermediates is critical in the design of catalysts that can efficiently deliver protons and electrons. Tautomerization can be determined by the electronic structure of the molecule or by external factors, such as secondary and outer sphere interactions. In this presentation, we report a combined experimental/theoretical investigation of the electrocatalytic production of H2 by FeP2N (P2N = (Et2PCH2)2NMe). We show how catalysis is impacted by tautomerization, whereby a proton migrates between the N of the ligand and the Fe center as a hydride. Specifically, we show how the nature of the counterion strongly affects the tautomeric equilibrium, altering the thermodynamics preference between possible protonation sites. Indeed, our investigation indicates that hydrogen bonding between N–H and the anionic counterions stabilize the complexes, driving preference for the N–H tautomer.

The 8 Ionization Potentials of Plutonium Revisited

Truong-Son Nguyen and Kirk A. Peterson
Washington State University

The ionization potential (IP), or ionization energy, is a well-known concept, and an investigation into a set of IPs can lend insight into the atom’s electronic structure, such as the role of nearby orbitals in stabilizing a given configuration, and in this case, which so-called “contributions” were significant along the way. The consistent and reliable Feller-Peterson-Dixon (FPD) method was used to accurately calculate the first eight IPs of plutonium (Pu). The contributions to the FPD method (namely, CBS limit, CV, QED, SO, Gaunt) are explicitly shown. It was found that the final three IPs adopted by NIST are much too large because of incorrect ground states.

Interfacial organization and diffusion of water at surfaces of boehmite (γ-AlO(OH)) and gibbsite (γ-Al(OH)3) with low miller indices

William Smith, Maxime Pouvreau, and Aurora Clark
Department of Chemistry, Washington State University

The precipitation of aluminum (oxy)hydroxides including boehmite (γ-AlO(OH)) and gibbsite (γ-Al(OH)3) is a major part of industrial applications like alumina refinement and the nuclear waste process at the Hanford Site, WA. The liquid/mineral interfaces have been widely studied at the basal surface while the edges have been largely unexplored despite most phenomena, such as crystal growth and dissolution, occurring at the edges. In this work we investigated the interface between water and the basal surface in addition to the predominant, low Miller index, edges of boehmite and gibbsite through the use of classical molecular dynamics and static density functional theory (DFT) calculations. Several unique adsorption modes of water have been identified at each surface, with adsorption mode population dependent on the surface termination. MD simulations provided the geometries for the calculation of DFT binding energies of adsorption modes on the basal plane suggest a statistically significant difference in energies. Additionally, the dynamics at the interface is of interest, primarily to study the reorientational and translational behavior of water molecules within the first hydration layer as well as the diffusion of water away from the surface. Understanding pure water’s structure and dynamics at the interface is an initial step in understanding the perturbation of the surface due to the electrolyte in highly alkaline conditions.

Friday, June 14th – Presentations


Role of Structural Dynamics in the Catalytic Activity of Inosine Triphosphate Pyrophosphatase (ITPA): A Molecular Dynamic Simulation Study

Yao Houndonougbo
 Department of Chemistry and Biochemistry, Eastern Washington University, Cheney, Washington

Inosine triphosphatase (ITPA) catalyzes the hydrolysis of non-canonical purine nucleotides inosine triphosphate (ITP), deoxy ITP (dITP), and xanthosine triphosphate (XTP) to the corresponding nucleoside monophosphate and pyrophosphate. Genome instability may occur as a result of the accumulation of ITP, dITP, and XTP in cell. The human ITPA acts as a homodimer which consists of central beta-sheet with two lobes having mixed alpha-beta structures. The variants P-32 and R-178 have been linked to epileptic encephalopathy and purine analog drug toxicity. We have investigated the structural dynamics of the human ITPA and its clinical mutants using molecular dynamics (MD) simulations method. The results of the micro-second simulations are used to give a possible structural explanation of the observed reduction of the enzymatic activity in the mutants. This study is an important starting step toward the protein engineering and design of ITPA.

Examination of Magnetism in a CrI3 Two-Dimensional Material

Ryan A. Beck, Hongbin Liu, Shichao Sun, Peter V. Sushko, Xiaodong Xu, and Xiaosong Li
University of Washington, Department of Chemistry

Two-dimensional van der Waals material have attracted attention for their potential to reduce the size and improve the efficiency in microelectronics given their magnetic properties. Chromium trihalides (CrX3) materials have received little attention in the past, however CrI3 has been shown to have the highest magnetic anisotropy of the CrX3 materials, is relatively simple to prepare (among the CrX3 compounds), has a band gap at 1.2 eV, is cleaved easily, and is stable in air. CrI3 also exhibits unique magnetic properties, specifically the layers of CrI3 are known to have an intralayer ferromagnetic (FM) state, but an interlayer antiferromagnetic (AFM) state. Modulation of the magnetic field has been shown to drive a magnetic transition between the layers which allows for a device with electrically tunable magnetism. As such a theoretical understanding of the magnetic interactions is required, however computational simulations of the CrI3 system have been unable to realize the AFM ordering of the material. As such, in this study the magnetic properties of multilayered CrI3 and their dependence on the geometry of the CrI3 lattice, as well as the presence of defects within the material are investigated via ab initio methods to gain understanding of the material to allow for future applications within microelectronics.

Exploring Magnetic Responses of Heavy Elements

Shichao Sun and Xiaosong Li
University of Washington

Molecular properties in the presence of magnetic field can provide important insights into the magnetic and spin characteristics of heavy elements. However, they are generally difficult to calculate mainly due to several theoretical challenges, including the gauge-origin problem and the spin non-collinearity. In this work, finite field gauge including atomic orbital (GIAO) based method is developed to address these challenges in both closed- and open-shell systems within a spin non-collinear framework. Nonperturbative effects of magnetic field and spin-orbit coupling are included variationally in a two-component method. With this method, the magnetic circular dichroism (MCD) spectra of closed shell system can be calculated with linear response, and the spin phase transition of transition metal system can be modeled in finite magnetic fields.

Roles of Hydration and Magnetism on the Structure of Ferrihydrite from First Principles

Michel Sassi and Kevin M. Rosso
Physical and Computational Sciences Directorate, Pacific Northwest National Laboratory, Richland, Washington

Ferrihydrite is one of the most common and essential mineralogic forms of ferric iron in the environment. However, subtleties of its atomic structure and the forces that dictate this structure or assemblage of structures remain unclear and are a matter of ongoing experimental focus. We report density functional theory calculations aimed at predicting thermodynamically stable structures for ferrihydrite across a range of possible compositions determined by the amount of structural water. Based on an assumed formula unit of Fe5O8H+nH2O, we performed ab initio calculations with evolutionary searching to find lowest enthalpy structures as a function of the water content up to n=2. This is the most exhaustive search for the ferrihydrite structure conducted so far; more than 5000 unique configurations were generated and evaluated over five states of hydration. Among them, the Michel akdalaite model was generated, along with several energetically comparable new structures at higher states of hydration. However, a direct comparison between calculated and experimental PDF and XRD patterns for the 50 lowest energy structures shows that none beyond the Michel model could be associated with ferrihydrite. Nevertheless, this energetically comparable structure set provides a novel basis for analyzing and understanding the effects of hydration and magnetism on the topology of ferrihydrite, from which we conclude that any tetrahedral Fe should be viewed as a metastable structural defect, created either as a result of the rapid kinetics of crystal growth or to accommodate a local magnetic stress between neighboring Fe atoms.



Rebecca Hanscam, Robert K. Szilagyi, Eric M. Shepard, Joan B. Broderick, and Valerie Copie
Department of Chemistry and Biochemistry, Montana State University, Bozeman, Montana

The [4Fe-4S]-maquettes are biomimetic compounds that encompass the most essential compositional and structural elements of metalloproteins that contain redox active [4Fe-4S] clusters. Experimental procedures for the reconstitution process of [4Fe-4S]-maquettes vary in the literature. Thus, there is a clear need for the development of quantitative structure/property relationships that can aid the optimization of the peptide sequence and reconstitution conditions for both oxidized and reduced cluster states. Moreover, the concept of [Fe-S] cluster nesting is a central theme in emergence-of-life theories, in which a critical step is linked to these clusters that bridges the inorganic FeS mineral-based catalysis of small-molecule activation to metalloenzymes in the Iron-Sulfur World.

Using force field-based (AMBER99SB) molecular dynamic simulations, we investigated the secondary structure of amino acids and peptides in explicit aqueous solution models with counter ions starting from a variety of initial peptide structures. During the simulations we followed the relative positions of the thiol groups of cysteine residues that can adopt arrangements poised to accept pre-assembled [Fe-S] clusters or smaller colloidal FeS nanoparticles. The presence of the latter has been established by detailed geochemical studies of low temperature, dissolved or electro-active FeS phases in aqueous solutions. The computational data sets (100k frames per peptide) were processed by an in-house developed secondary structure analysis code. The detailed description of the analysis toolkit is provided on a corresponding poster presentation presented at this meeting.

Based on the simulation results, we conclude that all peptides studied show similar secondary structure without notable variations as a function of amino acid sequence, in which
i. the presence of a preformed and stable [4Fe-4S] cluster nest is a rare event,
ii. 25-46% peptide conformations show the presence of [2Fe-2S] nests,
iii. the lifetime of the [2Fe-2S] peptide nest can be as long as 200 ps,
iv. 50 ns long simulations already provide meaningful statistics for secondary structure,and
v. peptide conformations do not match the biological dihedral angle distributions.

The molecular dynamics simulations imply that the tendency of aqueous peptides forming a [4Fe-4S] nest is rather different than that of the folded proteins with the “apo” cluster nest present due to the peptides experiencing a higher degrees of conformational flexibility.

The many-body expansion for aqueous systems revisited

Joseph P. Heindel and Sotiris S. Xantheas
University of Washington

We revisit the Many-Body Expansion (MBE) for water-water interactions by examining the effects of the basis set, including those resulting from the Basis Set Superposition Error (BSSE) correction, on the various terms for selected sizes of water clusters up to n = 21. The analysis is performed at the second order Møller-Plesset (MP2) perturbation theory with the family of augmented correlation consistent basis sets up to five zeta quality for the (H2O)n, n = 7, 10, 13, 16 and 21 clusters, for which we report either the complete MBE (n = 7, 10) or the one through the 6-body (n = 13) and the 5-body terms (n = 16, 21). Our results suggest that any sizeable contributions to the total cluster binding energy arising from the 5-body and larger terms are solely an artifact of the finite basis set. Indeed, all terms above the 4-body converge to practically zero at the Complete Basis Set (CBS) limit and this finding is accurately reproduced even with the smaller basis set of the series (aug-cc-pVDZ) once the BSSE correction is considered. The same level of theory (MP2/aug-cc-pVDZ, BSSE-corrected) also accurately reproduces the magnitude of the 3- and 4-body terms, for which we also find that the contributions of electron correlation are quite small. Our results unquestionably demonstrate that the MBE for water-water interactions vanishes monotonically with basis set size and can be safely truncated at the 4-body term once BSSE corrections are considered. We expect these findings to have important consequences in the pursuit of accurate many-body molecular dynamics simulations for aqueous systems.

Community analysis of molecular protrusions at liquid/liquid interfaces

Michael J. Servis and Aurora E. Clark
Washington State University

Surfactant-laden liquid/liquid interfaces are central to numerous industrially relevant processes. Understanding of the mechanisms of water transport from the aqueous surface to a nonpolar solvent, which result in surfactant-water adducts soluble in the nonpolar phase, are lacking and the subject of active investigation by studies using molecular dynamics simulations. These studies have noted the formation of “”protrusions”” or “”fingers”” of water extending into the organic phase as the surfactant pulls water from the aqueous surface. Forming and breaking these features represent a kinetically limiting process in the transport of water between phases. Understanding the role of the surfactant in their formation would facilitate design of efficient chemical processes. However, quantitative analysis of these features is limited if they cannot be algorithmically identified form simulation data. To achieve this, we employ the network analysis concept of communities. Communities are subgraphs which have high internal connectivity and sparse inter-community connectivity. Using modularity optimization, we partition a graphical representation of these molecules and their interactions such that protrusions form a community which can be isolated from structural and dynamic analysis. Using this methodology, the speciation and hydrogen bonding patterns of the protrusions are analyzed to identify water-surfactant structures that enable protrusion formation.

Computational observation of structure and solvation of the octanol-water interface

Zhu Liu and Aurora E. Clark
Department of Chemistry, Washington State University

Molecular dynamics simulation is used to study the prototypical water/1-octanol (denoted as octanol) liquid-liquid interface. What of particular interest is that an ordered bilayer octanol molecules is observed in the vicinity of the interface, where octanol molecules in the first layer present obvious orientation preference to point their hydroxyl headgroups toward the aqueous phase, while octanol in the second layer pointing in the opposite direction. This moderate bilayer preference is consistent with octanol’s density and end-to-end distance profiles herein but is rarely observed due to the significantly lower water miscibility in previous octanol-water interface simulations. 2-dimensional radial distribution functions provide a qualitative description of the octanol-octanol and octanol-water aggregates formation for octanol molecules within the bilayer. The occurence and morphology of these aggregates are investigated via analyzing the interfacial hydrogen bonding interactions at a molecular level. Numerical results will be presented for this specific interesting liquid-liquid interface system using the molecular dynamics simulation with the help of the state of art network theory focus on investigating pockets of enhanced water concentration and penetrating into the octanol phase due to the formation of the interfacial bilayer.


Ultrafast Intersystem Crossing from Exact Two-Component TD-DFT: Application to a Platinum Dimer Complex

Andrew J. S. Valentine
University of Washington

Spin-orbit coupling phenomena are important effects in quantum chemistry, particularly in the description of heavy elements. Among the most important of these is intersystem crossing (ISC), the spin-forbidden nonradiative transition between states of different multiplicities. ISC plays a critical role in photochemistry, where the rate of triplet formation will help determine the luminescent and phosphorescent properties of a material. Despite its importance, ISC is notoriously difficult to model computationally, as it requires both relativistic theory and accurate nuclear dynamics.

We present a novel method for calculating ISC rates using Exact Two-Component Time-Dependent Density Functional Theory (X2C-TDDFT). Unlike more common perturbative treatments of spin-orbit coupling, X2C is fully variational and includes relativistic effects at the molecular orbital level. By calculating excited states both with and without spin-orbit coupling, and mapping the spin-mixed states to spin-pure states, we are able to evaluate spin-orbit coupling matrix elements. We apply this approach to the study of a [Pt(ppy)(-tBu2pz)]2 complex, a challenging bimetallic system with very strong spin-orbit coupling. This platinum dimer complex exhibits ultrafast ISC, faster than the experimental resolution of 200 fs. We confirm the ultrafast ISC of the platinum dimer with an estimated rate of 25 fs, greatly shortening the window for triplet formation in this species.

Natural Transition Orbitals for Complex 2-Component TDDFT

Joseph M. Kasper and Xiaosong Li
University of Washington

The time-dependent Hartree-Fock (TDHF) and time-dependent density functional theory (TDDFT) equations are widely used as a first approximation to understanding electronic excited states. While the natural transition orbital method has allowed transitions to be viewed in a traditional orbital picture, the extension to multicomponent molecular orbitals such as those used in generalized Hartree-Fock (GHF) or generalized Kohn-Sham (GKS) is less straightforward due to mixing of spin-components and the inherent inclusion of spin-flip transitions in TDGHF/TDGKS. An extension of natural transition orbitals to the two-component framework is presented. Unlike the single-component analog, the method explicitly describes spin, which often yields multiple significant transition orbitals such as an equal mixture of α and β components in a singlet transition. The complex molecular orbitals can be plotted using mapped isosurfaces that display the magnitude and phase information. Calculations of a small ZnO semiconducting quantum dot illustrate the utility of the approach. Comparison with other techniques based on visualizing the density or density difference show that the approaches yield similar, though complimentary information.

Novel Phenomena in Quantum Thermodynamics: How to make Heat Flow from Cold to Hot

Phillip C. Lotshaw and Michael E. Kellman
Institute of Theoretical Science and Department of Chemistry and Biochemistry, University of Oregon

I will discuss three types of novel quantum thermodynamic behavior that are observed in models of system-environment thermalization and that might be observed in future studies of small molecular systems. First, I will discuss a quantum “entropy of the universe” that is used to formulate the second law for isolated system-environment “universes,” where the standard von Neumann entropy is zero. The new entropy gives standard results in a classical limit. Outside this limit, quantum time-dependence leads to “excess entropy production,” with consequences for thermodynamic behavior. Second, I will discuss a model for a small, variable-temperature environment that is similar to models for molecular vibrational motion. The finite size of the environment leads to non-standard analytical temperature behavior, which is manifested in simulations of system-environment thermalization. Finally, I will discuss simulations of heat exchange between two finite baths, where both finite size and quantum time-dependence are significant. Classically forbidden types of thermodynamic behavior are observed, including heat flow from cold to hot and an equilibrium state with different temperatures in the baths. These phenomena are found to be consistent with the new quantum entropy in the second law. These quantum thermodynamic phenomena might all be observed in future experimental and computational studies of small molecular systems.

Downfolding of many-body Hamiltonians using active-space models

Nicholas P. Bauman, Eric J. Bylaska,  Sriram Krishnamoorthy, Guang Hao Low, Nathan Wiebe, and Karol Kowalski
Pacific Northwest National Laboratory

Despite the grand effort in quantum chemistry and materials science communities to develop methods for describing complicated electron correlation effects, the applicability of these methods is still defined by a trade-off between accuracy and computational costs. Mathematically rigorous models where correlation effects are downfolded into a low-dimensional space offer a unique change to eliminate the inherent bias/biases of current many-body methods. In this presentation, I will discuss the extension of recently introduced the sub-system embedding sub-algebras coupled cluster (SES-CC) formalism [Kowalski, J. Chem. Phys. 148, 094104 (2018)] to unitary CC formalisms [Bauman, et al., arXiv preprint arXiv:1902.01553 and submitted to J. Chem. Phys.], which allows one to include the dynamical (outside the active space) correlation effects in an SES induced complete active space (CAS) effective Hamiltonian. In contrast to the standard single-reference SES-CC theory, the unitary CC approach results in a Hermitian form of the effective Hamiltonian. I will discuss a particular form of the unitary ansatz needed in order to accomplish this called the the double unitary CC formalism (DUCC) where the corresponding CAS eigenvalue problem provides a rigorous separation of external cluster amplitudes that describe dynamical correlation effects – used to define the effective Hamiltonian – from those corresponding to the internal (inside the active space) excitations that define the components of eigenvectors associated with the energy of the entire system. The proposed formalism can be viewed as an efficient way of downfolding many-electron Hamiltonian to the low-energy model represented by a particular choice of CAS. In principle, this technique can be extended to any type of complete active space representing an arbitrary energy window of a quantum system. The Hermitian character of low-dimensional effective Hamiltonians makes them an ideal target for several types of full configuration interaction (FCI) type eigensolvers, such as those used in quantum algorithms and DMRG methods. I will demonstrate the power and effectiveness of this downfolding technique with examples of energies for H2, H4, and Be systems obtained with the quantum phase estimator algorithm available in the Quantum Development Kit [Low, et al., arXiv preprint arXiv:1904.01131].


A Relativistic Two-Component Multireference Configuration Interaction Method

Hang Hu, Andrew J. Jenkins, Hongbin Liu, Joe M. Kasper, Michael J. Frisch, and Xiaosong Li
Molecular Engineering and Sciences, University of Washington

The description of the electron structure of heavy elements and transition metals is complicated by strong relativistic effects and strong electron correlations. We developed a relativistic multireference configuration interaction (MRCI) method as a post Complete Active Space Self-Consistent Field (CASSCF) procedure to capture the both the “dynamic” and “static” electron correlations. The relativistic effects, including scalar relativistic effects and one-electron spin-orbit coupling, are included here using an “exact two component” transformation of the solution of one-electron modified Dirac equation. A benchmark study of the fine structure of Sodium is presented. The effect of different correlation spaces on the description of the micro-states and the ‘d-line’ splitting is investigated.


Pradeep Gurunathan, Yen Bui, Lyudmila Slipchenko, Vanda Glezakou, and Roger Rousseau
Department of Chemistry, Purdue University, West Lafayette, Indiana
PCSD, Pacific Northwest National Laboratory, Richland ,Washington

The effective fragment potential (EFP) method is a computational approach designed to simulate accurate intermolecular interactions and solvent effects. Recently, we developed an extension of the EFP method to simulate non-covalent interactions in proteins, called BioEFP. In this approach, we make use of a divide-and-conquer approach to partition a biomolecule into smaller ‘effective fragments’. The parameters corresponding to these effective fragments are obtained using a set of preparatory ab-initio calculations.

The primary objective of rational drug design is to estimate the affinity of a molecule or a set of molecules to bind to a target both qualitatively and quantitatively. Molecular mechanics based force fields are considered the methods of choice for computing the drug-target binding by estimating the strength of intermolecular interactions, mainly due to the speeds that can be achieved in screening a large number of compounds as well as the ease in performing dynamics simulations. Such methods utilize a local model in which the interactions that are localized to the drug binding pocket are taken into account, while the interactions with the rest of the protein are often neglected. Here we propose an EFP-based energy decomposition analysis scheme for quantifying the drug-target interactions in proteins. We apply this scheme to understand the substituent effects in a -chloro vs -methyl aryl substitution in factor Xa inhibitor drugs. Our results indicate that the substituents are stabilized/destabilized due to interactions well beyond the binding pocket. While the primary contributor to the stability of the substituent functional group is electrostatic interactions, other terms such as exchange-repulsion play an important role as well. The extended contact model presented here accounts for interactions that are typically neglected in a local model and marks an improvement over the latter for reliable predictions of interaction energies. This also opens up possibilities for utilization of the EFP scheme to applications in bioproducts separations and computational catalysis.

Characterizing screening methods for solar thermal fuel properties of norbornadiene

Reuben Szabo, Khoa Le, and Tim Kowalczyk
Western Washington University

The norbornadiene-quadricyclane (NBD-QC) photoisomer system makes for a promising solar thermal fuel, but its absorption energy needs to be red-shifted by adding electron donating and withdrawing groups. However, in decreasing the absorption energy of norbornadiene compounds, we do not want to sacrifice the half-life, storage energy density, or photoisomerization quantum yield. With the aim of screening vast libraries of potential substituents, the suitability of density functional tight binding (DFTB) for characterizing solar thermal fuel properties of modified NBD-QC systems is herein evaluated. In addition, DFT methods are used to characterize transition states of the modified systems and estimate the reverse isomerization energy barrier and quantum yield.


Eric Beyerle and Marina Guenza
University of Oregon

Protein dynamics occurs over a range of time- and lengthscales. The Guenza group has developed an analytical method, the Langevin Equation for Protein Dynamics (LE4PD), that decomposes a protein’s motion into a set of linearly-independent normal modes of motion. Each mode describes collective fluctuations of the protein’s alpha-carbons over a characteristic time- and lengthscale. The modes predicted by the LE4PD can be visualized as diffusing over a free-energy landscape that possesses characteristic barriers to transport for each normal mode, and the kinetics on these free energy surfaces can be described using Markov state models (MSMs). We are able to construct informative MSMs for the slowest LE4PD internal modes, and we show that using an MSM to model barrier-crossing kinetics for the slowest, highest amplitude LE4PD modes improves the ability of the LE4PD to predict bond-bond autocorrelation functions, calculated from an equilibrium MD simulation, in flexible binding regions of the ubiquitin protein . These results demonstrate the importance of accurately describing barrier crossing events when modeling protein dynamics.