2025 Kansas Physical Chemistry Symposium

Gas Absorption into CO₂-Expanded Ionic Liquids

Rund Al-kofahi, Micah Welsch, and Brian B. Laird 

Department of Chemistry, University of Kansas

Ionic liquids represent environmentally friendly alternatives to traditional organic solvents due to their extremely low vapor pressure, which reduces the risk of inhalation exposure and air pollution. In this study, Gibbs Ensemble Monte Carlo (GEMC) molecular simulation is used to investigate the absorption of alkanes into the CO₂-expanded ionic liquid 1-butyl-3-methylimidazolium (bmim) PF₆. The coexistence curve for CO₂ is being calculated using GEMCsimulation to validate the simulation approach and ensure its accuracy. Furthermore, the model for bmim PF₆ is being built using Visual Molecular Dynamics. Currently, there are challenges in both the GEMC simulation for CO₂ and modeling of the ionic liquid. Once these issues are resolved, the absorption of CO₂ into bmim PF₆ will be measured using GEMC simulation, followed by the absorption of various alkanes into the expanded ionic liquid using the same method. This research aims to provide valuable insights into the solubility of gases in CO₂-expanded ionic liquids, thereby contributing to a deeper understanding of these systems and their potential applications as solvents for industrially relevant catalytic reactions.

Charged and Uncharged Dyes in Nanoporous Anodic Aluminum Oxide Membranes: Fluorescence Correlation Spectroscopy Studies of Nanoconfined Ethanol/Water Mixtures

Kumar Amgain, Takashi Ito, and Daniel A. Higgins

Department of Chemistry, Kansas State University

The diffusion of charged rhodamine B (RhB) and uncharged perylenediimide (PDI) dyes confined within nanoporous anodic aluminum oxide (AAO) membranes was investigated by fluorescence correlation spectroscopy (FCS). The AAO nanopores were filled with a series of water-ethanol mixtures of high ethanol content, near the composition of the bulk azeotrope. This study explores how nanoconfinement of the dyes as model solutes modulates their diffusive motions and interactions with the nanopore surfaces. Autocorrelation data obtained were fit to a model for two-component one-dimensional diffusion, revealing the occurrence of both fast and slow diffusion processes characterized by diffusion coefficients D_f and D_s, respectively, within the membrane nanopores. Both diffusion coefficients for both dyes were found to depend on ethanol-water mixture composition. The D_f values generally followed the viscosity-dependent trend diffusion in bulk solution for both dyes. In contrast, their D_s values exhibited different dependences on mixture composition. D_s for the charged RhB dye was found to increase with increasing water content in the mixture, while D_s for uncharged PDI decreased across the same series of mixtures. The former behavior is attributed to solvation energy effects that preclude strong RhB-pore interactions in intermediate water-ethanol mixtures, while the latter is assigned to enhanced hydrophobic effects in mixtures of higher water content, causing the dye to interact more strongly with the pore surface.

Nanoconfinement Effects on Solute Diffusion in Nanoporous Anodic Aluminum Oxide Membranes: A Study with n-Butanol/Water Mixtures Using Fluorescence Correlation Spectroscopy

Safoura Asadi, Hamid Rashidi, Takashi Ito, and Daniel A. Higgins 

Department of Chemistry, Kansas State University

This study examines nanoconfinement effects on solute diffusion within anodic aluminum oxide (AAO) membranes, using n-butanol/water mixtures as a model system. Fluorescence correlation spectroscopy (FCS) was employed to study the transport behavior of Rhodamine B (RhB) dye, offering insights into molecular diffusion in nanoporous environments. AAO membranes with pore diameters of 10 nm and 20 nm were analyzed using both a single-component anomalous one-dimensional (1D) diffusion model and a two-component 1D diffusion model. The two-component model more accurately described diffusion behavior, revealing two distinct components: fast diffusion characterized by diffusion coefficient D_f and slow diffusion characterized by D_s. Both were influenced by nanopore size.

Both D_f and D_s showed a clear dependence on solvent composition, with higher water content generally leading to increased diffusion rates, counter to expectations from mixture viscosity. Instead, the trend aligns with solubility measurements obtained via gravimetric analysis. However, while D_f consistently increased with water content, D_s showed a notable drop at 20% water fraction, suggesting possible phase separation under nanoconfinement. To investigate this further, temperature-dependent diffusion studies are now underway.

These findings highlight the role of solvent composition and pore size in governing molecular transport, with implications for improving membrane-based separation technologies in applications such as biofuel purification, water treatment, and chemical processing through betterunderstanding of diffusion behavior in confined systems.

Structural Dependency of Surfactants for Solubilization of Aliphatic Moieties in an Aqueous/Ethylene Glycol Solvent System 

Matthew A. Baird,¹ Dustin Willhite,¹ and Doug English² 

¹BG Products, Wichita, Kansas, and ²Department of Chemistry and Biochemistry, Wichita State University 

Surfactants (surface active agents) reduce the surface tension of a liquid which allows it to spread and wet surfaces easily and enables water and oil to mix. As the concentration of a surfactant increases in a hydrophilic solution the surfactants aggregate and self-assemble intomicelles, which consist of a hydrophobic core and a hydrophilic surface. The hydrophobic core of a micelle can solubilize hydrophobic molecules in an otherwise aqueous environment. Here we explore the structural dependency and efficacy of the hydrophobic tail group to solubilize aliphatic moieties in an aqueous/ethylene glycol solvent system.

Aliphatic carboxylic acids as coolant corrosion inhibitors were first patented in the US in 1945, this paved the way for modern coolant chemistry. Due to the hydrophobic effect, aliphatic carboxylic acids over five carbon atoms are difficult to dissolve in an aqueous system. Which effectively limits the number of aliphatic carboxylic acids available as corrosion inhibitors. Successful stabilization of long chain aliphatic moieties in an aqueous ethylene glycol solvent system would open the door to new types of corrosion inhibitors. Ten commercially available surfactants of varying tail chain lengths were obtained and screened against six aliphatic moieties. The effects of critical micelle concentration, aromaticity, branching, and chain length were measured. The stability of solvent system was measured between -7°C and 88°C. Solubility was observed to be structurally dependent on the tail group of the surfactant; suggesting that the environment of the micelle core is important for the solubility of the aliphatic species of interest.

Exploring the Dynamics of Supercooled Water through the Temperature Dependence of Activation Energies

Elizabeth R. Bartlett and Ward H. Thompson 

Department of Chemistry, University of Kansas

The anomalous properties of supercooled water have frequently been attributed to a high-pressure phase transition between a high-density liquid (HDL) and a low-density liquid (LDL), which terminates at a liquid-liquid critical point. The extension of this coexistence curve to ambient pressures represents the Widom line. The temperature- dependent behavior of supercooled water at ambient pressure can show a transition between more HDL-like and more LDL-like as the Widom line is crossed. Hestand and Skinner, in their 2018 perspective, propose a temperature dependence of the diffusion coefficient that is based on a transition from HDL behavior to low-density liquid LDL behavior in a so-called “two liquids” picture. We attempt to fit the parameters of this description with molecular dynamics simulations of TIP4P/2005 water and compare to experimental results. We also explore the behavior of water reorientational dynamics at supercooled temperatures through the timescale associated with hydrogen bond exchanges, τ₀.

Silanol Functionalization of Amorphous Silica Surface

Shreyaa Brahmachari and Marco Caricato 

Department of Chemistry, University of Kansas

Inspite of its wide use as support in heterogeneous catalysis, accurate atomically detailed models of amorphous silicates have been elusive to-date. Amorphous silica, which is composed of SiO₄ tetrahedral units in the form of siloxane rings of different sizes, shows local structural disorder and surface heterogeneity that play a key role in its reactivity. Although it is difficult to identify and understand the distribution of catalytically active sites and their structures from experimental characterization, density functional theory (DFT) simulations allow us to probe the properties of the individual sites. In this contribution, cluster models of amorphous silica are used to study the thermodynamics of the functionalization reaction on the silica surface. We investigate the variation of functionalization energy with the structural parameters of the unfunctionalized site, which are defined as descriptors. We use the random forests regression model to determine the most important descriptor(s) that correlates with the functionalization energy. Considering the different combinations, a two-descriptor set composed of r^L (longest Si-O bond at the site) and θ^P_diff (difference between the smallest and largest angles of the unfunctionalized ring) provides the lowest average root mean square error (RMSE), which is still 30% of the average functionalization energy in magnitude. While increasing the size of the data set should improve the machine learning fitting and reduce the RMSE, modelling large clusters is computationally expensive. Thus, for a quantitative simulation of the reaction we attempt to produce minimal models of the site of functionalization that effectively mimic the results of the large clusters. We truncate (and cap the dangling bonds with F atoms) or relax (and freeze the surroundings) the original large clusters. Repeating the calculations by increasing the radii of truncation (or relaxation) of the clusters from the functionalization site, it is seen that the energy changes with increase in size of the clusters until it converges. These results show that not only the local coordination environment but the bulk of amorphous silica significantly affect the functionalization energy.

Broadband Two-Photon Absorption (BB-2PA)

Chase Courbot and Christopher Elles 

Department of Chemistry, University of Kansas

Broadband two-photon absorption (BB-2PA) spectroscopy is a powerful but underutilized technique for measuring the nonlinear optical properties of materials. Most 2PA measurements use a single-wavelength laser beam to obtain the 2PA cross section at a single excitation energy, which requires tuning the laser to obtain a full spectrum and often results in systematic error due to variation of the laser focal conditions with wavelength. These approaches also rely on indirect measurements of 2PA, like fluorescence detection or attenuation of the transmitted beam to quantify the 2PA strength of the sample. We demonstrate a BB-2PA measurement by utilizing a broadband pump-probe technique to measure 2PA spectra directly and provide wide spectral coverage. Our BB-2PA technique utilizes stimulated Raman scattering (SRS) from the solvent as an internal standard. The internal standard eliminates uncertainty related to the overlap of pump and probe pulses, which is typically the largest source of error in 2PA measurements. This enables rapid data acquisition and increases the accuracy of absolute cross-sections compared to other approaches.

Probing Electronic Metal-Support Interactions of Cu/TiO₂ (110) 

N. Dissanayake, L.N. Penland, H. H. Hirushan and R. G. Farber 

Department of Chemistry, University of Kansas

Electronic metal-support interactions (EMSI) play a crucial role in determining the chemical selectivity and efficiency of oxide-supported metal nanoparticle catalysts. The non- oxidative dehydrogenation of ethanol to acetaldehyde over Cu/TiO₂ catalysts is known to be sensitive to EMSI. It has been shown that the Lewis acidity of the oxide support, support defect density, and Cu nanoparticle size can tune the EMSI of Cu/TiO₂ catalysts.

Using Cu/TiO₂(110) as a representative catalyst, we aim to develop a spatially resolved understanding of the structural and compositional features influencing EMSI of Cu/TiO₂ catalysts. A pristine TiO₂(110) surface was prepared through repeated cycles of Ar+ sputtering and annealing under ultra-high vacuum (UHV) conditions, resulting in the observation of the characteristic (1 ́1) and (1 ́2) surface reconstruction. Cu was then deposited onto the reconstructed TiO₂(110) surface resulting in Cu nanoparticles. The structural and electronic properties of both the pristine and Cu/TiO2(110) surfaces were characterized using Auger electron spectroscopy (AES), low energy electron diffraction (LEED), scanning tunneling microscopy (STM), and scanning tunneling spectroscopy (STS). These complementary techniques provide insights into the evolution of surface structure and local density of states (LDOS) upon Cu deposition. Our findings contribute to a spatially resolved understanding of how surface structure and nano-particle formation modulate EMSI at the atomic scale, providing a foundation for future studies investigating the structural, compositional, and electronic features guiding selective ethanol dehydrogenation over Cu/TiO₂ catalyst surface.

First Principles Simulations of Optical Rotation of Chiral Molecular Crystals

Emmanuel Forson, Taylor Parsons, and Marco Caricato 

Department of Chemistry, University of Kansas

In this work, we present simulations of the optical rotation (OR) for five molecular crystals at density functional theory level with periodic boundary conditions (DFT- PBC). Calculations are compared with experimental measurements and show semi- quantitative agreement with experimental data for three of the crystals: tartatic acid, benzil, and pentaerythritol. For the other two crystals, aspartic acid and glutamic acid, the experimental data are two orders of magnitude larger than the calculated data. We provide some arguments that support the theoretical predictions and suggest that the experiments should be revisited. We also find that the position of H centers provided in experimental X-ray data is not sufficiently reliable for simulating OR, and better results are obtained when H atoms are allowed to relax while keeping heavier elements fixed at the experimental positions. Comparison with molecular cluster calculations with a better functional and a larger basis set indicate that the role of intermolecular interactions (reproduced with the PBC technique) are as or more important than the choice of model chemistry. Despite the current limitations in the level of theory that can be employed, these simulations provide a promising avenue to investigate the effect of intermolecular interactions on this sensitive electronic property of molecules and materials.

The Computational Design of Peptides Derived from E-cadherin to Modulate the Blood Brain Barrier                                                                           

Jinyan He,¹ Krzysztof Kuczera,^1,2 and Teruna J. Siahaan³ 

¹Department of Chemistry, University of Kansas, ²Department of Molecular Biosciences, University of Kansas, ³Department of Pharmaceutical Chemistry - School of Pharmacy, University of Kansas

 The blood-brain barrier (BBB) is a selective barrier which allows nutrients to enter the brain but blocks pathogens and toxins from entering. At the same time, BBB also blocks brain entry of therapeutics used to treat brain diseases. Previously, HAV and ADT peptides, identified in E-cadherin crystal contacts, proved to be able to successfully modulate the BBB permeability both in vivo and in vitro. Here, we use computational methods to perform a more exhaustive and systematic search for novel peptides which were derived from E-cadherin protein and can effectively interfere with E-cadherin interactions and thus modulate the permeability of the BBB. In the first stage, computational protein-protein docking was employed to explore possible interactions between the first two domains of human E-cadherin (EC12). Promising peptide candidates were proposed by analyzing the explored interaction at docking interfaces. Next, the discovered peptides were re-docked to E- cadherin, and the binding modes and affinities of peptide-protein complexes were analyzed. Using different protein-peptide docking methods, peptides identified with strong binding affinity to EC12 were selected for future experimental validation and further sequence optimization. One of the proposed peptide candidates was selected, synthesized and its binding affinity to recombinant EC1 domain verified with Surface Plasmon Resonance.  Overall, our results present a systematic approach for generating novel peptides with high potential for disrupting the BBB and thus enabling drug delivery to the brain.  

Structural and Electronic Properties of Benzyl Isothiocyanate Self-Assembled Monolayers

H. H. Hirushan, Lindsey N. Penland, N. Dissanayake, Darya Moiny, Lily Tackett, Dmitry Ovchinnikov, Qunfei Zhou, and Rachael G. Farber

University of Kansas, Department of Chemistry

There is an intrinsic relationship between molecular structure and the chemical, physical, and electronic properties of self-assembled monolayers (SAMs). Benzyl isothiocyanate (BITC), an aromatic thiolate of biological importance, is commonly used in nano-emulsions, nanoencapsulation, and nanoparticle-based drug delivery applications. The coexistence of stabilizing π-π interactions and the destabilizing electron-withdrawing cyanide group is expected to influence the packing and formation of BITC SAMs. However, despite its applications in confined and condensed environments, the molecular arrangement of BITC SAMs on Au surfaces remains largely unexplored.

Using ambient tapping mode atomic force microscopy (AFM), it was determined that an ordered BITC SAM forms on Au(111)/mica following a 65 hour incubation in a 20 mM ethanolic BITC solution. This was confirmed by AFM which revealed a linear pattern across the Au(111) surface, indicative of an ordered BITC SAM. Once optimal incubation conditions were identified, scanning tunneling microscopy and spectroscopy (STM/STS) were used to determine the molecularly resolved packing structure of BITC. Ongoing computational studies aim to determine the molecular unit cell and projected density of states of BITC SAMs, providing physical insight for experimental observations.

Advanced Insights into Non-Adiabatic Dynamics and Ring-Opening Mechanisms of Oxazole and Isoxazole

Paul Javed,¹ Briony Downes-ward,³ Arthur G. Suits,³ Daniel Rolles,² and Christine M. Aikens¹ 

¹Department of Chemistry, Kansas State University, ²Department of Physics, Kansas State University, ³Department of Chemistry, University of Missouri

This study aims to investigate the photoinduced ring-opening dynamics in oxazole and isoxazole using ab initio non-adiabatic molecular dynamics simulations. Employing the State-Averaged CASSCF/aug-cc-pVDZ level of theory, we explored the excited-state dynamics and ring-opening mechanisms of these heterocyclic compounds. Initial conditions were generated using Wigner sampling, and trajectories were propagated for up to 1,000 fs. Our results reveal distinct photochemical behaviors for oxazole and isoxazole. In isoxazole, we observed O–N bond breaking within an average time of 41.46 fs, predominantly through the S2 → S1 → S0 relaxation pathway. For oxazole, although complete bond breaking was not observed within the simulation timeframe, significant C–O bond elongation with substantial geometric changes and ring puckering were noted, particularly at S1/S0 crossing points. We analyzed vertical excitation energies, simulated IR spectra, and tracked internal coordinates throughout the trajectory. We compared the results from our simulations with experimental ultrafast electron diffraction (UED) patterns and found good agreement between the two. These findings provide detailed insights into population dynamics, quantum amplitudes, and branching ratios for various photochemical pathways in both molecules. This work enhances our understanding of the fundamental photochemistry of five-membered heterocycles, which has significant implications for the design of photoactive materials and photochemical synthesis strategies, with potential applications in the development of advanced organic photovoltaics. By combining ab initio simulations with machine learning, this research opens new avenues for innovation in materials science and photochemistry.

Self-Assembly and Organization of Gold Binding Peptides: AuBP1 and AuBP2

Chris Johnson,¹ Hashitha Tharakee,¹ Nathalie Moro,¹ Trish Nair,¹ Atrooba Hashim,¹ Taylor Bader,^2,4 Candan Tamerler,^2,3,4 and Cindy L. Berrie¹

¹Department of Chemistry, University of Kansas, ²Bioengineering Program, University of Kansas, ³Department of Mechanical Engineering, University of Kansas, ⁴Institute for Bioengineering Research, University of Kansas

Solid binding peptides show significant potential for biohybrid materials technologies due to their ability to bind selectively to a specific target material. Their applications includebiocatalysis, biosensing, and agriculture. For the utilization of such materials, it is critical to understand the optimal conditions for these peptides and to further the fundamental knowledge base for controlling their binding and organization. Herein, the self-assembly of two peptides, AuBP1 (WAGAKRLVLRRE) and AuBP2 (WALRRSIRRQSY), were analyzed through atomic force microscopy (AFM). AFM studies were conducted on Gold, Graphite, octadecyltrichlorosilane (OTS) on Si, and Mica surfaces to gain insight into the assembly process as well as provide a framework for how materials can be designed to be used with such peptides. Peptide film coverage on the surfaces were monitored as a function of adsorption time at 100nM peptide concentration. pH and ionic strength studies were performed with AuBP1 on gold to understand how peptide binding is a function of the solution conditions. Our observations show a complexassembly process involving both peptide-peptide and peptide-surface interactions. Understanding the conditions that lead to robust self-assembled peptide films could extend the utilization of materials binding peptides into future technologies.

Calculation of the Surface Free Energy of a Lennard-Jones Fluid at a Hard Wall by Monte Carlo Simulation 
 

Jonah Ludiker and Brian B. Laird 

Department of Chemistry, University of Kansas

Despite the ubiquity of the solid liquid interface in nearly every field of chemistry, there are still major limitations in the theoretical descriptions. As both liquid and solid phases are condensed, the interface is very difficult to study experimentally. Surface free energy is an interfacial thermodynamic quantity that governs many phenomena of scientific and technical interest, such as wetting, crystal growth, dendrite morphology, and nanoparticle stability. In our work, we seek to improve connections between the surface free energy and intermolecular forces. The effects of repulsive forces on the surface free energy in interfacial systems are well studied, but the understanding of the attractive inter-molecular interactions role in surface thermodynamics is less understood. Through analysis of simple model systems, we aim to break down the complex surface-fluid interactions into simpler component problems which can be solved independently. Density profiles of Monte Carlo simulations of Lennard-Jones fluids at repulsive walls were sampled, the interfacial excess volume was calculated, and the surface free energy is calculated with Gibbs-Cahn integration. The results are compared to theoretical predictions.

Study of Transport Properties within Low-dimensional Quantum Materials via Fabrication of Magnetic Tunnel Junctions

Darel Pates, Grant Saggars, Abishai Mathai, Dmitry Ovchinnikov, Alex Guardiola, Ryan Luke, Salman Ahsanullah, and Vivek Jain 

Department of Physics and Astronomy, University of Kansas

Recent developments in two-dimensional (2D) materials have led to a growth in applications and understanding of the underlying physics in lower dimensional magnetism. Heterostructures known as magnetic tunnel junctions have been a special interest due to their flexibility in examining many material properties from spin injection efficiency, magnetoresistance, and spin resolved transmission. In this talk I will showcase our laboratory’s work in fabrication, measuring and testing of Magnetic Tunnel Junctions as well as highlighting potential applications and what we can learn about the underlying physics from these experiments.

Investigation of the Structure of Gold-binding Peptides in Solution, in the Solid state, and Bound to Gold using ATR-FTIR

Ruvini O. Rodrigo,¹ Emily Gordon,^1,2 Chris Johnson,¹ Taylor Bader,^3,4,5 Candan Tamerler,^3,4,5 and Cindy L. Berrie¹

¹Department of Chemistry, University of Kansas, ²Department of Chemistry and Biochemistry, Colorado College, ³Mechanical Engineering, University of Kansas, ⁴Bioengineering Program, University of Kansas, ⁵Institute of Bioengineering Research, University of Kansas

Biosensors often employ enzymes bound to solid support for the sensitive, selective detection of bioanalytes. However, the immobilization of multiple enzymes in an organized way to achieve coupled enzyme activity is a significant challenge, and methods are needed to selectively bind the different enzymes to specific locations on the surface. One approach for achieving this selective immobilization is the use of material-specific peptides, which are selected for their high binding affinity, selectivity, and specificity for binding to a particular material. Herein, we have used Attenuated Total Reflection Fourier Transform Infrared (ATR-FTIR) spectroscopy to investigate model gold-binding peptides (AuBP1 and AuBP2) under a variety of different conditions. Specifically, significant changes were observed in the shape of the amide I band when in solution, in the solid state, and when bound to gold nanoparticles. The structural differences responsible for the changes in the amide I band have been quantitatively assessed by standard curve fitting and large changes in the relative fraction of alpha helix, beta sheet, and random coil were observed. Overall, we show that these peptides bind differently to gold than they do in solution or deposited on a surface. These results help to understand the organization of enzymes bound to gold through gold-binding peptides in an ordered array, giving insight into how they may be organized for coupled enzyme activity in biosensors.

Fabrication and Characterization of High PMA and TC Spin-Orbit Torque Heterostructures – Fe₃GaTe₂ and WTe₂

Grant Saggars, Abishai Mathai, Darel Pates, Dmitry Ovchinnikov, Ryan Luke, Salman Ahsanullah, and Vivek Jain

Department of Physics and Astronomy, University of Kansas

The key to future, high-density, and energy-efficient spintronics for memory and logic are held by two-dimensional (2-D) van der Waals (vdW) materials. Recent breakthroughs have led to interest in materials that are promising enough for commercial integration; however, the need for room temperature strong ferromagnetism and perpendicular magnetic anisotropy (PMA) has not fully been realized. Recent discovery of these properties within WTe₂/Fe₃GaTe₂ interfaces1 may pave the way to fabricate dense, stable, and efficient memory and logic devices. This project is dedicated to the development of heterostructures based on topological insulators such as WTe2 that possess high spin-orbit coupling and 2D magnets with high critical temperatures such as Fe₃GaTe₂ and Fe₅GeTe₂. These heterostructures will be further employed for the demonstration of magnetic-field free switching of magnetic layers using Spin-Orbit Torque (SOT).

Single-Molecule Investigation of Rhodamine B (RhB) Adsorption and Diffusion on Fresh and Aged Polyethylene Terephthalate (PET) Plastics

Sorour Salehi and Daniel A. Higgins 

Department of Chemistry, Kansas State University

Microplastics (MPs) are a serious environmental threat due to their ability to adsorb and transport organic micropollutants (OMs). We employed advanced fluorescence-based single-molecule detection and tracking techniques to better understand the molecular mechanisms governing OM-MP interactions. By tracking individual Rhodamine B (RhB) dye molecules as a model for OMs, we directly examined their adsorption, desorption, and diffusion behaviors on the surfaces of thin plastic films. We studied polyethylene terephthalate (PET) films as models for PET MPs. We compared fresh samples to those that had been artificially aged in a UV-ozone chamber. Spectroscopic ellipsometry showed that aging reduced the PET film thickness, and water contact angle measurements indicated increased hydrophilicity with aging time. Super-resolved single-molecule tracking using wide-field microscopy in total internal reflection mode demonstrated significant effects of aging on RhB-PET interactions. Specifically, it was discovered that RhB accumulates on the PET surface by nonspecific hydrophobic interactions that allow the molecules to remain mobile and explore the plastic surface. When the molecules encounter reactive sites, they become strongly adsorbed and immobilized. Molecular diffusion analyses indicate that aging reduces the hydrophobic interactions and molecular mobility, as evidenced by changes in the dye diffusion coefficient, while adsorption to reactive sites continues to be observed. Our findings have implications for developing plastics engineered to minimize hazardous OM accumulation and may inform targeted remediation efforts for specific OM-MP combinations of environmental concern.

A Software Package for Performing Fluctuation Theory Calculations

Ibrahim Sufi,^1,2 Anjali Radhakrishnan,¹ Elizabeth R. Bartlett,¹ and Ward H. Thompson¹

¹Department of Chemistry, University of Kansas, ²Department of Electrical Engineering & Computer Science, University of Kansas

​The dynamics of many real-world chemical systems depend on temperature. For example, the activation energy of a diffusion process can be thought of as the derivative of the diffusion coefficient with respect to the inverse temperature. Typically, these derivatives are calculated via Arrhenius analysis by running experiments or simulations at multiple temperatures. However, Arrhenius analysis fails for many systems of interest because it requires that the dynamics do not change substantially in a broad enough temperature range to calculate the derivative.

​Fluctuation theory solves this problem as it allows calculation of the temperature derivative of any property from simulations at a single temperature. The key insight is to sample the distributions of energies by running an NVT ensemble simulation and then use those sampled configurations to calculate the dynamical properties by running an NVE ensemble simulation. We introduce a software package that automates much of this workflow, allowing users to simply input their Lammps input files that they would use for an NVT ensemble simulation as well as their analysis code. The package then automates the rest of the workflow and calculates the temperature derivative of said property for their system. The package is specifically designed for parallel use across slurm clusters allowing the many NVEs to be run in parallel savingcomputation time.

Surfactant-Vesicle Based Analysis of Inhibitors for Multivalent Lectin/Carbohydrate Binding

Carson Tucker, Kassidy Adams, Leo Hays, David Marck, and Douglas S. English, 

Department of Chemistry and Biochemistry, Wichita State University

We have developed a robust and inexpensive method for high throughput screening of inhibitors for the carbohydrate binding protein concanavalin-A (Con A). Our method uses inexpensive, easily prepared surfactant-based vesicles. These vesicles are highly stable with a long shelf life and are amenable for use in high-throughput 96-well screening assays. Compared to other methods such as surface plasmon resonance (SPR) which require chemical modification of gold substrates, our method uses a free-floating fluid bilayer that more closely matches a biological cell membrane. We have demonstrated our method using a variety of soluble inhibitors against Con A, and compared our results of previous studies using SPR. We are now exploring this method to evaluate carbohydrate mimicking peptides that could provide building blocks for easily synthesized multivalent inhibitors.

Exciton Dynamics in Non-fullerene Acceptor Materials for Photovoltaic Applications

Elizabeth Udeh, Kushal Rijal, Neno Fuller, Fatima Lariz, and Wai-Lun Chan, 

Department of Physics and Astronomy, University of Kansas

Non-fullerene acceptors (NFAs) are gaining increasing attention recently because organic photovoltaics (OPVs) made with NFAs significantly outperform those made with traditional fullerene acceptors. Despite its outstanding performance, it remains unclear how NFAs can enable free-carrier generation from bound-excitons to occur with a much lower energy loss as compared to fullerene acceptors. This work is focused on investigating the exciton dynamics in bulk heterojunctions made with PM6 donor polymer and Y6 NFA as a function of temperature. Time- and frequency- resolved photoluminescence spectroscopy is used to probe both the dynamics and energy relaxation of excitons. We found that significant energy relaxation occurs at a very low temperature (~ 77 K), but this relaxation process is absence at higher temperatures (> 150 K). This work is supported by US DOE under Award Number DE-SC0024525.

Arrayed Gold and Silver Nanostructures Designed as Cascade Enzyme Scaffold

Nimodi D. Uduwela,¹ Chris Johnson,¹ Taylor Bader,^3,4,5 Candan Tamerler,^3,4,5 and Cindy L. Berrie¹

¹Department of Chemistry, University of Kansas, ²Department of Chemistry and Biochemistry, Colorado College, ³Mechanical Engineering, University of Kansas, ⁴Bioengineering Program, University of Kansas, 5Institute of Bioengineering Research, University of Kansas

Cascade enzyme systems show great promise as both biocatalysis and biosensors. Noval cascade enzyme systems structure themselves with protein or DNA back bones to achieve the high spatial organization requires efficient systems. Such systems are limited by biological stability, as such we are interested in the design of arrayed multi material structures to be utilized with bioengineered peptide to fabricate highly stable cascade enzyme systems. Our approach for the fabrication of the nanostructure arrays utilized thermal evaporation with a particle lithograph mask, and electroless deposition with a particle lithography templated resist layer. Silica and polystyrene beads are being explored for lithographic mask formation to optimize packing efficiency and pattern reproducibility. The nanostructures are characterized using atomic force microscopy (AFM) and localized surface plasmon resonance (LSPR). Future work will focus on combining the two fabrication methods to create bimetallic nanostructure arrays. These integrated platforms are expected to enable the precise self-assembly of multi-enzyme systems, offering a promising route for the development of advanced biosensors and catalytic devices.

Water Diffusion Around a Hexameric Resorcin[4]arene Assembly: A Molecular Dynamics Investigation of Energetic Barriers and Intermolecular Interactions

Nnamdi Uzoukwu, Anjali Radhakrishnan, and Ward H. Thompson, Department of Chemistry, University of Kansas

The forces that drive the diffusion of water in a hexameric resorcin[4]arene assembly formed in wet chloroform are examined by molecular dynamics simulations to better understand the role of water in the dynamics of this system. Water diffusion coefficients, the corresponding activation energies, and the contribution to the activation energies from the intermolecular forces are calculated. A significant reduction in water diffusion is observed in this system, with a 33% decrease in diffusion coefficient relative to bulk water. This slowdown is attributed to an approximate 80% increase in the activation energy for diffusion, indicating a substantial energetic barrier to water mobility in this environment. The decomposition of the activation energy reveals that the majority of the activation energy comes from electrostatic interactions between water molecules and the supramolecular assembly. Furthermore, the Lennard-Jones interaction energy between water and the assembly is found to be significant, suggesting a more structured solvation shell around the assembly. These findings underscore the critical role of water in the dynamics of a resorcin[4]arene and imply that understanding its diffusion mechanism is essential for elucidating the guest-exchange process that occurs within the assembly.

Dependence of Water Adsorption on Aluminum Content in the BEA Polymorph A Zeolite using Monte Carlo Simulation

Micah Welsch and Brian B. Laird 

Department of Chemistry, University of Kansas

Zeolites are microporous aluminosilicate materials with numerous industrial and commercial applications whose versatility lie in their highly varied and modifiable framework structures. In this work, the effect of Al mole percent on TIP5P-Ew water adsorption into an H+ beta polymorph A [BEA(A)] zeolite as a representative framework was investigated. The effective water adsorption resulting from Al content has been previously studied in various zeolites but typically only with a focus on select adsorption characteristics or without holding other compositional or topological variables constant. This work aims to isolate the effect of varying Al content on water adsorption by maintaining these variables and providing a comprehensive analysis of adsorption properties. Water vapor adsorption isotherms, isosteric heats of adsorption, hydrogen bond geometries, and water oxygen – aluminum radial distribution functions (RDFs) were calculated and compared across several Al mole percents for the BEA(A) zeolite. Gibbs-Ensemble Monte Carlo (GEMC) and Grand Canonical Monte Carlo (GCMC) simulation were utilized for bulk water and adsorption calculations, respectively. It was found that although adsorption rapidly increases with rising Al content, the heats of adsorption and strength of the hydrogen bond network were observed to plateau at moderate Al compositions, suggesting an upper limit to the improvement of water affinity when increasing Al content alone. Obtaining a deeper understanding of how water adsorption into zeolites varies with Al mole percent is vital in the optimization of structures for many current and future applications of this important class of materials.

Charge Separation at Organic Interfaces with Near-Zero Energy Offset – A Step Towards Designing Inorganic-like Organic Semiconductors

Neno Fuller, Kushal Rijal, Elizabeth Udeh, and Wai-Lun Chan

Department of Physics and Astronomy, University of Kansas

For organic semiconductors, it is often presumed that the energy level offset at the donor/acceptor (D/A) interface provides the necessary driving force for charge separation (CS). In this work, by using zinc phthalocyanine (ZnPc) and fluorinated zinc phthalocyanine (F₄ZnPc) as a model D/A interface, we found that spontaneous CS, with an enthalpy increase of ~ 0.3 – 0.4 eV, can occur even with an interfacial energy offset as small as ~ 0.1 eV. This enthalpy-increase CS process can only be driven by entropy. We argue that the entropic driving force can be enhanced by two ingredients – 1) the minimal spatial contact between the delocalized electron and hole wavefunction in the interlayer exciton; 2) a small band bending near the interface originated from long-range electrostatic interaction.  Our work suggests that effective CS can be achieved by entropy instead of an interfacial energy level offset, which means that energy loss at the D/A interface can be avoided. 

Characterization study of magnetism and surface properties of high-Tc insulating 2D magnet - Sr₃Fe₂O₅Cl₂ and Ca₂FeO₃Cl,

Vivek Jain,¹ Salman Ahsanullah,¹ Alex Guardiola,¹ Grant Saggars,1 Darel Andrew Pates,¹ Andrew F. May,² Michael A. McGuire,² and  Dmitry Ovchinnikov¹

¹Department of Physics and Astronomy, University of Kansas, ²Oak Ridge National Laboratory

The field of 2D vdW magnets has seen significant growth in applications and fundamental understanding since the discovery of intrinsic ferromagnetism in monolayer Cr₂Ge₂Te₆; a topic of special interest has been the enhancement of Curie-point temperature of 2D magnets in the ultrathin limit. And to achieve this goal collaborative research has enabled the discovery of a wide variety of 2D magnets showing both electronic (insulating, semiconducting and metallic) and emergent properties (ferromagnetism, anti-ferromagnetism, ferroelectricity). In this talk I will showcase a new family of high 𝑇𝑐 Ruddleson Popper phase (n = 1) perovskites and work carried out in our lab to probe magnetism in these insulating magnets using nanoscale electron tunneling devices and study its surface characteristics using scanning probe microscopy.

Quantifying Ion-Ion Association in Mixed Electrolyte Systems Using Bulk Thermodynamic Experimental Data

Elizabeth A. Ploetz and Paul E. Smith 

Department of Chemistry, Kansas State University 

Activity coefficient models for electrolyte solutions, such as the equations of Pitzer or of eNRTL, have been used in the past to obtain the experimental values of the Kirkwood-Buff integrals (KBIs) for bulk single electrolyte solutions. The KBIs in a single electrolyte solution quantify the salt-solvent, salt-salt, and solvent-solvent net affinities, and are derived from bulk thermodynamic volumetric and chemical potential composition-derivative data. In this simplest case, it is also widely known how to re-interpret these KBIs to obtain the ion-specific KBIs (cation-solvent, cation-anion, etc.). However, this process has never been performed for systems with more than one species of cation and anion, which is a severe restriction. Here we show, for the first time, how to carry out the process for bulk mixed electrolyte solutions regardless of ionconcentration, valency, molecular complexity, etc., assuming one has correlating equations for the bulk thermodynamic data. This is made possible by combining Kirkwood-Buff theory and local electroneutrality constraints. We will use the Pitzer activity coefficient model to illustrate the process for bulk mixtures for the reciprocal salt systems NaCl+KBr (aq) at 298 K and 1 bar and MgCl2+KBr (aq) at 373 K and 1 bar as well as their subsystems; however, the process could be carried out for any number of ionic components. A comparison of the experimental ion-specific KBIs to those obtained from molecular dynamics simulations of the same systems is provided and are shown to be in excellent agreement for the ambient NaCl+KBr (aq) system using the KBFF+SPC/E force field.

Probing the Effect of Osmolytes on Water Dynamics through Activation Energies

Anjali Radhakrishnan,¹ Ashley K. Borkowski,¹ Khanh V. Le,² Alan M. Allgeier,² Sarah A. Neuenswander,³ Justin T. Douglas,³ Ward H. Thompson¹

¹Department of Chemistry, University of Kansas, ²Department of Chemical and Petroleum Engineering, University of Kansas, ³Nuclear Magnetic Resonance Core Lab, University of Kansas 

The changes to water structure and dynamics induced by the presence of osmolytes are still not fully understood. Yet these effects may be important in the stabilizing and destabilizing effects of osmolytes on proteins. These effects, which are concentration dependent, can occur through direct interactions with the protein or indirectly by perturbation of the water solvent. It is now well established that urea destabilizes folded protein structures while TMAO (trimethylamine N-oxide) stabilizes them. Here, we examine how urea and TMAO influence water structure and dynamics as a function of their concentration from 1 to 8 M. We consider key dynamical properties of water that are central to much of its behavior: the self-diffusion coefficient and the hydrogen-bond exchange time. Results were dissected based on water molecule location relative to the osmolyte and their hydrogen-bonding state. Activation energies for these dynamical properties were calculated using the fluctuation theory for dynamics approach. We find, consistent with prior measurements, that water dynamics is slowed upon addition of either osmolyte and provide new insight into the role of energetic and entropic factors in this behavior. The similarities and differences between the effects of TMAO and ureaare particularly discussed.

Single-Molecule Tracking Reveals the Mechanism of Molecular Diffusion Under Nanoconfinement

Akash Nathani, Daniel A. Higgins, and Takashi Ito

Department of Chemistry, Kansas State University

Understanding mass transport mechanisms in nanopores is essential for optimizing separation processes and developing advanced materials for sensing and energy storage applications. In this study, we directly observed the diffusion behavior of single fluorescent molecules within cylindrical nanopores filled with a series of different ethanol-water mixtures. Specifically, single-molecule diffusion in horizontally-oriented anodic aluminum oxide (AAO) nanopores (10 nm or 5 nm in diameter) was measured using highly-inclined and laminated optical light sheet microscopy. Individual rhodamine B (RhB) molecules exhibited one-dimensional motions, reflecting their diffusion along the long axis of the nanopores. Interestingly, the integrated squared-displacement distribution revealed three distinct diffusion components: fast (desorption-mediated hopping motion, ≈ 10-30 µm²/s), intermediate (crawling motion, 2-4 µm²/s) and slow (wiggling motion, < 1 µm²/s) diffusion components.  In 10 nm pores, the fast component was enhanced, while the slow component was suppressed across the solvent mixtures, resulting from weakened dye-surface interactions as previously shown using fluorescence correlation spectroscopy. In 5 nm pores, the fast component was suppressed for pure solvents as well as formixtures, attributed to rapid re-adsorption of the dye due to confinement. The contribution of the intermediate component to overall diffusion remained nearly constant across the solvent systems for both the nanopores. These findings highlight the interplay between solvation-mediated surface interactions and nanoconfinement in governing molecular transport, providing deeper insights into diffusion mechanisms within nanoporous environments.

DFT Investigation of Hydrogen Abstraction and Radical Coupling Pathways for 2,6-Substituted Pyridinium-Derived Radicals

Catherine Moraghan and Gail S. Blaustein

Department of Chemistry, Benedictine College

The efficiency of pyridinium-derived radicals as non-aqueous redox flow battery anolytes is hindered by their susceptibility to decomposition via hydrogen abstraction (HA) and radical-coupling (RC) pathways. A deeper understanding of these pathways—particularly the impact of alkyl substituents on anolyte stability—is needed to improve anolyte design. Density functional theory was used to analyze the HA and RC pathways of the methyl, ethyl, and isopropyl forms of a 2,6-substituted pyridinium-derived radical. Enthalpy trends for the three HA pathways, determined using isodesmic reactions, will be presented. Additionally, an overview will be provided of the decomposition mechanisms and the computational strategies employed to locate transition states. Current progress toward identifying transition states for the three HA pathways and for the methyl-form RC pathway will be presented.

Enhanced Low-Field Differential Ion Mobility Spectrometry of Pendular Macromolecules using Tunable Rectangular Waveforms

Egor Gusachenko,¹ Hayden A. Thurman,¹ Gordon A. Anderson,² Alexandre A. Shvartsburg1

¹Department of Chemistry and Biochemistry, Wichita State University, ²GAA Custom Electronics LLC

Differential IMS separates ions by the increment of mobility in gases between two electric field levels with orthogonality to MS and linear IMS relying on the absolute mobility. In low-field differential (LOD) IMS, the moderate fields align macroions with strong dipoles, revealing the ratio of directional to orientationally-averaged collision cross section (a_⟂). This provides the complementary structural descriptors: the dipole moment and molecular elongation. The initial work adopted the bisinusoidal waveforms common to FAIMS, with suboptimal profiles required by the high voltages and frequencies redundant for LODIMS.

We enabled LODIMS using near-ideal rectangular waveforms (0.1 μs rise time) with flexible frequency (5-250 kHz) and aspect ratio (f) up to 100. This modality improves the resolution, simplifies the alignment dynamics, and permits smaller drivers. A planar IMS cell (1.88 mm gap) was coupled to a Thermo LTQ Velos. Proteins were ionized by ESI under denaturing conditions.

We explored the ADH (37 kDa) and BSA (66 kDa) exhibiting monomers, dimers, and larger oligomers. Odd-z dimers were isolated to avoid the overlapping monomers. The aligned and rotary conformers appeared at extreme negative and small positive compensation field (EC), respectively. The curves of E_C vs. dispersion field (E_D) were cubic at positive E_C but linear at negative E_C, consistent with the discrete alignment model. The slopes were near-flat with frequency but expectedly decreased at higher f with the a_⟂ metric approaching an asymptote. The ensuing a_⟂ estimates are superior to those with the bisinusoidal waveforms a priori and match known extended protein conformations closer.

Probing Excited-State Dynamics of a Photoswitch with Resonance-Enhanced Raman Spectroscopy

Emmaline R. Lorenzo and Christopher G. Elles

Department of Chemistry, University of Kansas

Ultrafast femtosecond stimulated Raman spectroscopy (FSRS) is a powerful vibrational spectroscopy technique for probing the dynamics of a molecule in its excited state. This three-beam technique allows for a stimulated Raman spectrum to be obtained at a variable time delay after an initial photoexcitation to the first excited state. Here, we investigate the ring-opening reaction of a diarylethene-based photoswitch (DAE-BT). This type of photoswitch has distinct “open” and “closed” isomeric properties, which allow them to be used for optical data storage, energy harvesting, and as model complexes for photochromic reactions. DAE-BT exhibits a narrow excited-state absorption band at 710 nm in its transient absorptionspectrum, which we target to perform resonance-enhanced FSRS. We also implement a tunable etalon filter to create an asymmetric Raman pump pulse to further enhance the Raman scattering intensity and mitigate excess background signals. Following excitation with the actinic pump, we observe an oscillation of the center frequency of several vibrational modes over the first couple picoseconds. We also track the evolution of the vibrational bands on the longer time scale of the excited state lifetime, with the population decaying completely back to the ground state within 100 ps. Finally, we perform TD-DFT calculations of the off-resonance Raman spectrum of the first excited state to support making vibrational assignments for the experimental spectrum.

Fabrication of Nanoscale Devices for beyond CMOS Applications

Abishai Mathai, Salman Ahsanullah, Grant Saggar, Darel Pates, and Dmitry Ovchinnikov

Department of Physics and Astronomy, University Of Kansas

My research focuses on creating nanoscale devices that overcome the limitations of present transistor technology, leading to performance upgrades in new types of memory technology, energy storage devices, and quantum computing applications. Our laboratory is developing new materials and device architectures based on two-dimensional quantum materials (2D materials). One prominent example of such devices is the development of technologies that use spin instead of charge as an information carrier. This new architecture of devices leads us to spintronic devices such as Magnetic Tunnel Junction Diodes (MTJ’s) and Spin Orbit Torque Devices (SOT). Their advantages over current technology include low power consumption, non-volatility, and high Read-Write cycles making them perfect candidates for energy-efficient memory devices. Another important venue of exploration is tuning of materials’ properties with external stimuli, and designing magnetic, semiconducting, and superconducting materials with programmable properties.

I use the state-of-the-art microfabrication capabilities of our laboratory and KU Nanofabrication facilities to create our devices. I use lithography, metal deposition, and plasma etching techniques to create our structures. Furthermore, in line with best practices in semiconductor industry research, I use metrology tools such as atomic force microscope and scanning electron microscope to understand the surfaces and interfaces at very high resolution, down to a single nanometer. To access the fundamental properties and device performance I perform electrical transport measurements as a function of temperature and magnetic field.

These metrology, nanofabrication, and measurement methods allow us to create nanoscale devices for beyond CMOS applications.

Development of a Force Field for Gold Hydride Nanoclusters that Predicts Potential Energy with Quantum-level Precision

D. Sulalith N. D. Samarasinghe and Christine M. Aikens 

Department of Chemistry, Kansas State University

Fast performance of quantum-mechanically accurate molecular dynamics (MD) calculations for larger molecular systems is desirable, but density functional theory (DFT) calculations are computationally expensive. A fast method is potentially useful for studying gold nanoclusters, which can be considerably large systems, to understand more about their unique characteristics which lead to applications in biomedical, catalysis, and sensing. Therefore, in this project, we are generating a force field with DFT level accuracy to predict the energies and dynamics of gold hydride nanoclusters using the fast learning of atomistic rare events (FLARE) code. The training process is initiated using Au₁₃H₅ and Au₁₈H₁₄ clusters. Once the training was finished, we used the LAMMPS package to simulate MD using each trained force field. The potential energy values obtained from the MD simulation can be verified through DFT calculations using the VASP software package. The initial results indicated that the force fields need further improvement. Therefore, a combined sparse set including all data points originating from both gold hydride clusters was considered for another round of training. Iterating this procedure, which includes validating the new force field using DFT calculations, updating the sparse dataset with new configurations, and retraining the model, shows that accuracy increases with each training cycle. Our current force field is a product of four training cycles, with 647 gold hydride nanocluster configurations in its database. Our current force field indicates that it can predict potential energies with relative accuracy for gold hydride systems that are within and outside the database.

Theoretical Study of Solvation and Substituent Effects on Internal Hydrogen Bonding of Cyclophanes

Matt Knehans and Gail S. Blaustein 

Department of Chemistry, Benedictine College

This study addressed the effects of electron-donating (EDGs) and electron-withdrawing (EWGs) groups and solvation of carbazolopyrindophane (CP) and diphenylaminopyridinophane (DP) derivatives on the characteristics of their internal N-H···N hydrogen bonds. The M06/6-31G(d,p) model chemistry was used to predict the molecular geometry and N-H···N bond length of CP, with the results validated by comparison to X-ray crystallography data for CP. EDGs and EWGs were added to CP and DP and the results indicated that the addition of EDGs to the cyclophanes caused a decrease in N-H···N bond length, and the addition of EWGs increased the N-H···N bond length. However, it was observed in the CP and DP derivatives the effect of substituents on the twist angle was inconsistent and rather trivial. The effects of the solvation of the CP and DP derivatives in water, heptane, and chloroform shed light onto trends, where the CP derivatives exhibited shorter N-H···N bonds in more polar solvents, such as water. In contrast, solvation had a negligible impact on the N-H···N bond length in DP derivatives The calculated basis set superposition error corrected energies confirmed that EDGs have a decreasing effect on the internal hydrogen bond energies and EWGs have an increasing effect. The study furthermore demonstrated the importance of solvent polarity in modulating non-covalent interactions, suggesting potential implications on the tunability of the internal hydrogen bond in CP and DP derivatives. The findings highlight the complexity of modeling non-covalent interactions in various solvents, offering valuable insights for the development of reusable cyclophane-based hydrazine sensors.

Simulating Absorption Spectra using EOM-CCSD with a Polarizable Force Field: Application to the GFP Chromophore

Taylor Parsons,¹ Andrew Snider,² Arthur Pyuskulyan,² Christine Isborn,² and Marco Caricato¹¹Department of Chemistry, University of Kansas, ²Department of Chemistry and Biochemistry, University of California Merced

Accurately representing spectra for solvated systems is still problematic for simu- lations, particularly with polar solvents capable of forming hydrogen bonds. The spectral broadening induced by solvent-solute interactions can be difficult to repli- cate with simulations, even for fully QM methods such as TDDFT. One potential workaround is to utilize hybrid methods, where the solute of a system is treated with a higher level of theory, and the environment is treated with a lower level, cheaper method. An EOM-CCSD solute in a TDDFT environment has been shown to im- prove upon this broadening, but the unfavorable cost scaling of fully QM methods makes the study of many large systems difficult. One potential cost-saving method is to treat the solute with a QM level of theory, and treat the solvent with classical MM. Here, we present results for QM/MMPol (i.e., a polarizable force field) calculations applied to the anionic GFP chromophore solvated in both water and methanol, where the chromophore is treated with EOM-CCSD and the solvent is treated with MMPol. In MMPol, polarization is induced with a point dipole placed at each atomic center in the MM region, whose magnitude and direction are determined by the electric field from both QM and MM regions. We compare the QM/MM results with both gas- phase calculations and with the solvent treated as point charges. We show that, for both solvents under consideration, there is a consistent increase in broadness going from the gas-phase to point charges to MMPol. Additionally, we show that EOM- CCSD improves on broadness compared to TDDFT for each solvent representation.

Computational Insights into the Applicability of Marcus Theory to Hydrated Electron Transfer to Organic Molecules in Aqueous Solution

Wilberth A. Narvaez,¹ Pauf Neupane,¹ David M. Bartels,² and Ward H. Thompson¹ 

¹Department of Chemistry, University of Kansas, ²Radiation Laboratory, University of Notre Dame

Electron transfer reactions in solution can typically be rationalized through the lens of Marcus theory, which predicts a quadratic relationship between the free energy of reaction and the free energy of activation. The theory’s predictions, however, seem to be at odds with several organic reduction reactions involving the quintessential reductant species known as the hydrated electron, which is an excess electron localized within an excluded volume that is solvated by its nearest four to six water molecules. For example, the hydrated electron transfer reactions involving a series of carbonyl compounds with free energies of reactions ranging from -90 kJ/mol to 10 kJ/mol all have activation energies that fall within a 12 to 18 kJ/mol range. Even more, these electron transfer reactions have rate constants that vary by orders of magnitude despite having identical activation energies and involving structurally similar reactants. We address this apparent failure of Marcus theory through the study of the hydrated electron’s reduction of carbon dioxide and acetone in aqueous solution using ab initio molecular dynamics and constrained density functional theory calculations. Through a quantitative exploration of the reactions’ solvent and vibrational reorganization energies, diabatic couplings, attempt frequencies, and free energies of activation, we can establish that these reactions proceed adiabatically and would fall within the normal Marcus regime.