Empowering Capacitive Devices: Harnessing Transfer Learning for Enhanced Data-Driven Optimization.

Developing data-driven models has found successful applications in engineering tasks, such as material design, process modeling, and process monitoring. In capacitive devices like deionization and supercapacitors, there exists potential for applying this data-driven machine learning (ML) model in optimizing its potential use in energy-efficient separations or energy generation. However, these models are faced with limited datasets, and even in large quantities, the datasets are incomplete, limiting their potential use for successful data-driven modeling. Here, the success of transfer learning in resolving the challenges with limited datasets was exploited. A two-step data-driven ML modeling framework named ImputeNet involving training with ML-imputed datasets and then with clean datasets was explored. Through data imputation and transfer learning, it is possible to develop a data-driven model with acceptable metrics mirroring experimental measurements. By using the model, optimization studies using the genetic algorithm were implemented to analyze the solution under the Pareto optimality. This early insight can be used in the initial stage of experimental measurements to rapidly identify experimental conditions worthy of further investigation. Moreover, we expect that the insights from these results will drive accurate predictive modeling in other fields including healthcare, genomic data analysis, and environmental monitoring with incomplete datasets.

Probing the Electrode-Electrolyte Interface of Sodium/Glyme-Based Battery Electrolytes.

Sodium-ion batteries (NIBs) are promising systems for large-scale energy storage solutions; yet, further enhancements are required for their commercial viability. Improving the electrochemical performance of NIBs goes beyond the chemical description of the electrolyte and electrode materials as it requires a comprehensive understanding of the underlying mechanisms that govern the interface between electrodes and electrolytes. In particular, the decomposition reactions occurring at these interfaces lead to the formation of surface films. Previous work has revealed that the solvation structure of cations in the electrolyte has a significant influence on the formation and properties of these surface films. Here, an experimentally validated molecular dynamics study is performed on a 1 M NaTFSI salt in glymes of different lengths placed between two graphite electrodes having a constant bias potential. The focus of this study is on describing the solvation environment around the sodium ions at the electrode-electrolyte interface as a function of glyme chain length and applied potential. The results of the study show that the diglyme/TFSI system presents features at the interface that significantly differ from those of the triglyme/TFSI and tetraglyme/TFSI systems. These computational predictions are successfully corroborated by the experimentally measured capacitance of these systems. In addition, the dominant solvation structures at the interface explain the electrochemical stability of the system as they are consistent with cyclic voltammetry characterization.

Computational Investigations of the Water Structure at the α-Al2O3(0001)-Water Interface.

The α-Al2O3(0001)-water interface is investigated using ab initio molecular dynamics (AIMD) simulations. The spectral signatures of the vibrational sum frequency generation (vSFG) spectra of OH stretching mode for water molecules at the interface are related to the interfacial water orientation, hydrogen bond network, and water dissociation process at different water/alumina interfaces. Significant differences are found between alumina surfaces at different hydroxylation levels, namely, Al-terminated and O-terminated α-Al2O3(0001). By calculating the vibrational sum frequency generation spectrum and its imaginary component from AIMD results, the structure of interfacial waters as well as the termination of alumina slab are related to the spectral signatures of vSFG data.

Adsorption Studies at the Graphene Oxide-Liquid Interface: A Molecular Dynamics Study.

The adsorption of organic aromatic molecules, namely aniline, onto graphene oxide is investigated using molecular simulations. The effect of the oxidation level of the graphene oxide sheet as well as the presence of two different halide salts, sodium chloride and sodium iodide, were examined. The aniline molecule in the more-reduced graphene oxide case, in the absence of added salt, showed a slightly greater affinity for the graphene oxide-water interface as compared to the oxidized form. The presence of the iodide ion increased the affinity of the aniline molecule in the reduced case but had the opposite effect for the more-oxidized form. The effect of oxidation and added salt on the interfacial water layer was also examined.

Unraveling the Role of Charge Patterning in the Micellar Structure of Sequence-Defined Amphiphilic Peptoid Oligomers by Molecular Dynamics Simulations.

Electrostatic interactions play a significant role in regulating biological systems and have received increasing attention due to their usefulness in designing advanced stimulus-responsive materials. Polypeptoids are highly tunable N-substituted peptidomimetic polymers that lack backbone hydrogen bonding and chirality. Therefore, polypeptoids are suitable systems to study the effect of noncovalent interactions of substituents without complications of backbone intramolecular and intermolecular hydrogen bonding. In this study, all-atom molecular dynamics (MD) simulations were performed on micelles formed by a series of sequence-defined ionic polypeptoid block copolymers consisting of a hydrophobic segment and a hydrophilic segment in an aqueous solution. By combining the results from MD simulations and experimental small-angle neutron scattering data, further insights were gained into the internal structure of the formed polypeptoid micelles, which is not always directly accessible from experiments. In addition, information was gained into the physics of the noncovalent interactions responsible for the self-assembly of weakly charged polypeptoids in an aqueous solution. While the aggregation number is governed by electrostatic repulsion of the negatively charged carboxylate (COO-) substituents on the polypeptoid chain within the micelle, MD simulations indicate that the position of the charge on singly charged chains mediates the shape of the micelle through the charge-dipole interactions between the COO- substituent and the surrounding water. Therefore, the polypeptoid micelles formed from the single-charged series offer the possibility for tailorable micelle shapes. In contrast, the polypeptoid micelles formed from the triple-charged series are characterized by more pronounced electrostatic repulsion that competes with more significant charge-sodium interactions, making it difficult to predict the shape of the micelles. This work has helped further develop design principles for the shape and structure of self-assembled micelles by controlling the position of charged moieties on the backbone of polypeptoid block copolymers.

Reactive events at the graphene oxide-water interface.

Graphene oxide exhibits interesting reactive events at its interface with water, with water as an active participant. The reactive events are influenced by the level of oxidation of the graphene oxide sheet. The fully oxidized sheet tends to make the interfacial water media acidic leaving the sheet negatively charged, whereas the reduced sheet can form comparatively long lived carbocations as well as split water forming two alcohol groups on the sheet.

Influence of Temperature on Molecular Adsorption and Transport at Liposome Surfaces Studied by Molecular Dynamics Simulations and Second Harmonic Generation Spectroscopy.

A fundamental understanding of the kinetics and thermodynamics of chemical interactions at the phospholipid bilayer interface is crucial for developing potential drug-delivery applications. Here we use molecular dynamics (MD) simulations and surface-sensitive second harmonic generation (SHG) spectroscopy to study the molecular adsorption and transport of a small organic cation, malachite green (MG), at the surface of 1,2-dioleoyl-sn-glycero-3-phospho-(1'-rac-glycerol) (DOPG) liposomes in water at different temperatures. The temperature-dependent adsorption isotherms, obtained by SHG measurements, provide information on adsorbate concentration, free energy of adsorption, and associated changes in enthalpy and entropy, showing that the adsorption process is exothermic, resulting in increased overall entropy. Additionally, the molecular transport kinetics are found to be more rapid under higher temperatures. Corresponding MD simulations are used to calculate the free energy profiles of the adsorption and the molecular orientation distributions of MG at different temperatures, showing excellent agreement with the experimental results.

Heterogeneous assembly of water from the vapor phase-Physical experiments and simulations with binding trifunctional organosilanes at the vapor/solid interface.

A trace amount of interfacial water is required to initiate hydrosilation reactions of trifunctional organosilanes to form surface assemblies. In recent studies, we have learned that water also has a critical role in directing molecular placement on surfaces because water can react with silicon to provide oxygenated sites for surface binding. Consequently, the wettability nature of substrates influences the placement and density of organosilane films formed by vapor-phase reactions. Nanopatterning protocols were designed using vapor-phase organosilanes and colloidal lithography to compare the wettability differences of hydrophilic mica(0001) compared to relatively hydrophobic Si(100) as a strategy for tracking the location of water on surfaces. The competition between hydrophobic and hydrophilic domains for the adsorption and coalescence of water condensed from vapor can be mapped indirectly by mapping the organosilanes, which bind to water at the solid interface, using atomic force microscopy. Trifunctional octadecyltrichlorosilane (OTS) was used as a marker molecule to map out the areas of the surface where water was deposited. The effect of systematic changes in film thickness and surface coverage of OTS was evaluated at the vapor/solid interface by adding an incremental amount of water to sealed reaction vessels to wet the surface and assessing the outcome after reaction with vapor-phase trichlorosilane. Reactive molecular dynamics simulations of the silicon-water vapor interface combined with electronic structure calculations of oxygenated silicon clusters with methyltrichlorosilane provided insight of the mechanism for surface binding, toward understanding the nature of the interface and wettability factors, which influence the association and placement of silane molecules on surfaces.

Effect of anion identity on ion association and dynamics of sodium ions in non-aqueous glyme based electrolytes-OTf vs TFSI.

Sodium-based rechargeable battery technologies are being pursued as an alternative to lithium, in part due to the relative abundance of sodium compared to lithium. Despite their low dielectric constant, glyme-based electrolytes are particularly attractive for these sodium-based batteries due to their ability to chelate with the sodium ion and their high electrochemical stability. While the glyme chain length is a parameter that can be tuned to modify solvation properties, charge transport behavior, reactivity, and ultimately battery performance, anion identity provides another tunable variable. Trifluoromethanesulfonate (triflate/OTf) and bis(trifluoromethane)sulfonamide (TFSI) are chemically similar anions, which are often used in battery electrolytes for lithium-based batteries. In this paper, molecular simulations are used to examine the differences in ion association and charge transport between sodium salts of these two anions at different salt concentrations in glymes with the increasing chain length. The use of the modified force field developed for NaOTf in glymes for the NaTFSI electrolytes was validated by comparing the TFSI-sodium ion radial distribution functions to the results from ab initio molecular dynamics simulations on 1.5 M NaTFSI in diglyme. While the ion association behavior as a function of salt concentration showed similar trends for both NaOTf and NaTFSI in tetraglyme and triglyme electrolytes, the dominant solvation structures for the two sets of electrolytes are distinctly different in the monoglyme and diglyme cases. The conductivity is impacted by both the ion association behavior in these electrolytes and the non-vehicular or hopping transport of the anions in these systems.

Understanding the ionic activity and conductivity value differences between random copolymer electrolytes and block copolymer electrolytes of the same chemistry.

Herein, a systematic study where the macromolecular architectures of poly(styrene-block-2-vinyl pyridine) block copolymer electrolytes (BCE) are varied and their activity coefficients and ionic conductivities are compared and rationalized versus a random copolymer electrolyte (RCE) of the same repeat unit chemistry. By performing quartz crystal microbalance, ion-sorption, and ionic conductivity measurements of the thin film copolymer electrolytes, it is found that the RCE has higher ionic activity coefficients. This observation is ascribed to the fact that the ionic groups in the RCE are more spaced out, reducing the overall chain charge density. However, the ionic conductivity of the BCE is 50% higher and 17% higher after the conductivity is normalized by their ion exchange capacity values on a volumetric basis. This is attributed to the presence of percolated pathways in the BCE. To complement the experimental findings, molecular dynamics (MD) simulations showed that the BCE has larger water cluster sizes, rotational dynamics, and diffusion coefficients, which are contributing factors to the higher ionic conductivity of the BCE variant. The findings herein motivate the design of new polymer electrolyte chemistries that exploit the advantages of both RCEs and BCEs.

Effect of Oxidation Level on the Interfacial Water at the Graphene Oxide-Water Interface: From Spectroscopic Signatures to Hydrogen-Bonding Environment.

The interfacial region of the graphene oxide (GO)-water system is nonhomogenous due to the presence of two distinct domains: an oxygen-rich surface and a graphene-like region. The experimental vibrational sum-frequency generation (vSFG) spectra are distinctly different for the fully oxidized GO-water interface as compared to the reduced GO-water case. Computational investigations using ab initio molecular dynamics were performed to determine the molecular origins of the different spectroscopic features. The simulations were first validated by comparing the simulated vSFG spectra to those from the experiment, and the contributions to the spectra from different hydrogen bonding environments and interfacial water orientations were determined as a function of the oxidation level of the GO sheet. The ab initio simulations also revealed the reactive nature of the GO-water interface.

Ion Pairing in HCl-Water Clusters: From Electronic Structure Investigations to Multiconfigurational Force-Field Development.

In the bulk, condensed-phase HCl exists as a dissociated Cl- ion and a proton that is delocalized over solvating water molecules. However, in the gas phase, HCl is covalent, and even on the introduction of hydrating water molecules, the HCl covalent state dominates small clusters and is relevant at larger clusters including 21 water molecules. Electronic structure calculations (at the MP2 level) and ab initio metadynamics simulations (at the DFT level) have been carried out on HCl-(H2O)n clusters with n = 2-22 to investigate distinct solvation environments in clusters from covalent HCl structure, to contact ion pairs and solvent-separated ion pairs. The data were further used to train and validate a multiconfigurational force-field for HCl-water clusters that incorporates covalent HCl states into the MS-EVB3.2 formalism. Additionally, the many-body interaction of the Cl- ion with water and the excess proton was modeled by the introduction of two geometric three-body terms that incorporates the dominant many-body interaction in an efficient noniterative manner.

Molecular Adsorption and Transport at Liposome Surfaces Studied by Molecular Dynamics Simulations and Second Harmonic Generation Spectroscopy.

A fundamental understanding of the factors that determine the interactions with and transport of small molecules through phospholipid membranes is crucial in developing liposome-based drug delivery systems. Here we combine time-dependent second harmonic generation (SHG) measurements with molecular dynamics simulations to elucidate the events associated with adsorption and transport of the small molecular cation, malachite green isothiocyanate (MGITC), in colloidal liposomes of different compositions. The molecular transport of MGITC through the liposome bilayer is found to be more rapid in 1,2-dioleoyl-sn-glycero-3-phospho-(1'-rac-glycerol) and 1,2-dioleoyl-sn-glycero-3-phospho-l-serine (DOPG and DOPS, respectively) liposomes, while the molecular transport is slower in 1,2-dioleoyl-sn-glycero-3-phosphocholine (DOPC) liposomes. Interestingly, MGITC is observed to neither adsorb nor transport in trimethyl quinone-1,2-dioleoyl-sn-glycero-3-phosphoethanolamine (QPADOPE) liposomes due to shielding by the quinone group. The modified Langmuir adsorption isotherm model is used to determine the free energy of adsorption for MGITC, which is found to be less negative in DOPC than in DOPG and DOPS, caused by lower electrostatic interactions between the positively charged dye and the zwitterionic DOPC liposome surface. The results are compared to our previous investigations, which showed that malachite green (MG) adsorbs and transports in DOPG and DOPS liposomes but not in DOPC and QPADOPE liposomes. Molecular dynamics simulations are used to investigate the adsorption and transport properties of MG and MGITC in DOPC and DOPG liposomes using umbrella sampling to determine the free energy profiles and interfacial molecular orientations. Together, these time-resolved SHG studies and corresponding molecular dynamics simulations characterize the complicated chemical interactions at different lipid membranes to provide key molecular-level insights for potential drug delivery applications. The results also point toward understanding the role of chemical functional groups, in this case isothiocyanate, in controlling molecular adsorption at and transport through lipid bilayers.

Tumor-Targeting NIRF NanoGUMBOS with Cyclodextrin-Enhanced Chemo/Photothermal Antitumor Activities.

The near-infrared fluorescent (NIRF) dye, IR780, is recognized as a promising theranostic agent and has been widely investigated for imaging, chemotherapeutic, and phototherapeutic applications. However, its poor photostability and nonselective toxicities toward both cancer and normal cells limit its biological applications. Herein, we introduce the use of GUMBOS (a group of uniform materials based on organic salts) developed through counter-anion exchange with IR780 and subsequent nanomaterials (nanoGUMBOS) formed by complexation with cyclodextrin (CD) for enhanced chemo/photothermal therapy. Such CD-based nanoGUMBOS display improved aqueous stability, photostability, and photothermal effects relative to traditional IR780. The examination of in vitro cytotoxicity reveals that CD-based nanoGUMBOS are selectively toxic toward cancer cells and exhibit synergistically enhanced cytotoxicity toward cancer cells upon NIR laser irradiation. Additionally, in vivo NIRF imaging demonstrated selective accumulation of these nanoGUMBOS within the tumor site, indicating tumor-targeting properties. Further in vivo therapeutic study of these CD-based nanoGUMBOS suggests excellent chemo/photothermal antitumor effects. Using these studies, we herein demonstrate a promising strategy, via conversion of IR780 into nanoGUMBOS, that can be used for improved theranostic cancer treatment.

Coarse-Grained Models for Constant pH Simulations of Carboxylic Acids.

A model for carboxylic acids, in both the protonated and deprotonated states, is developed in which hydrogen interaction sites are not used and all interactions are short-ranged. A method for constant pH simulations, which exploits these features of the model, is developed. The constant pH method samples protonation states by making discrete Monte Carlo steps and is able to efficiently move between states in two steps. The method is applied to the polymer poly(methacrylic acid), a pH-responsive polymer that undergoes structural changes as a function of pH. The model is able to reproduce the structural changes induced by pH.

Mechanistic Perspectives in the Regioselective Indole Addition to Unsymmetrical Silyloxyallyl Cations.

Our investigations on the reaction mechanism to account for regioselectivity on the addition of indoles to unsymmetrical silyloxyallyl cations are reported. Using both experimental and computational methods, we confirmed the significance of steric effects from the silyl ether group toward directing the inward approach of indoles, leading to nucleophilic attack at the less substituted electrophilic α'-carbon. The role of residual water toward accelerating the rate of reaction is established through stabilization of the participating silyloxyallyl cation.

Interfacial Water at Graphene Oxide Surface: Ordered or Disordered?

The graphene oxide (GO)-water interface was simulated using Born-Oppenheimer molecular dynamics (BOMD) simulations with two different functionals, namely, revPBE-D3 and BLYP-D2, as well as a commonly used classical force field, namely, OPLS-AA. A number of different order parameters, including the orientation of the interfacial water molecules near the aromatic region of the GO surface as well as those near the oxygenated defects, were examined and compared. The BOMD interfacial waters are clearly much less structured as compared to the classical force field that shows a strongly ordered interface. Higher-level calculations, namely, symmetry adapted perturbation theory, were performed on representative clusters taken from the BOMD simulation. These calculations revealed not only that a number of conformations have similar interaction energies but also the importance of induction contribution to the interaction energies.

Towards a coarse-grained model of the peptoid backbone: the case of N,N-dimethylacetamide.

In this study, a coarse-grained (CG) model for N,N-dimethylacetamide (DMA), which represents the polypeptoid backbone, is developed as a step towards establishing a CG model of the complex polypeptoid system. Polypeptoids or poly N-substituted glycines are a type of peptidomimetic polymers that are highly tunable, and hence an ideal model system to study self-assembly as a function of chemical groups in aqueous soft matter systems. The DMA CG model is parameterized to reproduce the structural properties of DMA liquid as well as a dilute aqueous solution of DMA using a reference all atom model, namely the OPLS-AA force-field. The intermolecular forces are represented by the Stillinger-Weber potential, that consists of both two- and three-body terms that are very short-ranged. The model is validated on thermodynamic properties of liquid and aqueous DMA, as well as the vapor-liquid interface of liquid DMA and the structure of a concentrated aqueous solution of DMA in water as well as a simple peptoid in water. Without long-ranged interactions and the absence of interaction sites on hydrogen atoms, the CG DMA model is an order of magnitude faster than the higher resolution all-atom (AA) model.

Enantioselective Functionalization of Enamides at the ß-Carbon Center with Indoles.

We report an enantioconvergent approach for the functionalization of enamides at the ß-carbon atom, which involves a chiral Brønsted acid induced tautomerization of 2-amidoallyl into 1-amidoallyl cations. These putative reactive intermediates were produced by ionization of racemic α-hydroxy enamides with a chiral Brønsted acid and captured with substituted indoles in a highly regio- and enantioselective manner.

Influence of Sulfur on Acid-Mediated Enamide Formation.

The acid-mediated condensation of acetamide with butanal dimethylacetal and EtSCH2CH(OMe)2, followed by dehydration, was investigated by electronic structure calculations that supported the prediction that the Z-geometry would be favored in the product. The reaction was investigated experimentally using suitably functionalized cysteine building blocks. Some side reactions and optimization of reaction conditions are reported, en route to identifying a mild, inexpensive Lewis acid that achieves a reasonable yield of (Z)-thioenamide 21 with high stereoselectivity.