Anion-π Interactions in Monolayers Formed by Amphiphilic Electron-Deficient Aromatic Compounds at Air/Water Interfaces.

Biological systems precisely and selectively control ion binding through various chemical reactions, molecular recognition, and transport by virtue of effective molecular interactions with biological membranes and proteins. Because ion binding is inhibited in highly polar media, recognition systems for anions in aqueous media, which are relevant to biological and environmental systems, are still limited. In this study, we explored the anion binding of Langmuir monolayers formed by amphiphilic naphthalenediimide (NDI) derivatives with a series of substituents at air/water interfaces via anion-π interactions. Density functional theory (DFT) simulations revealed that the binding of anions originating from anion-π interactions is related to the electron density of the anions. At the air/water interfaces, amphiphilic NDI derivatives formed Langmuir monolayers, and the addition of anions caused expansion of the Langmuir monolayers. The anions with larger hydration energies related to electron density showed larger binding constants (Ka) for 1:1 stoichiometry with the NDI derivatives. The loosely packed monolayer formed by the amphiphilic NDI derivatives with bromine groups showed a better anion response. In contrast, the binding of NO3- was significantly enhanced in the highly packed monolayer. These results indicate that the packing of NDI derivatives with rigid aromatic rings influenced the binding of the anions. These results provide insight into ion binding using the air/water interface as a promising recognition site for mimicking biological membranes. In future, sensing devices can be developed using Langmuir-Blodgett films on electrodes. Furthermore, the capture of anions on electron-deficient aromatic compounds can lead to doping or composition technologies for n-type semiconductors.

Black-Box Optimization for Automated Discovery.

In chemistry and materials science, researchers and engineers discover, design, and optimize chemical compounds or materials with their professional knowledge and techniques. At the highest level of abstraction, this process is formulated as black-box optimization. For instance, the trial-and-error process of synthesizing various molecules for better material properties can be regarded as optimizing a black-box function describing the relation between a chemical formula and its properties. Various black-box optimization algorithms have been developed in the machine learning and statistics communities. Recently, a number of researchers have reported successful applications of such algorithms to chemistry. They include the design of photofunctional molecules and medical drugs, optimization of thermal emission materials and high Li-ion conductive solid electrolytes, and discovery of a new phase in inorganic thin films for solar cells.There are a wide variety of algorithms available for black-box optimization, such as Bayesian optimization, reinforcement learning, and active learning. Practitioners need to select an appropriate algorithm or, in some cases, develop novel algorithms to meet their demands. It is also necessary to determine how to best combine machine learning techniques with quantum mechanics- and molecular mechanics-based simulations, and experiments. In this Account, we give an overview of recent studies regarding automated discovery, design, and optimization based on black-box optimization. The Account covers the following algorithms: Bayesian optimization to optimize the chemical or physical properties, an optimization method using a quantum annealer, best-arm identification, gray-box optimization, and reinforcement learning. In addition, we introduce active learning and boundless objective-free exploration, which may not fall into the category of black-box optimization.Data quality and quantity are key for the success of these automated discovery techniques. As laboratory automation and robotics are put forward, automated discovery algorithms would be able to match human performance at least in some domains in the near future.

QCforever: A Quantum Chemistry Wrapper for Everyone to Use in Black-Box Optimization.

To obtain observable physical or molecular properties such as ionization potential and fluorescent wavelength with quantum chemical (QC) computation, multi-step computation manipulated by a human is required. Hence, automating the multi-step computational process and making it a black box that can be handled by anybody are important for effective database construction and fast realistic material design through the framework of black-box optimization where machine learning algorithms are introduced as a predictor. Here, we propose a Python library, QCforever, to automate the computation of some molecular properties and chemical phenomena induced by molecules. This tool just requires a molecule file for providing its observable properties, automating the computation process of molecular properties (for ionization potential, fluorescence, etc.) and output analysis for providing their multi-values for evaluating a molecule. Incorporating the tool in black-box optimization, we can explore molecules that have properties we desired within the limitation of QC computation.

Automatic Differentiation for the Direct Minimization Approach to the Hartree-Fock Method.

Automatic differentiation (AD) has become an important tool for optimization problems in computational science, and it has been applied to the Hartree-Fock method. Although the reverse-mode AD is more efficient than the forward-mode, eigenvalue calculation in the self-consistent field (SCF) method has impeded the use of the reverse-mode AD. Here, we propose a method to directly minimize Hartree-Fock energy under the orthonormality constraint of the molecular orbitals using reverse-mode AD by avoiding eigenvalue calculation. According to our validation, the proposed method was more stable than the conventional SCF method and achieved comparable accuracy.

Understanding the evolution of a de novo molecule generator via characteristic functional group monitoring.

Recently, artificial intelligence (AI)-enabled de novo molecular generators (DNMGs) have automated molecular design based on data-driven or simulation-based property estimates. In some domains like the game of Go where AI surpassed human intelligence, humans are trying to learn from AI about the best strategy of the game. To understand DNMG's strategy of molecule optimization, we propose an algorithm called characteristic functional group monitoring (CFGM). Given a time series of generated molecules, CFGM monitors statistically enriched functional groups in comparison to the training data. In the task of absorption wavelength maximization of pure organic molecules (consisting of H, C, N, and O), we successfully identified a strategic change from diketone and aniline derivatives to quinone derivatives. In addition, CFGM led us to a hypothesis that 1,2-quinone is an unconventional chromophore, which was verified with chemical synthesis. This study shows the possibility that human experts can learn from DNMGs to expand their ability to discover functional molecules.

High reactivity of H2O vapor on GaN surfaces.

Understanding the process of oxidation on the surface of GaN is important for improving metal-oxide-semiconductor (MOS) devices. Real-time X-ray photoelectron spectroscopy was used to observe the dynamic adsorption behavior of GaN surfaces upon irradiation of H2O, O2, N2O, and NO gases. It was found that H2O vapor has the highest reactivity on the surface despite its lower oxidation power. The adsorption behavior of H2O was explained by the density functional molecular dynamic calculation including the spin state of the surfaces. Two types of adsorbed H2O molecules were present on the (0001) (+c) surface: non-dissociatively adsorbed H2O (physisorption), and dissociatively adsorbed H2O (chemisorption) molecules that were dissociated with OH and H adsorbed on Ga atoms. H2O molecules attacked the back side of three-fold Ga atoms on the (0001̅) (-c) GaN surface, and the bond length between the Ga and N was broken. The chemisorption on the (101̅0) m-plane of GaN, which is the channel of a trench-type GaN MOS power transistor, was dominant, and a stable Ga-O bond was formed due to the elongated bond length of Ga on the surface. In the atomic layer deposition process of the Al2O3 layer using H2O vapor, the reactions caused at the interface were more remarkable for p-GaN. If unintentional oxidation can be resulted in the generation of the defects at the MOS interface, these results suggest that oxidant gases other than H2O and O2 should be used to avoid uncontrollable oxidation on GaN surfaces.

NMR-TS: de novo molecule identification from NMR spectra.

Nuclear magnetic resonance (NMR) spectroscopy is an effective tool for identifying molecules in a sample. Although many previously observed NMR spectra are accumulated in public databases, they cover only a tiny fraction of the chemical space, and molecule identification is typically accomplished manually based on expert knowledge. Herein, we propose NMR-TS, a machine-learning-based python library, to automatically identify a molecule from its NMR spectrum. NMR-TS discovers candidate molecules whose NMR spectra match the target spectrum by using deep learning and density functional theory (DFT)-computed spectra. As a proof-of-concept, we identify prototypical metabolites from their computed spectra. After an average 5451 DFT runs for each spectrum, six of the nine molecules are identified correctly, and proximal molecules are obtained in the other cases. This encouraging result implies that de novo molecule generation can contribute to the fully automated identification of chemical structures. NMR-TS is available at https://github.com/tsudalab/NMR-TS.

Theoretical study on photo-induced processes of 1-methyl-3-(N-(1,8-naphthalimidyl)ethyl)imidazolium halide species: an application of constrained density functional theory.

1-Methyl-3-(N-(1,8-naphthalimidyl)ethyl)imidazolium (MNEI) has potential as a versatile sensor that can measure the electronegativity of anions based on the fluorescence intensity upon irradiation. To clarify the factors that determine the fluorescence intensity, constrained density functional theory (CDFT) was applied to explore the electron transfer (ET) states of MNEI halide species (MNEI-X; X = F, Cl, Br, I). According to the CDFT potential energy surface, intra-molecular ET (SM1) states on MNEI are responsible for the intensity of absorption and fluorescence spectra. However, inter-molecular ET (SET) states between MNEI and X are certainly responsible for fluorescence quenching. Hence, the energetic difference between the SM1 state and the SET state (ΔEM1_ET) is a crucial factor that determines the fluorescence intensity in the spectra of MNEI-X complexes. ΔEM1_ET decreases as the electronegativity of X decreases (i.e., F > Cl > Br > I). This explains the fluorescence intensity of MNEI-X.

Manipulation of fullerene superstructures by complexing with polycyclic aromatic compounds.

Polycyclic aromatic compounds (naphthalene, anthracene and pyrene) have been intercalated into the superstructures of fullerene nanowhiskers, using a facile liquid-liquid interfacial precipitation (LLIP) method. Due to the interaction between polycyclic molecules and fullerene, the growth of fullerene crystals was interfered in comparison to the fullerene crystal growth without the polycyclic molecules, resulting in the formation of fullerene superstructures with various nanofeatures. Moreover, the fluorescence emissions of the fullerene superstructures were significantly changed due to the intercalation of the polycyclic molecules, implying the influence of molecular packing on the electron transfer within the nanostructures. These results may bring new insights on the control of fullerene nanostructures and to manipulate their optical properties in optoelectronic devices.

Multi-spin-state at a Li3PO4/LiCoO2 (104) interface.

We have found the disproportion between the intermediate spin (IS) and low spin (LS) configurations of Co atoms at a Li3PO4/LiCoO2 (104) interface through density functional molecular dynamics (DF-MD). The manifold of the spin state at the interface, however, does not affect the band alignment between the Li3PO4 and LiCoO2 regions.

First-principles analysis on role of spinel (111) phase boundaries in Li4+3xTi5O12 Li-ion battery anodes.

The practical anode material Li4+3xTi5O12 is known to undergo a two-phase separation into Li7Ti5O12 and Li4Ti5O12 during charging/discharging. This phase-separated Li4+3xTi5O12 exhibits electron conduction, although individual phases are expected to be insulators. To elucidate the role played by spinel (111) phase boundaries on these physical properties, first principles calculations were carried out using the GGA+U method. Two-phase Li7Ti5O12/Li4Ti5O12 models are found to exhibit metallic characteristics near their phase boundaries. These boundaries provide conduction paths not only for electrons, but also for Li ions. Judging from the formation energy of Li vacancies/interstitials, the phase boundaries preferentially uptake or release Li via in-plane conduction and then continuously shift in a direction perpendicular to the phase boundary planes. The continuous phase boundary shift leads to a constant electrode potential. A three-dimensional network of cubic {111} planes may contribute to smooth electrochemical reactions.

Intermolecular electron transfer states of 1-methyl-3-(N-(1,8-naphthalimidyl)ethyl)imidazolium iodide obtained by constrained density functional theory.

Electron transfer (ET) states of 1-methyl-3-(N-(1,8-naphthalimidyl)ethyl)imidazolium iodide are responsible for its photophysics. Investigation of an ET state based on constrained density functional theory (CDFT) revealed that nonradiative decay from the ET excited state is mediated by the interaction of the iodine atom with the 1,8-naphthalimide or the imidazolium group.

Theoretical studies of molecular orientation and charge recombination in poly-paraphenylenevinylene light-emitting diodes.

Poly-paraphenylenevinylene (PPV), a material used in organic light-emitting diodes (OLEDs), for which improving the efficiency is an important issue. In general, the molecular orientations of organic compounds in the crystal form are an essential factor determining electron and hole transfer, which are closely related to the efficiency of OLEDs. We have investigated the effects of the rotation of each molecule and the intermolecular distance in the dimer system of PPV, which consists of donor and acceptor molecules, on its charge-recombination process by performing constrained density functional calculations. Starting from the structure of the crystal, it was clarified that the rotation of the donor decreases the charge-recombination factor, to nearly zero, while that of the acceptor increases it to about 10(6) s(-1). We found that this is caused by the repulsive interaction between the donor and acceptor molecules and the formation of a transport pathway resulting from the acceptor rotation.

Theoretical study of singlet oxygen molecule generation via an exciplex with valence-excited thiophene.

Singlet-oxygen [O2((1)Δg)] generation by valence-excited thiophene (TPH) has been investigated using multireference Møller-Plesset second-order perturbation (MRMP2) theory of geometries optimized at the complete active space self-consistent field (CASSCF) theory level. Our results indicate that triplet TPH(1(3)B2) is produced via photoinduced singlet TPH(2(1)A1) because 2(1)A1 TPH shows a large spin-orbit coupling constant with the first triplet excited state (1(3)B2). The relaxed TPH in the 1(3)B2 state can form an exciplex with O2((3)Σg(-)) because this exciplex is energetically more stable than the relaxed TPH. The formation of the TPH(1(3)B2) exciplex with O2((3)Σg(-)) whose total spin multiplicity is triplet (T1 state) increases the likelihood of transition from the T1 state to the singlet ground or first excited singlet state. After the transition, O2((1)Δg) is emitted easily although the favorable product is that from a 2 + 4 cycloaddition reaction.

Micrometer-level naked-eye detection of caesium particulates in the solid state.

Large amounts of radioactive material were released from the Fukushima Daiichi nuclear plant in Japan, contaminating the local environment. During the early stages of such nuclear accidents, iodine I-131 (half-life 8.02 d) is usually detectable in the surrounding atmosphere and bodies of water. On the other hand, in the long-term, soil and water contamination by Cs-137, which has a half-life of 30.17 years, is a serious problem. In Japan, the government is planning and carrying out radioactive decontamination operations not only with public agencies but also non-governmental organizations, making radiation measurements within Japan. If caesium (also radiocaesium) could be detected by the naked eye then its environmental remediation would be facilitated. Supramolecular material approaches, such as host-guest chemistry, are useful in the design of high-resolution molecular sensors and can be used to convert molecular-recognition processes into optical signals. In this work, we have developed molecular materials (here, phenols) as an optical probe for caesium cation-containing particles with implementation based on simple spray-on reagents and a commonly available fluorescent lamp for naked-eye detection in the solid state. This chemical optical probe provides a higher spatial resolution than existing radioscopes and gamma-ray cameras.

Koopmans' Theorem-Compliant Long-Range Corrected (KTLC) Density Functional Mediated by Black-Box Optimization and Data-Driven Prediction for Organic Molecules.

Density functional theory (DFT) is a significant computational tool that has substantially influenced chemistry, physics, and materials science. DFT necessitates parametrized approximation for determining an expected value. Hence, to predict the properties of a given molecule using DFT, appropriate parameters of the functional should be set for each molecule. Herein, we optimize the parameters of range-separated functionals (LC-BLYP and CAM-B3LYP) via Bayesian optimization (BO) to satisfy Koopmans' theorem. Our results demonstrate the effectiveness of the BO in optimizing functional parameters. Particularly, Koopmans' theorem-compliant LC-BLYP (KTLC-BLYP) shows results comparable to the experimental UV-absorption values. Furthermore, we prepared an optimized parameter dataset of KTLC-BLYP for over 3000 molecules through BO for satisfying Koopmans' theorem. We have developed a machine learning model on this dataset to predict the parameters of the LC-BLYP functional for a given molecule. The prediction model automatically predicts the appropriate parameters for a given molecule and calculates the corresponding values. The approach in this paper would be useful to develop new functionals and to update the previously developed functionals.

Revealing factors influencing polymer degradation with rank-based machine learning.

The efficient treatment of polymer waste is a major challenge for marine sustainability. It is useful to reveal the factors that dominate the degradability of polymer materials for developing polymer materials in the future. The small number of available datasets on degradability and the diversity of their experimental means and conditions hinder large-scale analysis. In this study, we have developed a platform for evaluating the degradability of polymers that is suitable for such data, using a rank-based machine learning technique based on RankSVM. We then made a ranking model to evaluate the degradability of polymers, integrating three datasets on the degradability of polymers that are measured by different means and conditions. Analysis of this ranking model with a decision tree revealed factors that dominate the degradability of polymers.

Selective Detection of Toxic C1 Chemicals Using a Hydroxylamine-Based Chemiresistive Sensor Array.

Formaldehyde (FA) is a deleterious C1 pollutant commonly found in the interiors of modern buildings. C1 chemicals are generally more toxic than the corresponding C2 chemicals, but the selective discrimination of C1 and C2 chemicals using simple sensory systems is usually challenging. Here, we report the selective detection of FA vapor using a chemiresistive sensor array composed of modified hydroxylamine salts (MHAs, ArCH2ONH2·HCl) and single-walled carbon nanotubes (SWCNT). By screening 32 types of MHAs, we have identified an ideal sensor array that exhibits a characteristic response pattern for FA. Thus, trace FA (0.02-0.05 ppm in air) can be clearly discriminated from the corresponding C2 chemical, acetaldehyde (AA). This system has been extended to discriminate methanol (C1) from ethanol (C2) in combination with the catalytic conversion of these alcohols to their corresponding aldehydes. Our system offers portable and reliable chemical sensors that discriminate the subtle differences between C1 and C2 chemicals, enabling advanced environmental monitoring and healthcare applications.

Integrating Incompatible Assay Data Sets with Deep Preference Learning.

A large amount of bioactivity assay data is already accumulated in public databases, but the integration of these data sets for quantitative structure-activity relationship (QSAR) studies is not straightforward due to differences in experimental methods and settings. We present an efficient deep-learning-based approach called Deep Preference Data Integration (DPDI). For integrating outcome variables of different assay types, a surrogate variable is introduced, and a neural network is trained such that the total order induced by the surrogate variable is maximally consistent with given data sets. In a task of predicting efficacy of factor Xa inhibitors, DPDI successfully integrated 2959 molecules distributed in 129 assay data sets. In most of our experiments, data integration improved prediction accuracy strongly in interpolation and extrapolation tasks, indicating that DPDI is an effective tool for QSAR studies.

De novo creation of a naked eye-detectable fluorescent molecule based on quantum chemical computation and machine learning.

Designing fluorescent molecules requires considering multiple interrelated molecular properties, as opposed to properties that straightforwardly correlated with molecular structure, such as light absorption of molecules. In this study, we have used a de novo molecule generator (DNMG) coupled with quantum chemical computation (QC) to develop fluorescent molecules, which are garnering significant attention in various disciplines. Using massive parallel computation (1024 cores, 5 days), the DNMG has produced 3643 candidate molecules. We have selected an unreported molecule and seven reported molecules and synthesized them. Photoluminescence spectrum measurements demonstrated that the DNMG can successfully design fluorescent molecules with 75% accuracy (n = 6/8) and create an unreported molecule that emits fluorescence detectable by the naked eye.