Bayesian Optimization for Calibrating and Selecting Hybrid-Density Functional Models

Machine learning for accuracy in density functional approximations

Machine learning techniques have found their way into computational chemistry as indispensable tools to accelerate atomistic simulations and materials design. In addition, machine learning approaches hold the potential to boost the predictive power of computationally efficient electronic structure methods, such as density functional theory, to chemical accuracy and to correct for fundamental errors in density functional approaches. Here, recent progress in applying machine learning to improve the accuracy of density functional and related approximations is reviewed. Promises and challenges in devising machine learning models transferable between different chemistries and materials classes are discussed with the help of examples applying promising models to systems far outside their training sets.

Janus MoSH/WSi2N4 van der Waals Heterostructure: Two-Dimensional Metal/Semiconductor Contact

Constructing heterostructures from already synthesized two-dimensional materials is of significant importance. We performed a first-principles study to investigate the electronic properties and interfacial characteristics of Janus MoSH/WSi2N4 van der Waals heterostructure (vdWH) contacts. We demonstrate that the p-type Schottky formed by MoSH/WSi2N4 and MoHS/WSi2N4 has extremely low Schottky barrier heights (SBHs). Due to its excellent charge injection efficiency, Janus MoSH may be regarded as an effective metal contact for WSi2N4 semiconductors. Furthermore, the interfacial characteristics and electronic structure of Janus MoSH/WSi2N4 vdWHs can not only reduce/eliminate SBH, but also forms the transition from p-ShC to n-ShC type and from Schottky contact (ShC) to Ohmic contact (OhC) through the layer spacing and electric field. Our results can offer a fresh method for optoelectronic applications based on metal/semiconductor Janus MoSH/WSi2N4 vdW heterostructures, which have strong potential in optoelectronic applications.

In Silico Chemical Experiments in the Age of AI: From Quantum Chemistry to Machine Learning and Back

Computational chemistry is an indispensable tool for understanding molecules and predicting chemical properties. However, traditional computational methods face significant challenges due to the difficulty of solving the Schrödinger equations and the increasing computational cost with the size of the molecular system. In response, there has been a surge of interest in leveraging artificial intelligence (AI) and machine learning (ML) techniques to in silico experiments. Integrating AI and ML into computational chemistry increases the scalability and speed of the exploration of chemical space. However, challenges remain, particularly regarding the reproducibility and transferability of ML models. This review highlights the evolution of ML in learning from, complementing, or replacing traditional computational chemistry for energy and property predictions. Starting from models trained entirely on numerical data, a journey set forth toward the ideal model incorporating or learning the physical laws of quantum mechanics. This paper also reviews existing computational methods and ML models and their intertwining, outlines a roadmap for future research, and identifies areas for improvement and innovation. Ultimately, the goal is to develop AI architectures capable of predicting accurate and transferable solutions to the Schrödinger equation, thereby revolutionizing in silico experiments within chemistry and materials science.

Multilevel quantum mechanical calculations show the role of promoter molecules in the dehydration of methanol to dimethyl ether in H-ZSM-5

Methyl carboxylate esters promote the formation of dimethyl ether (DME) from the dehydration of methanol in H-ZSM-5 zeolite. We employ a multilevel quantum method to explore the possible associative and dissociative mechanisms in the presence, and absence, of six methyl ester promoters. This hybrid method combines density functional theory, with dispersion corrections (DFT-D3), for the full periodic system, with second-order Møller-Plesset perturbation theory (MP2) for small clusters representing the reaction site, and coupled cluster with single, double, and perturbative triple substitution (CCSD(T)) for the reacting molecules. The calculated adsorption enthalpy of methanol, and reaction enthalpies of the dehydration of methanol to DME within H-ZSM-5, agree with experiment to within chemical accuracy (∼4 kJ mol-1). For the promoters, a reaction pathway via an associative mechanism gives lower overall reaction enthalpies and barriers compared to the reaction with methanol only. Each stage of this mechanism is explored and related to experimental data. We provide evidence that suggests the promoter's adsorption to the Brønsted acid site is the most important factor dictating its efficiency.

Tailoring of High-Valent Sn-Doped Porous Na3V2(PO4)3/C Nanoarchitechtonics: An Ultra High-Rate Cathode for Sodium-Ion Batteries

NASICON structured Na3V2(PO4)3 (NVP) has captured enormous attention as a potential cathode for next-generation sodium-ion batteries (SIBs), owing to its sturdy crystal structure and high theoretical capacity. Nonetheless, its poor intrinsic electronic conductivity has led to inferior electrochemical performance in terms of rate capability and long cycling performance. To address this problem, a combined strategy is adopted, such as (1) carbon coating and (2) high valent Sn4+ ion doping in the lattice site of vanadium in the NVP cathode. Carbon coating can effectively enhance the surface electronic conductivity, wherein high-valent Sn4+ improves the bulk intrinsic electronic conductivity of the materials. Moreover, Sn is a well-known alloying/dealloying type anode for SIBs; thus, doping of such metal in cathode materials will assume the role of structure stabilizing pillars and establishing high-performing cathode materials. Herein, Na3V2-xSnx(PO4)3/C (denoted as Sn(x)-NVP/C, where x = 0.00, 0.03, 0.05, 0.07, 0.1) were synthesized via sol-gel route, followed by calcination at 800 °C. XRD, Raman, XPS, and electron microscopy data confirmed the high purity of the synthesized cathode. The optimized Sn(0.07)-NVP/C exhibited excellent electrochemical performance in terms of high rate capability and long cycling performance, a high appreciable capacity of 98 mAh g-1 with capacity retention of 85% after 500 cycles. Similarly, at a high current of 20C, it is still able to deliver a stable capacity of 76 mAh g-1 with 85% capacity retention after 3000 cycles. The rate capability study indicates the high current tolerance of Sn(0.07)-NVP/C up to 70 C with a capacity delivery of 55 mAh g-1. It is worth mentioning that CV and EIS analysis for Sn(0.07)-NVP/C cathode displayed minimum voltage polarization and enhanced diffusion coefficient. Moreover, DFT calculation also proved that the electronic and ionic conductivity of NVP is promoted by Sn doping. Hence, the present results demonstrated that Sn(0.07)-NVP/C is considered a promising cathode for sodium-ion battery application.

Reaction Mechanism and Performance of Innovative 2D Germanane-Silicane Alloys: SixGe1- xH Electrodes in Lithium-Ion Batteries

The adjustable structures and remarkable physicochemical properties of 2D monoelemental materials, such as silicene and germanene, have attracted significant attention in recent years. They can be transformed into silicane (SiH) and germanane (GeH) through covalent functionalization via hydrogen atom termination. However, synthesizing these materials with a scalable and low-cost fabrication process to achieve high-quality 2D SiH and GeH poses challenges. Herein, groundbreaking 2D SiH and GeH materials with varying compositions, specifically Si0.25Ge0.75H, Si0.50Ge0.50H, and Si0.75Ge0.25H, are prepared through a simple and efficient chemical exfoliation of their Zintl phases. These 2D materials offer significant advantages, including their large surface area, high mechanical flexibility, rapid electron mobility, and defect-rich loose-layered structures. Among these compositions, the Si0.50Ge0.50H electrode demonstrates the highest discharge capacity, reaching up to 1059 mAh g-1 after 60 cycles at a current density of 75 mA g-1. A comprehensive ex-situ electrochemical analysis is conducted to investigate the reaction mechanisms of lithiation/delithiation in Si0.50Ge0.50H. Subsequently, an initial assessment of the c-Li15(SixGe1- x)4 phase after lithiation and the a-Si0.50Ge0.50 phase after delithiation is presented. Hence, this study contributes crucial insights into the (de)lithiation reaction mechanisms within germanane-silicane alloys. Such understanding is pivotal for mastering promising materials that amalgamate the finest properties of silicon and germanium.

Inserting auxeticity into graphene oxide via bottom-up strategy

Carbon-based materials that process a wide bandgap, high mechanical performance, thermal stability and adjustable characteristics are in high demand. Auxeticity is one of the factors that helps enhances the mechanical performance. Based on this concept, two stable layered carbon-based materials, namely α-C2O and β-C2O, are proposed. A new mechanism (multi-directional negative Poisson's ratio (NPR) effect) is induced, which is attributed to the interaction of modified pz orbitals between interfacial layers. This effect introduces high mechanical properties into materials. Besides, all layered materials are ultrawide bandgap semiconductors, which endows them comparable dielectric properties to those of diamond. Furthermore, α-BK-C2O would maintain its configuration over 2000 K, thereby guaranteeing extremely high thermodynamic stability. So far, these advantages suggested that these carbon-based layer materials could be used in nanoelectronics, especially in electromechanical devices.

Ultra-Low-Loaded Platinum Bonded Hexagonal Boron Nitride as Stable Electrocatalyst for Hydrogen Generation

Chemical stability of hexagonal boron nitride (hBN) ultrathin layers in harsh electrolytes and the availability of nitrogen site in hBN to stabilize metals like Pt are used here to develop a high intrinsic activity hydrogen evolution reaction (HER) catalyst having low loaded Pt (5 weight% or <1 atomic%). A catalyst having a nonzero oxidation state for Pt (with a Pt-N bonding) is shown to be HER active even with low catalyst loadings (0.114 mgcm-2). Electronic modification of the shear exfoliated hBN sheets is achieved by Au nanoparticle-based surface decoration (hBN_Au), and further anchoring with Pt develops a catalyst (hBN_Au_Pt) with high turnover frequency for HER (∼15). The hBN_Au_Pt is shown to be a highly durable catalyst even after the accelerated durability test for 10000 cycles and temperature annealing at 100 °C. Density functional theory based calculations gave insights in to the electronic modifications of hBN with Au and the catalytic activity of the hBN_Au_Pt system, in line with the experimental studies, indicating the demonstration of a new class of catalyst system devoid of issues such as carbon corrosion and Pt leaching.

On Accelerating Substrate Optimization Using Computational Gibbs Energy Barriers: A Numerical Consideration Utilizing a Computational Data Set

Substrate optimization is a time- and resource-consuming step in organic synthesis. Recent advances in chemo- and materials-informatics provide systematic and efficient procedures utilizing tools such as Bayesian optimization (BO). This study explores the possibility of reducing the required experiments further by utilizing computational Gibbs energy barriers. To thoroughly validate the impact of using computational Gibbs energy barriers in BO-assisted substrate optimization, this study employs a computational Gibbs energy barrier data set in the literature and performs an extensive numerical investigation virtually regarding the Gibbs energy barriers as virtual experimental results and those with systematic and random noises as virtual computational results. The present numerical investigation shows that even the computational reactivity affected by noises of as much as 20 kJ/mol helps reduce the number of required experiments.

Fast cycling of lithium metal in solid-state batteries by constriction-susceptible anode materials

Interface reaction between lithium (Li) and materials at the anode is not well understood in an all-solid environment. This paper unveils a new phenomenon of constriction susceptibility for materials at such an interface, the utilization of which helps facilitate the design of an active three-dimensional scaffold to host rapid plating and stripping of a significant amount of a thick Li metal layer. Here we focus on the well-known anode material silicon (Si) to demonstrate that, rather than strong Li-Si alloying at the conventional solid-liquid interface, the lithiation reaction of micrometre-sized Si can be significantly constricted at the solid-solid interface so that it occurs only at thin surface sites of Si particles due to a reaction-induced, diffusion-limiting process. The dynamic interaction between surface lithiation and Li plating of a family of anode materials, as predicted by our constrained ensemble computational approach and represented by Si, silver (Ag) and alloys of magnesium (Mg), can thus more homogeneously distribute current densities for the rapid cycling of Li metal at high areal capacity, which is important in regard to solid-state battery application.

Laser-Driven Rapid Synthesis of Metal-Organic Frameworks and Investigation of UV-NIR Optical Absorption, Luminescence, Photocatalytic Degradation, and Gas and Ion Adsorption Properties

In this study, we designed a platform based on a laser-driven approach for fast, efficient, and controllable MOF synthesis. The laser irradiation method was performed for the first time to synthesize Zn-based MOFs in record production time (approximately one hour) compared to all known MOF production methods with comparable morphology. In addition to well-known structural properties, we revealed that the obtained ZnMOFs have a novel optical response, including photoluminescence behavior in the visible range with nanosecond relaxation time, which is also supported by first-principles calculations. Additionally, photocatalytic degradation of methylene blue with ZnMOF was achieved, degrading the 10 ppm methylene blue (MB) solution 83% during 1 min of irradiation time. The application of laser technology can inspire the development of a novel and competent platform for a fast MOF fabrication process and extend the possible applications of MOFs to miniaturized optoelectronic and photonic devices.

In Situ High-Temperature Raman Spectroscopy of UCl3: A Combined Experimental and Theoretical Study

Uranium trichloride (UCl3) has received growing interest for its use in uranium-fueled molten salt reactors and in the pyrochemical processing of used fuel. In this paper, we report for the first time the experimentally determined Raman spectra of UCl3, at both ambient condition and in situ high temperatures up to 871 K. The frequencies of five of the Raman-active vibrational modes (vi) of UCl3 exhibit a negative temperature derivative ((∂νi/∂T)P) with increasing temperature. This red-shift behavior is likely due to the elongation of U-Cl bonds. The average isobaric mode Grüneisen parameter (γiP = 0.91 ± 0.02) of UCl3 was determined through use of the coefficient of thermal expansion published in Vogel et al. (2021) and the (∂νi/∂T)P values determined in this study. These results are in general agreement with those calculated here by density functional theory (DFT+U). Finally, a comparison of the ambient band positions of UCl3 to those of isostructural lanthanide (La-Eu) and actinide chlorides (Am-Cf) has been made.

Enhancing Performance of Ultraviolet C Photodetectors Through Single-Domain Epitaxy of Monoclinic β-Ga2 O3 Films and Tailored Anti-Reflection Coating

Implementing high-performance ultraviolet C photodetectors (UVC PDs) based on β-Ga2 O3 films is challenging owing to the anisotropic crystal symmetry between the epitaxial films and substrates. In this study, highly enhanced state-of-the-art photoelectrical performance is achieved using single-domain epitaxy of monoclinic β-Ga2 O3 films on a hexagonal sapphire substrate. Unlike 3D β-Ga2 O3 films with twin domains, 2D β-Ga2 O3 films exhibit a single domain with a smooth surface and low concentration of point defects, which enable efficient charge separation by suppressing boundary-induced recombination. Furthermore, a tailored anti-reflection coating (ARC) is adopted as a light-absorbing medium to improve charge generation. The tailored nanostructure, which features a gradient refractive index, not only substantially reduces the reflection, but also suppresses the surface leakage current as a passivation layer. This study provides fundamental insights into the single-domain epitaxy of β-Ga2 O3 films and the application of ARC for the development of high-performance UVC PDs.

Structural Evolution of Air-Exposed Layered Oxide Cathodes for Sodium-Ion Batteries: An Example of Ni-doped NaxMnO2

Sodium-ion batteries have recently aroused the interest of industries as possible replacements for lithium-ion batteries in some areas. With their high theoretical capacities and competitive prices, P2-type layered oxides (NaxTMO2) are among the obvious choices in terms of cathode materials. On the other hand, many of these materials are unstable in air due to their reactivity toward water and carbon dioxide. Here, Na0.67Mn0.9Ni0.1O2 (NMNO), one of such materials, has been synthesized by a classic sol-gel method and then exposed to air for several weeks as a way to allow a simple and reproducible transition toward a Na-rich birnessite phase. The transition between the anhydrous P2 to the hydrated birnessite structure has been followed via periodic XRD analyses, as well as neutron diffraction ones. Extensive electrochemical characterizations of both pristine NMNO and the air-exposed one vs sodium in organic medium showed comparable performances, with capacities fading from 140 to 60 mAh g-1 in around 100 cycles. Structural evolution of the air-exposed NMNO has been investigated both with ex situ synchrotron XRD and Raman. Finally, DFT analyses showed similar charge compensation mechanisms between P2 and birnessite phases, providing a reason for the similarities between the electrochemical properties of both materials.

Design and Realization of Ni Clusters in MoS2@Ni/RGO Catalysts for Alkaline Efficient Hydrogen Evolution Reaction

Due to their almost zero relative hydrogen atom adsorption-free energy, MoS2-based materials have received substantial study. However, their poor electronic conductivity and limited number of catalytic active sites hinder their widespread use in hydrogen evolution reactions. On the other hand, metal clusters offer numerous active sites. In this study, by loading Ni metal clusters on MoS2 and combining them with the better electrical conductivity of graphene, the overpotential of the hydrogen evolution reaction was reduced from 165 mV to 92 mV at 10 mA·cm-2. This demonstrates that a successful method for effectively designing water decomposition is the use of synergistic interactions resulting from interfacial electron transfer between MoS2 and Ni metal clusters.

Lattice strain controlled Ni@NiCu efficient anode catalysts for direct borohydride fuel cells

We successfully fabricated a novel tensile lattice strained Ni@NiCu catalyst with a popcorn-like morphology, which is composed of a crystalline Ni core and a NiCu alloy shell. It exhibits outstanding catalytic activity, selectivity, and stability towards borohydride electrooxidation. Moreover, a direct borohydride fuel cell (DBFC) with a Ni@NiCu anode can deliver a power density of 433 mW cm-2 and an open circuit voltage of 1.94 V, much better than the performances of DBFCs employing other anode catalysts reported in the literature. This could be attributed to the fact that the tensile lattice strain generated by the introduction of Cu leads to a rise in the d-band center of the Ni metal and promotes the final B-H decoupling, which is the rate-determining step in the borohydride oxidation reaction, thus improving remarkably the catalytic performances of Ni@NiCu.

Engineering ligand reactivity enables high-temperature operation of stable perovskite solar cells

Perovskite solar cells (PSCs) consisting of interfacial two- and three-dimensional heterostructures that incorporate ammonium ligand intercalation have enabled rapid progress toward the goal of uniting performance with stability. However, as the field continues to seek ever-higher durability, additional tools that avoid progressive ligand intercalation are needed to minimize degradation at high temperatures. We used ammonium ligands that are nonreactive with the bulk of perovskites and investigated a library that varies ligand molecular structure systematically. We found that fluorinated aniliniums offer interfacial passivation and simultaneously minimize reactivity with perovskites. Using this approach, we report a certified quasi-steady-state power-conversion efficiency of 24.09% for inverted-structure PSCs. In an encapsulated device operating at 85°C and 50% relative humidity, we document a 1560-hour T₈₅ at maximum power point under 1-sun illumination.

3D integral imaging depth estimation of partially occluded objects using mutual information and Bayesian optimization

Integral imaging (InIm) is useful for passive ranging and 3D visualization of partially-occluded objects. We consider 3D object localization within a scene and in occlusions. 2D localization can be achieved using machine learning and non-machine learning-based techniques. These techniques aim to provide a 2D bounding box around each one of the objects of interest. A recent study uses InIm for the 3D reconstruction of the scene with occlusions and utilizes mutual information (MI) between the bounding box in this 3D reconstructed scene and the corresponding bounding box in the central elemental image to achieve passive depth estimation of partially occluded objects. Here, we improve upon this InIm method by using Bayesian optimization to minimize the number of required 3D scene reconstructions. We evaluate the performance of the proposed approach by analyzing different kernel functions, acquisition functions, and parameter estimation algorithms for Bayesian optimization-based inference for simultaneous depth estimation of objects and occlusion. In our optical experiments, mutual-information-based depth estimation with Bayesian optimization achieves depth estimation with a handful of 3D reconstructions. To the best of our knowledge, this is the first report to use Bayesian optimization for mutual information-based InIm depth estimation.

Catalytic Ozonation of Polluter Benzene from -20 to >50 °C with High Conversion Efficiency and Selectivity on Mullite YMn2O5

Catalytic decomposition of aromatic polluters at room temperature represents a green route for air purification but is currently challenged by the difficulty of generating reactive oxygen species (ROS) on catalysts. Herein, we develop a mullite catalyst YMn₂O₅ (YMO) with dual active sites of Mn³⁺ and Mn⁴⁺ and use ozone to produce a highly reactive O* upon YMO. Such a strong oxidant species on YMO shows complete removal of benzene from -20 to >50 °C with a high CO_x selectivity (>90%) through the generated reactive species O* on the catalyst surface (60 000 mL g^-1 h^-1). Although the accumulation of water and intermediates gradually lowers the reaction rate after 8 h at 25 °C, a simple treatment by ozone purging or drying in the ambient environment regenerates the catalyst. Importantly, when the temperature increases to 50 °C, the catalytic performance remains 100% conversion without any degradation for 30 h. Experiments and theoretical calculations show that such a superior performance stems from the unique coordination environment, which ensures high generation of ROS and adsorption of aromatics. Mullite's catalytic ozonation degradation of total volatile organic compounds (TVOC) is applied in a home-developed air cleaner, resulting in high efficiency of benzene removal. This work provides insights into the design of catalysts to decompose highly stable organic polluters.

Modulating the Combinatorial Target Power of MgSnN2 via RF Magnetron Sputtering for Enhanced Optoelectronic Performance: Mechanistic Insights from DFT Studies

The unique structural features of many ternary nitride materials with strong chemical bonding and band gaps above 2.0 eV are limited and are experimentally unexplored. It is important to identify candidate materials for optoelectronic devices, particularly for light-emitting diodes (LEDs) and absorbers in tandem photovoltaics. Here, we fabricated MgSnN₂ thin films, as promising II-IV-N₂ semiconductors, on stainless-steel, glass, and silicon substrates via combinatorial radio-frequency magnetron sputtering. The structural defects of the MgSnN₂ films were studied as a function of the Sn power density, while the Mg and Sn atomic ratios remained constant. Polycrystalline orthorhombic MgSnN₂ was grown on the (120) orientation within a wide optical band gap range of ∼2.20-2.17 eV. The carrier densities of 2.18× 10²⁰ to 1.02 × 10²¹ cm^-3, mobilities between 3.75 and 2.24 cm²/Vs, and a decrease in resistivity from 7.64 to 2.73 × 10^-3 Ω cm were confirmed by Hall-effect measurements. These high carrier concentrations suggested that the optical band gap measurements were affected by a Burstein-Moss shift. Furthermore, the electrochemical capacitance properties of the optimal MgSnN₂ film exhibited an areal capacitance of 152.5 mF/cm² at 10 mV/s with high retention stability. The experimental and theoretical results showed that MgSnN₂ films were effective semiconductor nitrides toward the progression of solar absorbers and LEDs.

Conversion between Metavalent and Covalent Bond in Metastable Superlattices Composed of 2D and 3D Sublayers

Reversible conversion over multimillion times in bond types between metavalent and covalent bonds becomes one of the most promising bases for universal memory. As the conversions have been found in metastable states, an extended category of crystal structures from stable states via redistribution of vacancies, research on kinetic behavior of the vacancies is highly in demand. However, it remains lacking due to difficulties with experimental analysis. Herein, the direct observation of the evolution of chemical states of vacancies clarifies the behavior by combining analysis on charge density distribution, electrical conductivity, and crystal structures. Site-switching of vacancies of Sb₂Te₃ gradually occurs with diverged energy barriers owing to their own activation code: the accumulation of vacancies triggers spontaneous gliding along atomic planes to relieve electrostatic repulsion. Studies on the behavior can be further applied to multiphase superlattices composed of Sb₂Te₃ (2D) and GeTe (3D) sublayers, which represent superior memory performances, but their operating mechanisms were still under debate due to their complexity. The site-switching is favorable (suppressed) when Te-Te bonds are formed as physisorption (chemisorption) over the interface between Sb₂Te₃ (2D) and GeTe (3D) sublayers driven by configurational entropic gain (electrostatic enthalpic loss). Depending on the type of interfaces between sublayers, phases of the superlattices are classified into metastable and stable states, where the conversion could only be achieved in the metastable state. From this comprehensive understanding on the operating mechanism via kinetic behaviors of vacancies and the metastability, further studies toward vacancy engineering are expected in versatile materials.

First principles study on thickness dependent structural and electronic properties unveiling the growth and stability of 2D layered II-VI semiconducting compounds

By employing first principles density functional theory calculations on thickness dependent structural and electronic properties of (0001) surface slabs of wurtzite MX compounds, our study demonstrated the possibility of the existence of 2D layered materials from II-VI group traditional semiconducting compounds that are widely used in various fields. Our calculations revealed that (0001) surface slabs of wurtzite ZnO and CdO compounds prefer to stabilize as sp² hybridized - atomically thin graphitic layers as observed in earlier work, which are separated by van der Waals distances, when compared to the respective wurtzite slabs. On the other hand, for surface slabs of other ZnX and CdX (X = S, Se, Te) compounds, sp³ hybridized bilayers, which comprise an X-Zn(Cd)-Zn(Cd)-X structural arrangement, are energetically stable until certain thicknesses of the slabs. Both 2D layered MO and MX systems are electronically insulating in nature. When increasing the number of layers in these systems, the band gap decreases due to the widening of the energy bands. Our calculations further confirmed that all of these 2D systems possess structural, elastic, and lattice dynamical stabilities, depicting their compatibility in optoelectronics and photovoltaic applications, as confirmed by their effective mass and mobility.

A nanoarchitecture of a gold cluster conjugated gold nanorod hybrid system and its application in fluorescence imaging and plasmonic photothermal therapy

Engineering different nanomaterials into a single functional material can impart unique properties of the parental nanoparticles, especially in the field of bio imaging and therapy. Gold nanomaterials having different sizes, shapes and dimensionalities exhibit exceptional properties apart from their non-toxicity and hence are strong candidates in the biomedical field. Designing a hybrid nanomaterial of two gold nanostructures retaining the individual properties of the parental nanomaterials is challenging. Here, we demonstrate the synthesis of a hybrid nanomaterial (GQC@GNR), comprising an extremely small gold nanocluster and a representative of the asymmetric gold nanostructure, i.e., a gold nanorod, both having their own different exclusive optical properties like tuneable emission and NIR absorption characteristics, respectively. The hybrid system is designed to retain its emission and absorption in the NIR region to use it as an agent for simultaneous imaging and therapy. The formation of GQC@GNR and its architectonics heavily depend on the synthesis route and the parameters adopted which in turn have a direct influence on its properties. The architecture and its connection to the optical properties are explained using UV-Vis absorption and photoluminescence spectroscopy, zeta potential, transmission electron microscopy, etc. DFT-based computational modelling supports architectonics as explained by the experimental findings. The formation of the gold-gold hybrid system witnessed interesting science with a strong indication that materials with desired properties can be designed by appropriately modulating the architectonics of hybrid formation. Finally, folate conjugated GQC@GNR demonstrated its efficacy for targeted imaging and photothermal therapy in HeLa cells and tumor-bearing animal models. The detailed therapeutic efficacy of GQC@GNR is also explained based on Raman spectroscopy.

Machine learning, artificial intelligence, and chemistry: how smart algorithms are reshaping simulation and the laboratory

Machine learning and artificial intelligence are increasingly gaining in prominence through image analysis, language processing, and automation, to name a few applications. Machine learning is also making profound changes in chemistry. From revisiting decades-old analytical techniques for the purpose of creating better calibration curves, to assisting and accelerating traditional in silico simulations, to automating entire scientific workflows, to being used as an approach to deduce underlying physics of unexplained chemical phenomena, machine learning and artificial intelligence are reshaping chemistry, accelerating scientific discovery, and yielding new insights. This review provides an overview of machine learning and artificial intelligence from a chemist’s perspective and focuses on a number of examples of the use of these approaches in computational chemistry and in the laboratory.

Machine Learning Accelerated Discovery of Promising Thermal Energy Storage Materials with High Heat Capacity

Thermal energy storage offers numerous benefits by reducing energy consumption and promoting the use of renewable energy sources. Thermal energy storage materials have been investigated for many decades with the aim of improving the overall efficiency of energy systems. However, finding solid materials that meet the requirement of high heat capacity has been a grand challenge for material scientists. Herewith, by training various machine learning models on 3377 high-quality data from full density functional theory (DFT) calculations, we efficiently search for potential materials with high heat capacity. We build four traditional machine learning models and two graph neural network models. Cross-comparison of the prediction performance and model accuracy was conducted among different models. The deeperGATGNN model exhibits high prediction accuracy and is used for predicting the heat capacity of 32,026 structures screened from the open quantum material database. We gain deep insight into the correlation between heat capacity and structure descriptors such as space group, prototype, lattice volume, atomic weight, etc. Twenty-two structures were predicted to possess high heat capacity, and the results were further validated with DFT calculations. We also identified one special structure, namely, MnIn₂Se₄, with space group no. 227 (Fd3̅m), that exhibits extremely high heat capacity, even higher than that of the Dulong-Petit limit at room temperature. This study paves the way for accelerating the discovery of novel thermal energy storage materials by combining machine learning with minimal DFT inquiry.

First-principle study of neutron irradiation induced performance degradation of amorphous porous silica

Neutron irradiation induced degradation of porous silica film is studied by Molecular Dynamics and Density-Functional theory-based methods. The degradation of microscopic structure, thermal property, and optical property of porous silica film are systematically investigated. Low-energy recoil is used to simulate the neutron irradiation effect. The pair and bond angle distributions, and coordination number distributions reveal that, under neutron irradiation, the microscopic structure of porous silica film is obviously modified, and the coordination defects are induced. We find that the higher recoil energy, the more coordination defects are formed in the film. The increased defects lead to a decrease in thermal conductivity. In addition, neutron irradiation induces additional optical absorption peaks in UV region and increasement in refractive index, resulting in a noticeable reduction in light transmittance. The detailed calculation of density of states reveals that these optical absorption peaks originate from the irradiation induced defect states in band gap. Our work shows that low-energy neutron irradiation can induce obvious defect density and degrade thermal and optical properties of porous silica film, which are responsible for subsequent laser-induced damage.

From Termination Dependent Chemical Sensitivity of Spin Orientation in All-bcc Fe/Co Magnetic Superlattices toward the Concept of an Artificial Surface of a Ferromagnet

Adsorption of gases on the surface of all-bcc (Fe/Co)_N superlattices drives the in-plane, 90° magnetization rotation of the bulk-like Fe(110) supporting ferromagnet. Both experimental and theoretical results prove that terminating the surface of (Fe/Co)_N superlattices either by Co or by Fe switches "ON" or "OFF" the spin orientation sensitivity to adsorption. Results indicate that purely surface limited adsorption processes strongly modify the magnetic anisotropy of the entire (Fe/Co)_N superlattice, which acts as a kind of "artificial" surface of the bulky Fe(110) ferromagnet. Such an artificial magnetic surface anisotropy concept not only enhances the surface contribution in classical surface-bulk competition but also provides its additional chemical sensitivity.

Evolving symbolic density functionals

Systematic development of accurate density functionals has been a decades-long challenge for scientists. Despite emerging applications of machine learning (ML) in approximating functionals, the resulting ML functionals usually contain more than tens of thousands of parameters, leading to a huge gap in the formulation with the conventional human-designed symbolic functionals. We propose a new framework, Symbolic Functional Evolutionary Search (SyFES), that automatically constructs accurate functionals in the symbolic form, which is more explainable to humans, cheaper to evaluate, and easier to integrate to existing codes than other ML functionals. We first show that, without prior knowledge, SyFES reconstructed a known functional from scratch. We then demonstrate that evolving from an existing functional ωB97M-V, SyFES found a new functional, GAS22 (Google Accelerated Science 22), that performs better for most of the molecular types in the test set of Main Group Chemistry Database (MGCDB84). Our framework opens a new direction in leveraging computing power for the systematic development of symbolic density functionals.

Construction of a 2D/2D g-C3N5/NiCr-LDH Heterostructure to Boost the Green Ammonia Production Rate under Visible Light Illumination

Photocatalytic N₂ fixation has emerged as one of the most useful ways to produce NH₃, a useful asset for chemical industries and a carbon-free energy source. Recently, significant progress has been made toward designing efficient photocatalysts to achieve this objective. Here, we introduce a highly active type-II heterojunction fabricated via integrating two-dimensional (2D) nanosheets of exfoliated g-C₃N₅ with nickel-chromium layered double hydroxide (NiCr-LDH). With an optimized loading of NiCr-LDH on exfoliated g-C₃N₅, excellent performance is realized for green ammonia synthesis under ambient conditions without any noble metal cocatalyst(s). Indeed, the g-C₃N₅/NiCr-LDH heterostructure with 2 wt % of NiCr-LDH (CN-NCL-2) exhibits an ammonia yield of about 2.523 mmol/g/h, which is about 7.51 and 2.86 times higher than that of solo catalysts, i.e., NiCr-LDH (NC-L) and exfoliated g-C₃N₅ (CN-5), respectively, where methanol is used as a sacrificial agent. The enhancement of NH₃ evolution by the g-C₃N₅/NiCr-LDH heterostructure can be attributed to the efficient charge transfer, a key factor to the photocatalytic N₂ fixation rate enhancement. Additionally, N₂ vacancies present in the system help adsorb N₂ on the surface, which improves the ammonia production rate further. The best-performing heterostructure also shows long-term stability with the NH₃ production rate remaining nearly constant over 20 h, demonstrating the excellent robustness of the photocatalyst.

Enhancing the kinetics of vanadium oxides via conducting polymer and metal ions co-intercalation for high-performance aqueous zinc-ions batteries

Aqueous zinc-ions batteries with low cost, reliable safety, high theoretical specific capacity and eco-friendliness have captured conspicuous attention in large-scale energy storage. However, the developed cathodes often suffer from low electrical conductivity and sluggish Zn²⁺ diffusion kinetics, which severely hampers the development of aqueous zinc-ions batteries. Herein, we successfully prepare Mg/PANI/V₂O₅•nH₂O (MPVO) nanosheets through conducting polymers (polyaniline) and metal ions (Mg²⁺) co-intercalated strategy and systematically explore its electrochemical performance as cathode materials for aqueous zinc-ion batteries. Benefitting from the synergistic effect of polyaniline and Mg²⁺ co-intercalated, the MPVO exhibits larger interlayer spacing and higher electrical conductivity than the single guest intercalation, which significantly enhances the electrochemical kinetics. As a consequence, the MPVO cathodes deliver superior specific capacity, rate capability and long-term cycling performance. Moreover, multiple characterizations and theoretical calculations are executed to expound the relevant mechanism.Therefore, this work provides a novel thought for the design of high-performance cathode materials for aqueous ZIBs.

One-Step Growth of Bilayer 2H-1T' MoTe2 van der Waals Heterostructures with Interlayer-Coupled Resonant Phonon Vibration

2H-1T' MoTe₂ van der Waals heterostructures (vdWHs) have promising applications in optoelectronics due to a seamlessly homogeneous semiconductor-metal coupled interface. However, the existing methods to fabricate such vdWHs involved complicated steps that may deteriorate the interfacial coupling and are also lacking precise thickness control capability. Here, a one-step growth method was developed to controllably grow bilayer 2H-1T' MoTe₂ vdWHs in the small growth window overlapped for both phases. Atomic-resolution low-voltage transmission electron microscopy shows the distinct moiré patterns in the bilayer vdWHs, revealing the epitaxial nature of the top 2H phase with the lattice parameters regulated by the underneath 1T' phase. Such epitaxially stacked bilayer vdWHs modulate the interlayer coupling by resonating their vibration modes, as unveiled by the angle-resolved polarized Raman spectroscopy and first-principles calculations.

Surface-Assisted Synthesis of N-Containing π-Conjugated Polymers

On-surface synthesis has recently emerged as a powerful strategy to design conjugated polymers previously precluded in conventional solution chemistry. Here, an N-containing pentacene-based precursor (tetraazapentacene) is ex-professo synthesized endowed with terminal dibromomethylene (:CBr₂ ) groups to steer homocoupling via dehalogenation on metallic supports. Combined scanning probe microscopy investigations complemented by theoretical calculations reveal how the substrate selection drives different reaction mechanisms. On Ag(111) the dissociation of bromine atoms at room temperature triggers the homocoupling of tetraazapentacene units together with the binding of silver adatoms to the nitrogen atoms of the monomers giving rise to a N-containing conjugated coordination polymer (P1). Subsequently, P1 undergoes ladderization at 200 °C, affording a pyrrolopyrrole-bridged conjugated polymer (P2). On Au(111) the formation of the intermediate polymer P1 is not observed and, instead, after annealing at 100 °C, the conjugated ladder polymer P2 is obtained, revealing the crucial role of metal adatoms on Ag(111) as compared to Au(111). Finally, on Ag(100) the loss of :CBr2 groups affords the formation of tetraazapentacene monomers, which coexist with polymer P1. Our results contribute to introduce protocols for the synthesis of N-containing conjugated polymers, illustrating the selective role of the metallic support in the underlying reaction mechanisms.

Role of Polarons in Single-Atom Catalysts: Case Study of Me1 [Au1, Pt1, and Rh1] on TiO2(110)

The local environment of metal-oxide supported single-atom catalysts plays a decisive role in the surface reactivity and related catalytic properties. The study of such systems is complicated by the presence of point defects on the surface, which are often associated with the localization of excess charge in the form of polarons. This can affect the stability, the electronic configuration, and the local geometry of the adsorbed adatoms. In this work, through the use of density functional theory and surface-sensitive experiments, we study the adsorption of Rh1, Pt1, and Au1 metals on the reduced TiO2(110) surface, a prototypical polaronic material. A systematic analysis of the adsorption configurations and oxidation states of the adsorbed metals reveals different types of couplings between adsorbates and polarons. As confirmed by scanning tunneling microscopy measurements, the favored Pt1 and Au1 adsorption at oxygen vacancy sites is associated with a strong electronic charge transfer from polaronic states to adatom orbitals, which results in a reduction of the adsorbed metal. In contrast, the Rh1 adatoms interact weakly with the excess charge, which leaves the polarons largely unaffected. Our results show that an accurate understanding of the properties of single-atom catalysts on oxide surfaces requires a careful account of the interplay between adatoms, vacancy sites, and polarons.

Supplementary Information

The online version contains supplementary material available at 10.1007/s11244-022-01651-0.

Wearable anti-temperature interference strain sensor with metal nanoparticle thin film and hybrid ligand exchange

Anti-interference characteristics, whereby undesirable signal interference is minimized, are required for multifunctional sensor platforms. In this study, an anti-temperature-interference resistive-type strain sensor, which does not respond to temperature but only to strain, is designed. Anti-interference properties were achieved by modulating the temperature coefficient of resistance (TCR) of metal nanoparticles (NPs) through hybrid chemical treatment with organic and halide ligands that induce negative and positive TCRs, respectively. Consequently, a very low TCR of 1.9 × 10^-5 K^-1 was obtained. To investigate the origin of this near-zero TCR, analyses of correlated electrical, thermal, and mechanical properties were performed in addition to structural characterization and analysis. Density functional theory calculations and electrical percolation modeling were performed to illuminate the transport behavior in the near-zero-TCR NP thin films. Finally, we fabricated a high-performance anti-temperature-interference strain sensor using a solution process. The sensors detect a variety of strains, including those arising from large movements, such as wrist and knee movements, and fine movements, such as artery pulses or movements made during calligraphy, and did not respond to temperature changes.

Investigation of Ordering on Oxygen-Deficient LiNi0.5 Mn1.5 O4-δ Thin Films for Boosting Electrochemical Performance in All-Solid-State Thin-Film Batteries

All-solid-state thin-film batteries (ASSTFBs) are promising next-generation battery systems, but critical challenges such as low-energy-density remain. The low-energy-density might persist with low-voltage cathode material; hence, high-voltage cathode material development is required. While LiNi_0.5 Mn_1.5 O₄ (LNM) has been considered a promising high-voltage cathode material. This study investigates the electrochemical properties of LNM thin films based on the correlation between the ordering of cations (Ni and Mn) and oxygen vacancies (V_O ). The authors find that the cations' order changes from a disordered structure to an ordered structure with an increased oxygen flow rate during deposition. The optimized LNM fabricated using a 60:40 ratio of Ar to O₂ exhibits the highest rate capability (321.4 mAh cm^-3 @ 20 C) and most prolonged cycle performance for 500 cycles. The role of V_O within the LNM structure and the lower activation energy of ordered LNM compared to disordered LNM through first-principles density functional theory calculations is elucidated. The superior electrochemical performance (276.9 mAh cm^-3 @ 0.5 C) and high cyclic performance (at 93.9%, 500 cycles) are corroborated by demonstrating flexible ASSTFB cells using LiPON solid-state electrolyte and thin-film Li anode. This work paves the way for future research on the fabrication of high-performance flexible ASSTFBs.

Improving Density Functional Prediction of Molecular Thermochemical Properties with a Machine-Learning-Corrected Generalized Gradient Approximation

The past decade has seen an increasing interest in designing sophisticated density functional approximations (DFAs) by integrating the power of machine learning (ML) techniques. However, application of the ML-based DFAs is often confined to simple model systems. In this work, we construct an ML correction to the widely used Perdew-Burke-Ernzerhof (PBE) functional by establishing a semilocal mapping from the electron density and reduced gradient to the exchange-correlation energy density. The resulting ML-corrected PBE is immediately applicable to any real molecule and yields significantly improved heats of formation while preserving the accuracy for other thermochemical and kinetic properties. This work highlights the prospect of combining the power of data-driven ML methods with physics-inspired derivations for reaching the heaven of chemical accuracy.

In situ DRIFT studies on N2O formation over Cu-functionalized zeolites during ammonia-SCR

The influence of the zeolite framework structure on the formation of N2O during ammonia-SCR of NOx was studied for three different copper-functionalized zeolite samples, namely Cu-SSZ-13 (CHA), Cu-ZSM-5 (MFI), and Cu-BEA (BEA).

Hydrogen-Bonding 2D Coordination Polymer for Enzyme-Free Electrochemical Glucose Sensing

Regular detection of blood glucose levels is a critical indicator for effective diabetes management. Owing to the intrinsic highly sensitive nature of enzymes, the performance of enzymatic glucose sensors is...

Enhancement of the power factor of SnSe by adjusting the crystal and energy band structures

A high power factor 686 μW m−1 K−2 at 773 K for the SnSe sample origins from temperature dependence of energy valley degeneration and m*DOSvia Ab initio molecular dynamics (AIMD) simulations on basis of the lattice contraction model.

Multi-objective optimization for retinal photoisomerization models with respect to experimental observables

The fitting of physical models is often done only using a single target observable. However, when multiple targets are considered, the fitting procedure becomes cumbersome, there being no easy way to quantify the robustness of the model for all different observables. Here, we illustrate that one can jointly search for the best model for each desired observable through multi-objective optimization. To do so, we construct the Pareto front to study if there exists a set of parameters of the model that can jointly describe multiple, or all, observables. To alleviate the computational cost, the predicted error for each targeted objective is approximated with a Gaussian process model as it is commonly done in the Bayesian optimization framework. We applied this methodology to improve three different models used in the simulation of stationary state cis-trans photoisomerization of retinal in rhodopsin, a significant biophysical process. Optimization was done with respect to different experimental measurements, including emission spectra, peak absorption frequencies for the cis and trans conformers, and energy storage. Advantages and disadvantages of previously proposed models are exposed.

Facile Synthesis of Ti2AC (A = Zn, Al, In, and Ga) MAX Phases by Hydrogen Incorporation into Crystallographic Voids

While using hydride precursors, such as TiH₂, can promote the formation of some MAX phases, the mechanism for this stabilization effect by hydrogen has been unsolved. Herein, we report a facile synthesis method of Ti₂AC (A = Zn, Al, In, and Ga) MAX phases using hydrogen as the phase stabilizer at their crystallographic voids. DFT calculations revealed that hydrogen could be incorporated in the center of the Ti₃A (A = Zn, Al, Ga, and In) cages of Ti₂AC MAX phases. The hydrogen is accommodated as an anion as a result of electron transfer from the surrounding Ti and A to H, leading to the stabilized state through Coulomb interaction between (Ti₃A)^δ+ and H^-. Consequently, high-purity Ti₂AC (A = Zn, Al, Ga, and In) was directly synthesized under pressure-less and milder temperature conditions by simply employing TiH₂ as the precursor. These findings indicate that utilizing hydrogen could be one of the experimental parameters to facilitate the formation of materials having crystallographic voids.

Enhancement of Photoresponse on Narrow-Bandgap Mott Insulator α-RuCl3 via Intercalation

Charge doping to Mott insulators is critical to realize high-temperature superconductivity, quantum spin liquid state, and Majorana fermion, which would contribute to quantum computation. Mott insulators also have a great potential for optoelectronic applications; however, they showed insufficient photoresponse in previous reports. To enhance the photoresponse of Mott insulators, charge doping is a promising strategy since it leads to effective modification of electronic structure near the Fermi level. Intercalation, which is the ion insertion into the van der Waals gap of layered materials, is an effective charge-doping method without defect generation. Herein, we showed significant enhancement of optoelectronic properties of a layered Mott insulator, α-RuCl₃, through electron doping by organic cation intercalation. The electron-doping results in substantial electronic structure change, leading to the bandgap shrinkage from 1.2 eV to 0.7 eV. Due to localized excessive electrons in RuCl₃, distinct density of states is generated in the valence band, leading to the optical absorption change rather than metallic transition even in substantial doping concentration. The stable near-infrared photodetector using electronic modulated RuCl₃ showed 50 times higher photoresponsivity and 3 times faster response time compared to those of pristine RuCl₃, which contributes to overcoming the disadvantage of a Mott insulator as a promising optoelectronic device and expanding the material libraries.

Biocompatible Ruthenium Single-Atom Catalyst for Cascade Enzyme-Mimicking Therapy

Rationally constructing single-atom enzymes (SAEs) with superior activity, robust stability, and good biocompatibility is crucial for tumor therapy but still remains a substantial challenge. In this work, we adopt biocompatible carbon dots as the carrier material to load Ru single atoms, achieving Ru SAEs with superior multiple enzyme-like activity and stability. Ru SAEs behave as oxidase, peroxidase, and glutathione oxidase mimics to synchronously catalyze the generation of reactive oxygen species (ROS) and the depletion of glutathione, thus amplifying the ROS damage and finally causing the death of cancer cells. Notably, Ru SAEs exhibit excellent peroxidase-like activity with a specific activity of 7.5 U/mg, which surpasses most of the reported SAEs and is 20 times higher than that of Ru/C. Theoretical results reveal that the electrons of the Ru 4d orbital in Ru SAEs are transferred to O atoms in H₂O₂ and then efficiently activate H₂O₂ to produce ^•OH. Our work may provide some inspiration for the design of SAEs for cancer therapy.

Effects of quarantine disobedience and mobility restrictions on COVID-19 pandemic waves in dynamical networks

For the global COVID-19 pandemic it is still not adequately understood how quarantine disobedience and change in mobility restrictions influence the pandemic spreading and waves. Here, we propose a new metapopulation epidemiological model as a network composed of equal clusters to predict the course of the epidemic based on the contiguous spreading between the neighbours, the probability of quarantine misbehaviour, and the probability of mobility, which control contacts outside the cluster. We exemplify the model by comparing simulation results with real data on COVID-19 pandemic in Croatia. Fitting the data over the first and second pandemic waves, when the probability of mobility is set by the stringency index, the probability of quarantine misbehaviour is found by a Bayesian optimization yielding a fascinating agreement between the daily COVID-19 deaths and model output and efficiently predicting the timing of pandemic bursts. A sudden increase in the probability of quarantine misbehaviour alongside the sudden increase in the probability of mobility generate the model third wave in good agreement with daily COVID-19 deaths.