Bayesian Optimization for Efficient Prediction of Gas Uptake in Nanoporous Materials.

The discovery and optimization of novel nanoporous materials (NPMs) such as Metal-Organic Frameworks (MOFs) are crucial for addressing global challenges. Traditional experimental approaches for optimizing these materials are time-consuming and resource-intensive. This research paper presents a strategy using Bayesian optimization (BO) to efficiently navigate the NPMs for gas storage. For a MOF dataset drawn from 19 different sources, we present a quantitative evaluation of BO. In our study, we employed machine learning (ML) techniques to conduct regression analysis on many models. Following this, we identified the three ML models that exhibited the highest accuracy, which were subsequently chosen as surrogates in our investigation, including the conventional Gaussian Process (GP) model. We found that GP with expected improvement (EI) as the acquisition function but without a gamma prior which is standard in Bayesian Optimization python library outperforms other surrogate models. Additionally, it should be noted that while the machine learning model that exhibits superior performance in predicting the target variable may be considered the best choice, it may not necessarily serve as the most suitable surrogate model for BO. This observation has significant importance and warrants further investigation. This comprehensive framework accelerates the pace of materials discovery.

Exploring Hydrogen Storage Capacity in Metal-Organic Frameworks: A Bayesian Optimization Approach.

Metal-organic Frameworks (MOFs) can be employed for gas storage, capture, and sensing. Finding the MOF with the best adsorption property from a large database is usual for adsorption calculations. In high-throughput computational research, the expense of computing thermodynamic quantities limits the finding of MOFs for separations and storage. In this work, we demonstrate the usefulness of Bayesian optimization (BO) for estimating the H2 uptake capability of MOFs by using an existing dataset containing 98000 real and hypothetical MOFs. We demonstrate that in order to recover the best candidate MOFs, less than 0.027 % of the database needs to be screened using the BO method. This allows future adsorption experiments on a small sample of MOFs to be undertaken with minimal experimental effort by effectively screening MOF databases. In addition, the presented BO can provide comprehensible material design insights, and the framework will be transferable to optimizing other target properties. We also suggest using Particle Swarm Optimisation (PSO), a swarm intelligence technique in artificial intelligence, to estimate MOFs' H2 uptake potential to achieve results comparable to BO. In addition, we implement a novel modification of PSO called Evolutionary-PSO (EPSO) to compare and find interesting outcomes.

Structural and Electronic Properties of Two-Dimensional Materials: A Machine-Learning-Guided Prediction.

The growing number of studies and interest in two-dimensional (2D) materials has not yet resulted in a wide range of material applications. This is a result of difficulties in getting the properties, which are often determined through numerical experiments or through first-principles predictions, both of which require lots of time and resources. Here we provide a general machine learning (ML) model that works incredibly well as a predictor for a variety of electronic and structural properties such as band gap, fermi level, work function, total energy and area of unit cell for a wide range of 2D materials derived from the Computational 2D Materials Database (C2DB). Our predicted model for classification of samples works extraordinarily well and gives an accuracy of around 99 %. We are able to successfully decrease the number of studied features by employing a strict permutation-based feature selection method along with the sure independence screening and sparsifying operator (SISSO), which further supports the design recommendations for the identification of novel 2D materials with the desired properties.

Acetylene-Mediated Borophosphene Dirac Materials as Efficient Anode Materials for Lithium-Ion Batteries.

Generally, graphynes have been generated by the insertion of acetylenic content (-C≡C-) in the graphene network in different ratios. Also, several aesthetically pleasing architectures of two-dimensional (2D) flatlands have been reported with the incorporation of acetylenic linkers between the heteroatomic constituents. Prompted by the experimental realization of boron phosphide, which has provided new insights on the boron-pnictogen family, we have modelled novel forms of acetylene-mediated borophosphene nanosheets by joining the orthorhombic borophosphene stripes with different widths and with different atomic constituents using acetylenic linkers. Structural stabilities and properties of these novel forms have been assessed using first-principles calculations. Investigation of electronic band structure elucidates that all the novel forms show the linear band crossing closer to the Fermi level at Dirac point with distorted Dirac cones. The linearity in the hole and electronic bands impose the high Fermi velocity to the charge carriers close to that of graphene. Finally, we have also unravelled the propitious features of acetylene-mediated borophosphene nanosheets as anodes in Li-ion batteries.

Deciphering the Role of Substitution in Transition-Metal Phosphorous Trisulfide (100) Surface: A Highly Efficient and Durable Pt-free ORR Electrocatalyst.

The rational design and development of earth-abundant, cost-effective, environmentally benign, and highly robust oxygen reduction reaction (ORR) electrocatalysts can circumvent the obstacles associated with the large-scale commercialization of fuel cells. Here, using first-principles-based density functional theory (DFT), we have computationally screened the potential and feasibility of transition-metal phosphorous trisulfides (TMPS3 ) (100) surfaces as efficient ORR electrocatalyst in acidic fuel cell application. MnPS3 (100) surface emerges to be the best among TMPS3 surfaces with optimal O2 activation resulting in very low overpotential. The study reveals that ORR occurs on the MnPS3 surface via 4e reduction associative pathway where the kinetically rate-determining step (RDS) is the formation of O*+H2 O with an activation barrier of 0.66 eV. Additionally, high CO tolerance and easy desorption of H2 O make MnPS3 a robust catalyst. Substitution in half of the Mn sites of MnPS3 (100) surface with Co considerably enhances the ORR activity. Mn0.5 Co0.5 PS3 (100) surface exhibits an ultralow overpotential of 0.39 V vs RHE switching ORR pathway from associative to dissociative. Spontaneous dissociation of H2 O2 on Mn0.5 Co0.5 PS3 proves 4e reduction pathway excluding 2e one. Electronic structure analysis reveals that pristine MnPS3 (100) surface is a narrow band gap semiconductor which upon Co substitution transforms into a conducting metallic surface enhancing ORR activity. Besides, Mn0.5 Co0.5 PS3 (100) surface obtains the apex of the volcano plot due to its optimal position of the d-band center which further justifies the improved ORR activity. With Pt-like onset potential, facile H2 O desorption ability, and robust dynamic and thermal stability, this CO tolerant Mn0.5 Co0.5 PS3 catalyst can be a potential alternative to Pt with encouraging practical viability.

Evolutionary structure prediction-assisted design of anode materials for Ca-ion battery based on phosphorene.

The utilization of multivalent ions such as Ca(ii), Mg(ii), and Al(iii) in energy storage devices opens up new opportunities to store energy density in a more efficient manner rather than monovalent Li or Na ion batteries. Active research on Ca(ii) has been limited due to the low diffusion rate of Ca within the lattice as well as the difficulty of the reversible electrodeposition of Ca in standard electrolytes at room temperature. Herein, using first-principles calculations, we have studied the applications of various allotropes of phosphorene (Pn) as potential materials for Ca(ii) battery (CIB). It is seen that among different forms, α and δ phases are suitable to act as anode materials for Ca ion battery. Apart from this, we have also studied the possible formation of various CaxPy phases during the calcination process since it is assumed that during metal insertion and extraction, anodes form non-equilibrium structures. Evolutionary Structure Prediction methods are extensively utilized to determine if the formation of these different CaxPy phases have a significant impact on the anodic performances of Pn or not. It is found that the CaxPy phases formed during the calcination process show reasonable average voltages as well as low volume change and high specific capacity, thus confirming the suitability of Pn as an excellent support for anodes in the Ca(ii) ion battery.

Transition-Metal Phosphorus Trisulfides and its Vacancy Defects: Emergence of a New Class of Anode Material for Li-Ion Batteries.

In the search of suitable anode candidates with high specific capacity, favorable potential, and structural stability for lithium-ion batteries (LIBs), transition-metal phosphorus trisulfides (TMPS3 ) can be considered as one of the most promising alternatives to commercial graphite. Here, it was demonstrated that the limitations of commercial anode materials (i.e., low specific capacity, large volume change, and high lithium diffusion barrier as well as nucleation) can be circumvented by using TMPS3 monolayer surfaces. The study revealed that lithium binds strongly to TMPS3 monolayers (-2.31 eV) without any distortion of the surface, with Li@TMPS3 exhibiting enhanced stability compared with other 2D analogues (graphene, phosphorene, MXenes, transition-metal sulfides and phosphides). The binding energy of lithium was overwhelmingly enhanced with vacancy defects. The vacancy-mediated TMPS3 surfaces showed further amplification of Li binding energy from -2.03 to -2.32 eV and theoretical specific capacity of 441.65 to 484.34 mAh g-1 for MnPS3 surface. Most importantly, minimal change in volume (less than 2 %) after lithiation makes TMPS3 monolayers a very effective candidate for LIBs. Additionally, the ultralow lithium diffusion barrier (0.08 eV) compared with other existing commercial anode material proves the superiority of TMPS3 .

Doped boron nitride surfaces: potential metal free bifunctional catalysts for non-aqueous Li-O2 batteries.

The development of novel cathode catalysts is crucial for the practical application of lithium-oxygen (Li-O2) batteries. In this paper, we have evaluated the catalytic mechanism and activity of doped hexagonal boron nitride (h-BN) surfaces as cathode catalysts for nonaqueous Li-O2 batteries. From the free energy diagrams it is evident that the CN doped h-BN surface shows the best catalytic activity among the others and this arises due to its considerably lower oxygen reduction reaction (ORR) overpotential and lower oxygen evolution reaction (OER) overpotential. This is due to the weaker binding of the first product (LiO2) and stronger binding with the inserted Li atom. The computations predict that among the considered doped h-BN surfaces, the CN doped h-BN surface can be an efficient metal-free cathode material for nonaqueous Li-O2 batteries.

Density functional theory study of defective silicenes as anode materials for lithium ion batteries.

In this contribution, we explore Li adsorption and diffusion on defective silicenes using first principles calculations. Defect formation energy (Ef) values showed that silicenes with 5105 and 5559 vacancy defects (Si-5559 and Si-5105) are likely to form during the fabrication process and Ef values are about one-third of graphenes. Calculation of Li adsorption energy indicated that Si-5559 and Si-5105 are better than pristine silicene for Li dispersion in the half-lithiated state. The diffusion barrier of Li on the surface of Si-5559 and Si-5105 and in the proximity of defected zone were obtained to be 0.24eV and 0.29eV, respectively. Diffusion barrier values show the easy motion of Li on these silicenes in comparison with defective graphenes. Ab-initio molecular dynamic (AIMD) simulations revealed that fully lithiated Si-5559 is not stable and can not accommodate lithium atoms. On the contrary, Si-5105 is stable and could store a certain amount of lithium atoms. The theoretical capacity of Si-5105 was calculated to be 664mAhg-1.

Exotic Physics and Chemistry of Two-Dimensional Phosphorus: Phosphorene.

Phosphorene, the monolayer form of black phosphorus, is the most recent addition to graphene-like van der Waals two-dimensional (2D) systems. Due to its several interesting properties, namely its tunable direct band gap, high carrier mobility, and unique in-plane anisotropy, it has emerged as a promising candidate for electronic and optoelectronic devices. Phosphorene (Pn) reveals a much richer phase diagram than graphene, and it comprises the two forms namely the stapler-clip like (black Pn, α form) and chairlike (blue Pn, ß form) structures. Regardless of its favorable properties, black Pn suffers from instability in oxygen and water, which limits its successful applications in electronic devices. In this Perspective, the cause of structural diversity of Pn, which leads to different properties of both black and blue Pn, is discussed. We provide possible solutions for protecting phosphorene from chemical degradation and its applications in the field of energy storage namely for Li and Na ion batteries.

Steric and electric field driven distortions in aromatic molecules: spontaneous and non-spontaneous symmetry breaking.

The structures of molecules form the cornerstone of our chemical knowledge. Lowering of symmetry in closed-shell molecules is often attributed to the Pseudo Jahn-Teller (PJT) distortions wherein non-adiabatic coupling (NAC) between the ground state and excited states creates vibrational instability along specific normal modes. Nevertheless, other factors like steric interactions are also well known in the literature to induce structural distortions. In this article, we consider two specific cases of molecular distortions - the first one being spontaneous for contorted polyaromatic hydrocarbons (c-PAH) where non-bonded repulsions between the two pairs of syn H-atoms in tribenzopyrene, TBP (1), can enforce either a C2v → C2 or C2v → Cs distortion. PJT-effects account for the correct preference of the Cs structure over C2 (by 4.6 kcal mol-1). The second case (non-spontaneous symmetry breaking) is that of benzene (2) and coronene (3) which upon application of sufficiently strong static external electric field develop vibrational instability along q(a2u) to cause D6h → C6v and D6h → C2 distortions for 2 and 3 respectively. An external electric field (FZ) was applied parallel to the aromatic ring of 2-3 for investigation of non-spontaneous symmetry breaking. Such electric field induced structural distortion is understood on the basis of excess charge accumulation of the planar rings which is circumvented by symmetry lowering. PJT effects seem to have significant consequences for identification of global minima amongst several local minimal molecular structures.

Pseudo-Jahn-Teller Distortion in Two-Dimensional Phosphorus: Origin of Black and Blue Phases of Phosphorene and Band Gap Modulation by Molecular Charge Transfer.

Phosphorene (Pn) is stabilized as a layered material like graphite, yet it possess a natural direct band gap (Eg = 2.0 eV). Interestingly, unlike graphene, Pn exhibits a much richer phase diagram which includes distorted forms like the stapler-clip (black Pn, α form) and chairlike (blue Pn, ß form) structures. The existence of these phases is attributed to pseudo-Jahn-Teller (PJT) instability of planar hexagonal P6(6-) rings. In both cases, the condition for vibronic instability of the planar P6(6-) rings is satisfied. Doping with electron donors like tetrathiafulvalene and tetraamino-tetrathiafulvalene and electron acceptors like tetracyanoquinodimethane and tetracyanoethylene convert blue Pn into N-type and black Pn into efficient P-type semiconductors, respectively. Interestingly, pristine blue Pn, an indirect gap semiconductor, gets converted into a direct gap semiconductor on electron or hole doping. Because of comparatively smaller undulation in blue Pn (with respect to black Pn), the van der Waals interactions between the dopants and blue Pn is stronger. PJT distortions for two-dimensional phosphorus provides a unified understanding of structural features and chemical reactivity in its different phases.