Statistical learning goes beyond the d-band model providing the thermochemistry of adsorbates on transition metals

Hydrogen, Oxygen, and Lead Adsorbates on Al13Co4(100): Accurate Potential Energy Surfaces at Low Computational Cost by Machine Learning and DFT-Based Data

Intermetallic compounds are promising materials in numerous fields, especially those involving surface interactions, such as catalysis. A key factor to investigate their surface properties lies in adsorption energy maps, typically built using first-principles approaches. However, exploring the adsorption energy landscapes of intermetallic compounds can be cumbersome, usually requiring huge computational resources. In this work, we propose an efficient method to predict adsorption energies, based on a Machine Learning (ML) scheme fed by a few Density Functional Theory (DFT) estimates performed on n sites selected through the Farthest Point Sampling (FPS) process. We detail its application on the Al13Co4(100) quasicrystalline approximant surface for several atomic adsorbates (H, O, and Pb). On this specific example, our approach is shown to outperform both simple interpolation strategies and the recent ML force field MACE [arXiv.2206.07697], especially when the number n is small, i.e., below 36 sites. The ground-truth DFT adsorption energies are much more correlated with the predicted FPS-ML estimates (Pearson R-factor of 0.71, 0.73, and 0.90 for H, O and Pb, respectively, when n = 36) than with interpolation-based or MACE-ML ones (Pearson R-factors of 0.43, 0.39, and 0.56 for H, O, and Pb, in the former case and 0.22, 0.35, and 0.63 in the latter case). The unbiased root-mean-square error (ubRMSE) is lower for FPS-ML than for interpolation-based and MACE-ML predictions (0.15, 0.17, and 0.17 eV, respectively, for hydrogen and 0.17, 0.25, and 0.22 eV for lead), except for oxygen (0.55, 0.47, and 0.46 eV) due to large surface relaxations in this case. We believe that these findings and the corresponding methodology can be extended to a wide range of systems, which will motivate the discovery of novel functional materials.

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.

Valence Tautomerism Induced Proton Coupled Electron Transfer:X-H Bond Oxidation with a Dinuclear Au(II) Hydroxide Complex

We report the preparation and characterization of the dinuclear AuII hydroxide complex AuII 2(L)2(OH)2 (L=N,N'-bis (2,6-dimethyl) phenylformamidinate) and study its reactivity towards weak X-H bonds. Through the interplay of kinetic analysis and computational studies, we demonstrate that the oxidation of cyclohexadiene follows a concerted proton-coupled electron transfer (cPCET) mechanism, a rare type of reactivity for Au complexes. We find that the Au-Au σ-bond undergoes polarization in the PCET event leading to an adjustment of oxidation levels for both Au centers prior to C(sp3)-H bond cleavage. We thus describe the oxidation event as a valence tautomerism-induced PCET where the basicity of one reduced Au-OH unit provides a proton acceptor and the second more oxidized Au center serves as an electron acceptor. The coordination of these events allows for unprecedented radical-type reactivity by a closed shell AuII complex.

Latent Variable Machine Learning Framework for Catalysis: General Models, Transfer Learning, and Interpretability

Machine learning has been successfully applied in recent years to screen materials for a variety of applications. However, despite recent advances, most screening-based machine learning approaches are limited in generality and transferability, requiring new models to be created from scratch for each new application. This is particularly apparent in catalysis, where there are many possible intermediates and transition states of interest in addition to a large number of potential catalytic materials. In this work, we developed a new machine learning framework that is built on chemical principles and allows the creation of general, interpretable, reusable models. Our new architecture uses latent variables to create a set of submodels that each take on a relatively simple learning task, leading to higher data efficiency and promoting transfer learning. This architecture infuses fundamental chemical principles, such as the existence of elements as discrete entities. We show that this architecture allows for the creation of models that can be reused for many different applications, providing significant improvements in efficiency and convenience. For example, our architecture allows simultaneous prediction of adsorption energies for many adsorbates on a broad array of alloy surfaces with mean absolute errors (MAEs) around 0.20-0.25 eV. The integration of latent variables provides physical interpretability, as predictions can be explained in terms of the learned chemical environment as represented by the latent space. Further, these latent variables also serve as new feature representations, allowing efficient transfer learning. For example, new models with useful levels of accuracy can be created with less than 10 data points, including transfer learning to an experimental data set with an MAE less than 0.15 eV. Lastly, we show that our new machine learning architecture is general and robust enough to handle heterogeneous and multifidelity data sets, allowing researchers to leverage existing data sets to speed up screening using their own computational setup.

Machine Learning-Driven Discovery of Key Descriptors for CO2 Activation over Two-Dimensional Transition Metal Carbides and Nitrides

Fusing high-throughput quantum mechanical screening techniques with modern artificial intelligence strategies is among the most fundamental ─yet revolutionary─ science activities, capable of opening new horizons in catalyst discovery. Here, we apply this strategy to the process of finding appropriate key descriptors for CO₂ activation over two-dimensional transition metal (TM) carbides/nitrides (MXenes). Various machine learning (ML) models are developed to screen over 114 pure and defective MXenes, where the random forest regressor (RFR) ML scheme exhibits the best predictive performance for the CO₂ adsorption energy, with a mean absolute error ± standard deviation of 0.16 ± 0.01 and 0.42 ± 0.06 eV for training and test data sets, respectively. Feature importance analysis revealed d-band center (ε_d), surface metal electronegativity (χ_M), and valence electron number of metal atoms (M_V) as key descriptors for CO₂ activation. These findings furnish a fundamental basis for designing novel MXene-based catalysts through the prediction of potential indicators for CO₂ activation and their posterior usage.

Fast evaluation of the adsorption energy of organic molecules on metals via graph neural networks

Modeling in heterogeneous catalysis requires the extensive evaluation of the energy of molecules adsorbed on surfaces. This is done via density functional theory but for large organic molecules it requires enormous computational time, compromising the viability of the approach. Here we present GAME-Net, a graph neural network to quickly evaluate the adsorption energy. GAME-Net is trained on a well-balanced chemically diverse dataset with C1-4 molecules with functional groups including N, O, S and C6-10 aromatic rings. The model yields a mean absolute error of 0.18 eV on the test set and is 6 orders of magnitude faster than density functional theory. Applied to biomass and plastics (up to 30 heteroatoms), adsorption energies are predicted with a mean absolute error of 0.016 eV per atom. The framework represents a tool for the fast screening of catalytic materials, particularly for systems that cannot be simulated by traditional methods.

Catalytic modification of corn straw facilitates the remediation of Cd contaminated water and soil

Turning postharvest residue into high-value-added products is crucial for agricultural waste management and environmental remediation. In this proof-of-concept study, nanosized Pt/TiO₂ was used as a model catalyst to modify corn straw (CS) materials through a simple low-temperature oxidation process. This method was demonstrated to be self-sustainable, waste-free, and with high yields. At an optimal temperature of 220 °C, O₂ treatment with 1 wt% Pt/TiO₂ greatly changed ultra-micropore and mesopore structures, dissolved organic carbon, aromatic contents and surface oxygen (O)-containing functional groups in CS products. This treatment resulted in an approximately 5-fold increase of cadmium (Cd) adsorption from aqueous solution and immobilization rate of 43.1% at 7d for bioavailable Cd in soil. Spectroscopic and linear regression analysis demonstrated that both acidic and basic functional groups in CS contributed to Cd adsorption, suggesting chemical adsorption. According to the d-band theory, the unexpected role of catalysts in CS modification could be associated with dissociative adsorption of molecular O₂ on the Pt surface. These results provide insights for the development of economic and sustainable technologies to reutilize agricultural waste biomass for water and soil remediation.

Interpretable Deep Learning Model for Analyzing the Relationship between the Electronic Structure and Chemisorption Property

The use of machine learning (ML) is exploding in materials science as a result of its high predictive performance of material properties. Tremendous trainable parameters are required to build an outperforming predictive model, which makes it impossible to retrace how the model predicts well. However, it is necessary to develop a ML model that can extract human-understandable knowledge while maintaining performance for a universal application to materials science. In this study, we developed a deep learning model that can interpret the correlation between surface electronic density of states (DOSs) of materials and their chemisorption property using the attention mechanism that provides which part of DOS is important to predict adsorption energies. The developed model constructs the well-known d-band center theory without any prior knowledge. This work shows that human-interpretable knowledge can be extracted from complex ML models.

Stick or Spill? Scaling Relationships for the Binding Energies of Adsorbates on Single-Atom Alloy Catalysts

Single-atom alloy catalysts combine catalytically active metal atoms, present as dopants, with the selectivity of coinage metal hosts. Determining whether adsorbates stick at the dopant or spill over onto the host is key to understanding catalytic mechanisms on these materials. Despite a growing body of work, simple descriptors for the prediction of spillover energies (SOEs), i.e., the relative stability of an adsorbate on the dopant versus the host site, are not yet available. Using Density Functional Theory (DFT) calculations on a large set of adsorbates, we identify the dopant charge and the SOE of carbon as suitable descriptors. Combining them into a linear surrogate model, we can reproduce DFT-computed SOEs within 0.06 eV mean absolute error. More importantly, our work provides an intuitive theoretical framework, based on the concepts of electrostatic interactions and covalency, that explains SOE trends and can guide the rational design of future single-atom alloy catalysts.

Machine Learning for Electrocatalyst and Photocatalyst Design and Discovery

Electrocatalysts and photocatalysts are key to a sustainable future, generating clean fuels, reducing the impact of global warming, and providing solutions to environmental pollution. Improved processes for catalyst design and a better understanding of electro/photocatalytic processes are essential for improving catalyst effectiveness. Recent advances in data science and artificial intelligence have great potential to accelerate electrocatalysis and photocatalysis research, particularly the rapid exploration of large materials chemistry spaces through machine learning. Here a comprehensive introduction to, and critical review of, machine learning techniques used in electrocatalysis and photocatalysis research are provided. Sources of electro/photocatalyst data and current approaches to representing these materials by mathematical features are described, the most commonly used machine learning methods summarized, and the quality and utility of electro/photocatalyst models evaluated. Illustrations of how machine learning models are applied to novel electro/photocatalyst discovery and used to elucidate electrocatalytic or photocatalytic reaction mechanisms are provided. The review offers a guide for materials scientists on the selection of machine learning methods for electrocatalysis and photocatalysis research. The application of machine learning to catalysis science represents a paradigm shift in the way advanced, next-generation catalysts will be designed and synthesized.

Deep Learning-Assisted Investigation of Electric Field-Dipole Effects on Catalytic Ammonia Synthesis

External electric fields can modify binding energies of reactive surface species and enhance catalytic performance of heterogeneously catalyzed reactions. In this work, we used density functional theory (DFT) calculations-assisted and accelerated by a deep learning algorithm-to investigate the extent to which ruthenium-catalyzed ammonia synthesis would benefit from application of such external electric fields. This strategy allows us to determine which electronic properties control a molecule's degree of interaction with external electric fields. Our results show that (1) field-dependent adsorption/reaction energies are closely correlated to the dipole moments of intermediates over the surface, (2) a positive field promotes ammonia synthesis by lowering the overall energetics and decreasing the activation barriers of the potential rate-limiting steps (e.g., NH2 hydrogenation) over Ru, (3) a positive field (>0.6 V/Å) favors the reaction mechanism by avoiding kinetically unfavorable N≡N bond dissociation over Ru(1013), and (4) local adsorption environments (i.e., dipole moments of the intermediates in the gas phase, surface defects, and surface coverage of intermediates) influence the resulting surface adsorbates' dipole moments and further modify field-dependent reaction energetics. The deep learning algorithm developed here accelerates field-dependent energy predictions with acceptable accuracies by five orders of magnitudes compared to DFT alone and has the capacity of transferability, which can predict field-dependent energetics of other catalytic surfaces with high-quality performance using little training data.

Modeling Operando Electrochemical CO2 Reduction

Since the seminal works on the application of density functional theory and the computational hydrogen electrode to electrochemical CO₂ reduction (eCO₂R) and hydrogen evolution (HER), the modeling of both reactions has quickly evolved for the last two decades. Formulation of thermodynamic and kinetic linear scaling relationships for key intermediates on crystalline materials have led to the definition of activity volcano plots, overpotential diagrams, and full exploitation of these theoretical outcomes at laboratory scale. However, recent studies hint at the role of morphological changes and short-lived intermediates in ruling the catalytic performance under operating conditions, further raising the bar for the modeling of electrocatalytic systems. Here, we highlight some novel methodological approaches employed to address eCO₂R and HER reactions. Moving from the atomic scale to the bulk electrolyte, we first show how ab initio and machine learning methodologies can partially reproduce surface reconstruction under operation, thus identifying active sites and reaction mechanisms if coupled with microkinetic modeling. Later, we introduce the potential of density functional theory and machine learning to interpret data from Operando spectroelectrochemical techniques, such as Raman spectroscopy and extended X-ray absorption fine structure characterization. Next, we review the role of electrolyte and mass transport effects. Finally, we suggest further challenges for computational modeling in the near future as well as our perspective on the directions to follow.

Automated exploitation of the big configuration space of large adsorbates on transition metals reveals chemistry feasibility

Mechanistic understanding of large molecule conversion and the discovery of suitable heterogeneous catalysts have been lagging due to the combinatorial inventory of intermediates and the inability of humans to enumerate all structures. Here, we introduce an automated framework to predict stable configurations on transition metal surfaces and demonstrate its validity for adsorbates with up to 6 carbon and oxygen atoms on 11 metals, enabling the exploration of ~10⁸ potential configurations. It combines a graph enumeration platform, force field, multi-fidelity DFT calculations, and first-principles trained machine learning. Clusters in the data reveal groups of catalysts stabilizing different structures and expose selective catalysts for showcase transformations, such as the ethylene epoxidation on Ag and Cu and the lack of C-C scission chemistry on Au. Deviations from the commonly assumed atom valency rule of small adsorbates are also manifested. This library can be leveraged to identify catalysts for converting large molecules computationally.

The discovery of heterogeneous catalysts for large molecule conversion has been lagging due to the combinatorial inventory of intermediates. Here, the author presents an automated framework to explore the chemical space of reaction intermediates.

Applications of Machine Learning in Alloy Catalysts: Rational Selection and Future Development of Descriptors

At present, alloys have broad application prospects in heterogeneous catalysis, due to their various catalytic active sites produced by their vast element combinations and complex geometric structures. However, it is the diverse variables of alloys that lead to the difficulty in understanding the structure-property relationship for conventional experimental and theoretical methods. Fortunately, machine learning methods are helpful to address the issue. Machine learning can not only deal with a large number of data rapidly, but also help establish the physical picture of reactions in multidimensional heterogeneous catalysis. The key challenge in machine learning is the exploration of suitable general descriptors to accurately describe various types of alloy catalysts, which help reasonably design catalysts and efficiently screen candidates. In this review, several kinds of machine learning methods commonly used in the design of alloy catalysts is introduced, and the applications of various reactivity descriptors corresponding to different alloy systems is summarized. Importantly, this work clarifies the existing understanding of physical picture of heterogeneous catalysis, and emphasize the significance of rational selection of universal descriptors. Finally, the development of heterogeneous catalytic descriptors for machine learning are presented.

Enhanced Catalytic Ozonation for Eliminating CH3SH via Stable and Circular Electronic Metal-Support Interactions of Si-O-Mn Bonds with Low Mn Loading

Catalytic ozonation of methyl mercaptan (CH₃SH) can effectively control this unbearable odorous sulfur-containing volatile organic compound (S-VOC). The construction of an electronic metal-support interaction (EMSI) coordination structure to maximize the number of active sites and increase the intrinsic activity of active sites is an effective means to improve catalytic performance. In this work, the abundant Si-OH groups on PSBA-15 (SBA-15 before calcination) were used to anchor Mn to form a Si-O-Mn-based EMSI coordination structure. Detailed characterizations and theoretical simulations reveal that the strong EMSI effect significantly adjusts and stabilizes the electronic structure of Mn 3d states, resulting in an electron-rich center on the Si-O-Mn bond to promote the specific adsorption/activation of ozone (O₃) and an electron-poor center on the (Si-O-)Mn-O bond to adsorb a large amount of CH₃SH accompanied by its own oxidative degradation. In situ Raman and in situ Fourier transform infrared (FTIR) analyses identify that catalytic ozonation over 3.0Mn-PSBA generates atomic oxygen species (AOS/*O) and reactive oxygen species (ROS/^•O₂^-) to achieve efficient decomposition of CH₃SH into CO₂/SO₄^2-. Furthermore, the electrons obtained from CH₃SH in electron-poor centers are transferred to maintain the redox cycle of Mn^2+/3+ → Mn⁴⁺ → Mn^2+/3+ through the internal bond bridge, thus accomplishing the efficient and stable degradation of CH₃SH prolonged to 180 min. Therefore, the rational design of catalysts with abundant active sites and optimized inherent activity via the EMSI effect can provide significant potential to improve catalytic performance and eliminate odorous gases.

Catlas: an automated framework for catalyst discovery demonstrated for direct syngas conversion

Catlas may be used with off-the-shelf pretrained models to explore large design spaces for catalyst discovery and has been used here to identify promising materials for the direct conversion of syngas to multi-carbon oxygenates.

Targeted Regulation of the Electronic States of Nickel Toward the Efficient Electrosynthesis of Benzonitrile and Hydrogen Production

Highly efficient electro-oxidation of benzylamine to generate value-added chemicals coupled with the hydrogen evolution reaction (HER) is crucial but challenging. Herein, targeted regulation of the electronic states of Ni sites was realized via simple yet precise nitridation engineering. Benefiting from the insertion of N atoms into the Ni lattice, the Ni₃N electrode exhibits superior activity, selectivity, and stability for the benzylamine oxidation reaction (BOR). Especially, under the industrially relevant current (∼250 mA), the Ni₃N catalyst remains ∼95% selective for benzonitrile production, reaching 1.43 mmol h^-1 cm^-2. Experimental and theoretical findings reveal that the formation of Ni-N bonds upshifts the Ni d-band center and optimizes the electrophilic properties of Ni sites, which contributes to the adsorption and dehydrogenations process of benzylamine. Furthermore, due to the work function difference between Ni and Ni₃N, a strong mutual interaction occurs at the heterogeneous interface for Ni-Ni₃N, which endows it with the appropriate H* adsorption energy and thus excellent HER performance. Impressively, the integrated solar-energy-driven BOR coupled with the HER electrolyzer affords 10 mA cm^-2 at an ultralow voltage of 1.4 V and exhibits a promising practical application (η_{solar-to-hydrogen} = 13.8%). This work offers a new perspective for the bifunctional design of nitrides in the field of electrosynthesis.

Screening and Understanding Li Adsorption on Two-Dimensional Metallic Materials by Learning Physics and Physics-Simplified Learning

Understanding and broad screening Li interaction energetics with surfaces are key to the development of materials for a wide range of applications including Li-based electrochemical capacitors, Li sensors, Li separation membranes, and Li-ion batteries. In this work, we build a high-throughput screening scheme to screen Li adsorption energetics on 2D metallic materials. First, density functional theory and graph convolution networks are utilized to calculate the minimum Li adsorption energies for some 2D metallic materials. The data is then used to find a dependence of the minimum Li adsorption energies on the sum of ionization potential, work function of the 2D metal, and coupling energy between Li⁺ and substrate, and the dependence is used to screen all 2D metallic materials. Physics-simplified learning by splitting the property into different contributions and learning or calculating each component is shown to have higher accuracy and transferability for machine learning of complex materials properties.

Tri-atomic Pt clusters induce effective pathways in a Cocore-Pdshell nanocatalyst surface for a high-performance oxygen reduction reaction

The crux of the hot topic concerning the widespread replacement of fuel cells (FCs) with traditional petrochemical energy is to balance improving the oxygen reduction reaction (ORR) and reducing the cost. The present study employs density functional theory (DFT) to investigate the effect of Pt ensemble size regulation from a single atom to full coverage on the physio-chemical properties, oxygen adsorption energies and overall ORR efficiency of bimetallic nanocatalysts (NCs) with a Co_core-Pd_shell structure. Our results reveal that the electronegativity difference and lattice strain between neighboring heteroatoms are enhanced to trigger a synergetic effect in local domains, with the Pt cluster size reduced from nanometers to subnanometers. They induce a directed and tunable charge relocation mechanism from deep Co to topmost Pt to optimize the adsorption energies of O₂/O* and achieve excellent ORR kinetics performance with minimum Pt usage but maximum Pt atom utilization (i.e., Pt₁ to Pt₃) compared with benchmark Pt(111). Such a dependency between the cluster size and corresponding ORR performance for the established Co@Pd-Pt_n system can be applied to accurately guide the experimental synthesis of ordered heterogeneous catalysts (e.g., other core@shell-clusters structures) toward low Pt, high efficiency and green economy.

Infusing theory into deep learning for interpretable reactivity prediction

Despite recent advances of data acquisition and algorithms development, machine learning (ML) faces tremendous challenges to being adopted in practical catalyst design, largely due to its limited generalizability and poor explainability. Herein, we develop a theory-infused neural network (TinNet) approach that integrates deep learning algorithms with the well-established d-band theory of chemisorption for reactivity prediction of transition-metal surfaces. With simple adsorbates (e.g., *OH, *O, and *N) at active site ensembles as representative descriptor species, we demonstrate that the TinNet is on par with purely data-driven ML methods in prediction performance while being inherently interpretable. Incorporation of scientific knowledge of physical interactions into learning from data sheds further light on the nature of chemical bonding and opens up new avenues for ML discovery of novel motifs with desired catalytic properties.

Low-Cost Pt Alloys for Heterogeneous Catalysis Predicted by Density Functional Theory and Active Learning

Pt is a key high-performing catalyst for important chemical conversions, such as biomass conversion and water splitting. Limited Pt reserves, however, demand that we identify more sustainable alternative catalyst materials for these processes. Here, we combine state-of-the-art graph neural networks and crystal graph machine learning representations with active learning to discover new, low-cost Pt alloy catalysts for biomass reforming and hydrogen evolution reactions. We identify 12 Pt-based alloys which have comparable catalytic activity to that of the exemplar Pt(111) surface. Notably, Cu₃Pt and FeCuPt₂ exhibit near identical catalytic performance as that of Pt(111). These results demonstrate the potential of machine learning for predicting new catalytic materials without recourse to expensive DFT geometry optimizations, the current bottleneck impeding high-throughput materials discovery. We also examine the performance of d-band theory in elucidating trends in binary and ternary Pt alloys.

Adsorption Enthalpies for Catalysis Modeling through Machine-Learned Descriptors

Heterogeneous catalysts are rather complex materials that come in many classes (e.g., metals, oxides, carbides) and shapes. At the same time, the interaction of the catalyst surface with even a relatively simple gas-phase environment such as syngas (CO and H₂) may already produce a wide variety of reaction intermediates ranging from atoms to complex molecules. The starting point for creating predictive maps of, e.g., surface coverages or chemical activities of potential catalyst materials is the reliable prediction of adsorption enthalpies of all of these intermediates. For simple systems, direct density functional theory (DFT) calculations are currently the method of choice. However, a wider exploration of complex materials and reaction networks generally requires enthalpy predictions at lower computational cost.The use of machine learning (ML) and related techniques to make accurate and low-cost predictions of quantum-mechanical calculations has gained increasing attention lately. The employed approaches span from physically motivated models over hybrid physics-ΔML approaches to complete black-box methods such as deep neural networks. In recent works we have explored the possibilities for using a compressed sensing method (Sure Independence Screening and Sparsifying Operator, SISSO) to identify sparse (low-dimensional) descriptors for the prediction of adsorption enthalpies at various active-site motifs of metals and oxides. We start from a set of physically motivated primary features such as atomic acid/base properties, coordination numbers, or band moments and let the data and the compressed sensing method find the best algebraic combination of these features. Here we take this work as a starting point to categorize and compare recent ML-based approaches with a particular focus on model sparsity, data efficiency, and the level of physical insight that one can obtain from the model.Looking ahead, while many works to date have focused only on the mere prediction of databases of, e.g., adsorption enthalpies, there is also an emerging interest in our field to start using ML predictions to answer fundamental science questions about the functioning of heterogeneous catalysts or perhaps even to design better catalysts than we know today. This task is significantly simplified in works that make use of scaling-relation-based models (volcano curves), where the model outcome is determined by only one or two adsorption enthalpies and which consequently become the sole target for ML-based high-throughput screening or design. However, the availability of cheap ML energetics also allows going beyond scaling relations. On the basis of our own work in this direction, we will discuss the additional physical insight that can be achieved by integrating ML-based predictions with traditional catalysis modeling techniques from thermal and electrocatalysis, such as the computational hydrogen electrode and microkinetic modeling, as well as the challenges that lie ahead.

Inverse-designed semiconductor nanocatalysts for targeted CO2 reduction in water

The most commonly used photocatalyst for CO₂ reduction is TiO₂. However, this semiconductor material is far from being ideally suited for this purpose, owing to its inefficient energy harvesting (it absorbs in the UV), low reduction rates (it exhibits short carrier lifetimes), and lack of selectivity with respect to competing reactions (such as the nearly isoenergetic and kinetically more favourable water reduction). In this work we compile a wish-list of properties for the ideal photocatalyst (including high reaction selectivity, availability of multiple redox equivalents at one time, large contact area for CO₂ adsorption with independently tunable band gap, and availability of electrons and holes at different locations on the surface for the two redox reactions to take place), and, using the principles of inverse design, we engineer a semiconductor nanostructure that not only meets all the necessary fundamental criteria to act as a catalyst for CO₂ reduction, but also exhibits all the wish-list properties, as confirmed by our state-of-the-art atomistic semi-empirical pseudopotential modelling. The result is a potentially game-changing material.

Perspective on computational reaction prediction using machine learning methods in heterogeneous catalysis

Heterogeneous catalysis plays a significant role in the modern chemical industry. Towards the rational design of novel catalysts, understanding reactions over surfaces is the most essential aspect. Typical industrial catalytic processes such as syngas conversion and methane utilisation can generate a large reaction network comprising thousands of intermediates and reaction pairs. This complexity not only arises from the permutation of transformations between species but also from the extra reaction channels offered by distinct surface sites. Despite the success in investigating surface reactions at the atomic scale, the huge computational expense of ab initio methods hinders the exploration of such complicated reaction networks. With the proliferation of catalysis studies, machine learning as an emerging tool can take advantage of the accumulated reaction data to emulate the output of ab initio methods towards swift reaction prediction. Here, we briefly summarise the conventional workflow of reaction prediction, including reaction network generation, ab initio thermodynamics and microkinetic modelling. An overview of the frequently used regression models in machine learning is presented. As a promising alternative to full ab initio calculations, machine learning interatomic potentials are highlighted. Furthermore, we survey applications assisted by these methods for accelerating reaction prediction, exploring reaction networks, and computational catalyst design. Finally, we envisage future directions in computationally investigating reactions and implementing machine learning algorithms in heterogeneous catalysis.

Collaboration between a Pt-dimer and neighboring Co-Pd atoms triggers efficient pathways for oxygen reduction reaction

The development of electrocatalysts with reconcilable balance between the cost and performance in oxygen reduction reaction (ORR) is an imperative task for the widespread adoption of fuel cell technology. In this study, we proposed a unique model of diatomic Pt-cluster (Pt-dimer) in the topmost layer of the Co/Pd bimetallic slab (Co@Pd-Pt2) for mimicking the Cocore@Pdshell nanocatalysts (NCs) surface and systematically investigating its local-regional collaboration pathways in ORR by density functional theory (DFT). The results demonstrate that the Pt-dimer produces local differentiation from both ligand and geometric effects on the Co@Pd surface, which forms adsorption energy (Eads) gradients for relocating the ORR-adsorbates. Our calculations for Eads-variations of ORR-species, reaction coordinates, and intraparticle charge injection propose and confirm a novel local synergetic collaboration around the Pt-dimer in the Co@Pd-Pt2 system with the best-performing ORR behavior compared with all reference models. With proper selection of the composition in intraparticle components, the proposed DFT assessments could be adopted for developing economical and high-performance catalysts in various heterogeneous reactions.

Machine learned features from density of states for accurate adsorption energy prediction

Materials databases generated by high-throughput computational screening, typically using density functional theory (DFT), have become valuable resources for discovering new heterogeneous catalysts, though the computational cost associated with generating them presents a crucial roadblock. Hence there is a significant demand for developing descriptors or features, in lieu of DFT, to accurately predict catalytic properties, such as adsorption energies. Here, we demonstrate an approach to predict energies using a convolutional neural network-based machine learning model to automatically obtain key features from the electronic density of states (DOS). The model, DOSnet, is evaluated for a diverse set of adsorbates and surfaces, yielding a mean absolute error on the order of 0.1 eV. In addition, DOSnet can provide physically meaningful predictions and insights by predicting responses to external perturbations to the electronic structure without additional DFT calculations, paving the way for the accelerated discovery of materials and catalysts by exploration of the electronic space.

Computational catalysis would strongly benefit from general descriptors applicable for predicting adsorption energetics. Here the authors propose a machine-learning approach for adsorption energy predictions based on learning the relevant descriptors in a surface atom's density of states as part of the training.

Iterative Supervised Principal Component Analysis Driven Ligand Design for Regioselective Ti-Catalyzed Pyrrole Synthesis

The rational design of catalysts remains a challenging endeavor within the broader chemical community owing to the myriad variables that can affect key bond-forming events. Designing selective catalysts for any reaction requires an efficient strategy for discovering predictive structure-activity relationships. Herein, we describe the use of iterative supervised principal component analysis (ISPCA) in de novo catalyst design. The regioselective synthesis of 2,5-dimethyl-1,3,4-triphenyl-1H-pyrrole (C) via a Ti-catalyzed formal [2 + 2 +1] cycloaddition of phenylpropyne and azobenzene was targeted as a proof of principle. The initial reaction conditions led to an unselective mixture of all possible pyrrole regioisomers. ISPCA was conducted on a training set of catalysts, and their performance was regressed against the scores from the top three principal components. Component loadings from this PCA space and k-means clustering were used to inform the design of new test catalysts. The selectivity of a prospective test set was predicted in silico using the ISPCA model, and optimal candidates were synthesized and tested experimentally. This data-driven predictive-modeling workflow was iterated, and after only three generations the catalytic selectivity was improved from 0.5 (statistical mixture of products) to over 11 (>90% C) by incorporating 2,6-dimethyl-4-(pyrrolidin-1-yl)pyridine as a ligand. The origin of catalyst selectivity was probed by examining ISPCA variable loadings in combination with DFT modeling, revealing that ligand lability plays an important role in selectivity. A parallel catalyst search using multivariate linear regression (MLR), a popular approach in catalysis informatics, was also conducted in order to compare these strategies in a hypothetical catalyst scouting campaign. ISPCA appears to be more robust and predictive than MLR when sparse training sets are used that are representative of the data available during the early search for an optimal catalyst. The successful development of a highly selective catalyst without resorting to long, stochastic screening processes demonstrates the inherent power of ISPCA in de novo catalyst design and should motivate the general use of ISPCA in reaction development.

Progress in Computational and Machine-Learning Methods for Heterogeneous Small-Molecule Activation

The chemical conversion of small molecules such as H₂ , H₂ O, O₂ , N₂ , CO₂ , and CH₄ to energy and chemicals is critical for a sustainable energy future. However, the high chemical stability of these molecules poses grand challenges to the practical implementation of these processes. In this regard, computational approaches such as density functional theory, microkinetic modeling, data science, and machine learning have guided the rational design of catalysts by elucidating mechanistic insights, identifying active sites, and predicting catalytic activity. Here, the theory and methodologies for heterogeneous catalysis and their applications for small-molecule activation are reviewed. An overview of fundamental theory and key computational methods for designing catalysts, including the emerging data science techniques in particular, is given. Applications of these methods for finding efficient heterogeneous catalysts for the activation of the aforementioned small molecules are then surveyed. Finally, promising directions of the computational catalysis field for further outlooks are discussed, focusing on the challenges and opportunities for new methods.

Predicting metal-metal interactions. II. Accelerating generalized schemes through physical insights

Operando-computational frameworks that integrate descriptors for catalyst stability within catalyst screening paradigms enable predictions of rates and selectivity on chemically faithful representations of nanoparticles under reaction conditions. These catalyst stability descriptors can be efficiently predicted by density functional theory (DFT)-based models. The alloy stability model, for example, predicts the stability of metal atoms in nanoparticles with site-by-site resolution. Herein, we use physical insights to present accelerated approaches of parameterizing this recently introduced alloy-stability model. These accelerated approaches meld quadratic functions for the energy of metal atoms in terms of the coordination number with linear correlations between model parameters and the cohesive energies of bulk metals. By interpolating across both the coordination number and chemical space, these accelerated approaches shrink the training set size for 12 fcc p- and d-block metals from 204 to as few as 24 DFT calculated total energies without sacrificing the accuracy of our model. We validate the accelerated approaches by predicting adsorption energies of metal atoms on extended surfaces and 147 atom cuboctahedral nanoparticles with mean absolute errors of 0.10 eV and 0.24 eV, respectively. This efficiency boost will enable a rapid and exhaustive exploration of the vast material space of transition metal alloys for catalytic applications.