Mitrofanov, Artem A

Dudakov I.V., Savelev S.A., Nevolin I.M., Mitrofanov A.A., Korolev V.V., Gorbunova Y.G.

The presented multimodal transformer networks quantitatively reproduce experimental proton conductivity and the underlying conduction mechanism and provide predictive uncertainty estimates.

Yakovenko E.I., Nevolin I.M., Chasovskikh A.A., Mitrofanov A.A., Korolev V.V.

Korolev V., Mitrofanov A.

While artificial intelligence drives remarkable progress in natural sciences, its broader societal implications are mostly disregarded. In this study, we evaluate environmental impacts of deep learning in materials science through extensive benchmarking. In particular, a set of diverse neural networks is trained for a given supervised learning task to assess greenhouse gas (GHG) emissions during training and inference phases. A chronological perspective showed diminishing returns, manifesting themselves as a 28% decrease in mean absolute error and nearly a 15,000% increase in the carbon footprint of model training in 2016-2022. By means of up-to-date graphics processing units, it is possible to partially offset the immense growth of GHG emissions. Nonetheless, the practice of employing energy-efficient hardware is overlooked by the materials informatics community, as follows from a literature analysis in the field. On the basis of our findings, we encourage researchers to report GHG emissions together with standard performance metrics.

Korolev V., Mitrofanov A.

Zubenko A.D., Shchukina A.A., Chernikova E.Y., Egorova B., Ikonnikova I., Priselkova A.B., Larenkov A.A., Bubenshchikov V.B., Mitrofanov A., Fedorov Y.V., Fedorova O.

In this article, we present the synthesis and characterization of new acyclic pyridine-containing polyaminocarboxylate ligands H4aPyta and H6aPyha, which differ in structural rigidity and the number of chelating groups. Their...

Matazova E.V., Egorova B.V., Zubenko A.D., Pashanova A.V., Mitrofanov A.A., Fedorova O.A., Ermolaev S.V., Vasiliev A.N., Kalmykov S.N.

Vlasov S.I., Smirnova A.A., Ponomarev A.V., Uchkina D.A., Sholokhova A.Y., Mitrofanov A.A.

Zamurueva L.S., Egorova B., Ikonnikova I., Zubenko A.D., Pashanova A.V., Karnoukhova V.A., Mitrofanov A., Trigub A., Moiseeva A.A., Priselkova A.B., Fedorova O., Kalmykov S.N.

In this work, we synthesized two new benzo-18-azacrown-6 ethers bearing picolinate and pyridine pendant arms and studied the copper complexes of these ligands, as well as those of an acetate analog.

Chernysheva M.G., Popov A.G., Dzianisik M.G., Egorov A.V., Egorova T.B., Gopin A.V., Mitrofanov A.A., Badun G.A.

Hydrogen treatment is a popular way of surface modification of nanodiamonds. Here, we used atomic hydrogen treatment of the functionalized surface to increase its hydrophobicity gently and maintain its overall composition. Corresponding mechanism was revealed via combination of theoretical and experimental methods.

Smirnova A., Yablonskiy M., Petrov V., Mitrofanov A.

Radiolytic stability is one of the main requirements of the ligands for the reprocessing of spent nuclear fuel. The prediction of radiation stability based only on the 2D molecular structural formula allows us to accelerate and simplify the development of new ligands. Here, we used quantum chemistry to investigate the radiolytic behavior of water-soluble diglycolamides as one of the most popular ligands for spent nuclear fuel reprocessing. The accurate accounting of conformational mobility in the descriptors based on the Frontier Orbital Fukui theory allowed us to obtain a good correlation between theoretical and experimental data.

Andreadi N., Zankov D., Karpov K., Mitrofanov A.

AbstractFinding global and local minima on the potential energy surface is a key task for most studies in computational chemistry. Having a set of possible conformations for chemical structures and their corresponding energies, one can judge their chemical activity, understand the mechanisms of reactions, describe the formation of metal‐ligand and ligand‐protein complexes, and so forth. Despite the fact that the interest in various minima search algorithms in computational chemistry arose a while ago (during the formation of this science), new methods are still emerging. These methods allow to perform conformational analysis and geometry optimization faster, more accurately, or for more specific tasks. This article presents the application of a novel global geometry optimization approach based on the Tree Parzen Estimator method. For benchmarking, a database of small organic molecule geometries in the global minimum conformation was created, as well as a software package to perform the tests.

Egorova B.V., Zamurueva L.S., Zubenko A.D., Pashanova A.V., Mitrofanov A.A., Priselkova A.B., Fedorov Y.V., Trigub A.L., Fedorova O.A., Kalmykov S.N.

A synthetic procedure for the synthesis of azacrown ethers with a combination of pendant arms has been developed and the synthesized ligand, characterized by various techniques, was studied. The prepared benzoazacrown ether with hybrid pendant arms and its complexes with copper and lead cations were studied in terms of biomedical applications. Similarly to a fully acetate analog, the new one binds both cations with close stability constants, despite the decrease in both constants. The calculated geometry of the complexes correlate with the data from X-ray absorption and NMR spectroscopy. Coordination of both cations differs due to the difference between the ionic radii. However, these chelation modes provide effective shielding of cations in both cases, that was shown by the stability of their complexes in the biologically relevant media towards transchelation and transmetallation.

Nugmanova A.G., Safonova E.A., Baranchikov A.E., Tameev A.R., Shkolin A.V., Mitrofanov A.A., Eliseev A.A., Meshkov I.N., Kalinina M.A.

• Photocatalysts are self-assembled from zinc porphyrins MOFs and graphene oxide. • These hybrids show ambivalent ability to red/ox photodegradation of dyes. • Photooxidation of dyes by singlet oxygen occurs in the oxygen-rich solutions. • Photoreduction of the same dyes is initiated under anaerobic conditions. • Size matching between the dyes and MOF micropores controls photoreduction. New photocatalysts were synthesized from graphene oxide (GO) and zinc porphyrins via non-covalent self-assembly in Pickering emulsions. The formation of surface-attached metal organic frameworks (SURMOFs) with different size of mesopores (1.1. and 1.6 nm) was confirmed by X-ray powder diffraction and BET nitrogen absorption methods. The activity of the SURMOF/GO materials in photodegradation of rhodamine 6G (Rh6G) and 1,5-dihydroxynaphtalene (DHN) were studied spectroscopically. The photocatalysts initiate aerobic oxidative photodestruction with k up to 2.3 × 10 −1 min −1 through generation of singlet oxygen on porphyrin centers. Under anaerobic conditions, these materials assist photoreduction of the same dyes in the SURMOF micropores. The mechanisms of photodegradation assisted by SURMOF/GO hybrids were confirmed by a combination of MALDI-TOF spectroscopy, Sensor Green and terephthalic acid probing. The size of the SURMOF pores controls the reduction, which occur due to the effective charge separation between porphyrin SURMOFs and GO. The photocatalyst with larger pores can transform both Rh6G and DHN, whereas that one with smaller pores is active only with respect to small DHN molecules. The ability of as-formed SURMOF/GOs to exploit two mechanisms yielding different products of photodestruction provides a basis for creating novel ambivalent photocatalysts for selective transformations of targeted compounds in molecular mixtures.

Mitrofanov A., Andreadi N., Korolev V., Kalmykov S.

Actinide chemistry often lies beyond the applicability domain of the majority of modern theoretical tools due to high computational costs, relativistic effects, or just the absence of actinide data for semiempirical method fitting. On the other hand, radioactivity pushes the usage of computational methods instead of experimental ones. Here, we would like to present a novel relPBE functional as an actinide-fitted version of the PBE0 functional.

Karpov K., Mitrofanov A., Korolev V., Tkachenko V.

The use of machine learning in chemistry has become a common practice. At the same time, despite the success of modern machine learning methods, the lack of data limits their use. Using a transfer learning methodology can help solve this problem. This methodology assumes that a model built on a sufficient amount of data captures general features of the chemical compound structure on which it was trained and that the further reuse of these features on a data set with a lack of data will greatly improve the quality of the new model. In this paper, we develop this approach for small organic molecules, implementing transfer learning with graph convolutional neural networks. The paper shows a significant improvement in the performance of the models for target properties with a lack of data. The effects of the data set composition on the model's quality and the applicability domain of the resulting models are also considered.

Yan S., Xu P., Li G., Zhu Y., Wu Y., Chen Q., Liu S., Li Q., Tang M.

Wang B., Xu Z., Han Z., Nie Q., Chen X., Xiao H., Yan G.

Enikeeva M.O., Proskurina O.V., Lopatin S.I., Gusarov V.V.

Abdolmaleki S., Aliabadi A., Khaksar S.

Wei Q., Qiu J., Wang R., Sun Z., Xie Y., Chen Y., Wan Y.

Liang Q., Yang J.

Norquist A.J.

Kang Y., Zheng Y., Zhang Z., Li B., Liu Y., Xie Y., Li Z., Tang X., Chen J., Wang L., Xu C.

Nurdillayeva R.N., Moral R.F., Pols M., Lee D., Altoe V., Schwartz C.P., Tao S., Sutter-Fella C.M.

Chen W., Mularso K.T., Jo B., Jung H.S.

This review explores the evolution of indoor perovskite solar cells driven by recent advances in material optimization and machine learning, fostering sustainable indoor energy solutions for interconnected smart technologies.

Lacerda S., de Kruijff R.M., Djanashvili K.

Recent years have seen a swift rise in the use of α-emitting radionuclides such as 225Ac and 223Ra as various radiopharmaceuticals to treat (micro)metastasized tumors. They have shown remarkable effectiveness in clinical practice owing to the highly cytotoxic α-particles that are emitted, which have a very short range in tissue, causing mainly double-stranded DNA breaks. However, it is essential that both chelation and targeting strategies are optimized for their successful translation to clinical application, as α-emitting radionuclides have distinctly different features compared to β−-emitters, including their much larger atomic radius. Furthermore, upon α-decay, any daughter nuclide irrevocably breaks free from the targeting molecule, known as the recoil effect, dictating the need for faster targeting to prevent healthy tissue toxicity. In this review we provide a brief overview of the current status of targeted α-therapy and highlight innovations in α-emitter-based chelator design, focusing on the role of click chemistry to allow for fast complexation to biomolecules at mild labeling conditions. Finally, an outlook is provided on different targeting strategies and the role that pre-targeting can play in targeted alpha therapy.

Leniak A., Pietruś W., Świderska A., Kurczab R.

Semelak J.A., Pickering I., Huddleston K., Olmos J., Grassano J.S., Clemente C.M., Drusin S.I., Marti M., Gonzalez Lebrero M.C., Roitberg A.E., Estrin D.A.

Han X., Wang X., Xu M., Feng Z., Yao B., Guo P., Gao Z., Lu Z.

Abstract The discovery of advanced materials is the cornerstone of human technological development and progress. The structures of materials and their corresponding properties are essentially the result of a complex interplay of multiple degrees of freedom such as lattice, charge, spin, symmetry, and topology. This poses significant challenges for the inverse design methods of materials. Humans have long explored new materials through a large number of experiments and proposed corresponding theoretical systems to predict new material properties and structures. With the improvement of computational power, researchers have gradually developed various electronic structure calculation methods, such as the density functional theory and high-throughput computational methods. Recently, the rapid development of artificial intelligence technology in the field of computer science has enabled the effective characterization of the implicit association between material properties and structures, thus opening up an efficient paradigm for the inverse design of functional materials. A significant progress has been made in inverse design of materials based on generative and discriminative models, attracting widespread attention from researchers. Considering this rapid technological progress, in this survey, we look back on the latest advancements in AI-driven inverse design of materials by introducing the background, key findings, and mainstream technological development routes. In addition, we summarize the remaining issues for future directions. This survey provides the latest overview of AI-driven inverse design of materials, which can serve as a useful resource for researchers.

Dong L., Xu X., Li W.

Han S., Lee B.G., Lim D., Kim J.

Terrones G.G., Huang S., Rivera M.P., Yue S., Hernandez A., Kulik H.J.

Sriram A., Choi S., Yu X., Brabson L.M., Das A., Ulissi Z., Uyttendaele M., Medford A.J., Sholl D.S.

Korolev V., Mitrofanov A.

Moghadam P.Z., Chung Y.G., Snurr R.Q.

Metal–organic frameworks (MOFs) are a class of nanoporous material precisely synthesized from molecular building blocks. MOFs could have a critical role in many energy technologies, including carbon capture, separations and storage of energy carriers. Molecular simulations can improve our molecular-level understanding of adsorption in MOFs, and it is now possible to use realistic models for these complicated materials and predict their adsorption properties in quantitative agreement with experiments. Here we review the predictive design and discovery of MOF adsorbents for the separation and storage of energy-relevant molecules, with a view to understanding whether we can reliably discover novel MOFs computationally prior to laboratory synthesis and characterization. We highlight in silico approaches that have discovered new adsorbents that were subsequently confirmed by experiments, and we discuss the roles of high-throughput computational screening and machine learning. We conclude that these tools are already accelerating the discovery of new applications for existing MOFs, and there are now several examples of new MOFs discovered by computational modelling. Metal–organic frameworks (MOFs) are porous materials that may find application in numerous energy settings, such as carbon capture and hydrogen-storage technologies. Here, the authors review predictive computational design and discovery of MOFs for separation and storage of energy-relevant gases.

Korolev V., Protsenko P.

Property prediction accuracy has long been a key parameter of machine learning in materials informatics. Accordingly, advanced models showing state-of-the-art performance turn into highly parameterized black boxes missing interpretability. Here, we present an elegant way to make their reasoning transparent. Human-readable text-based descriptions automatically generated within a suite of open-source tools are proposed as materials representation. Transformer language models pretrained on 2 million peer-reviewed articles take as input well-known terms such as chemical composition, crystal symmetry, and site geometry. Our approach outperforms crystal graph networks by classifying four out of five analyzed properties if one considers all available reference data. Moreover, fine-tuned text-based models show high accuracy in the ultra-small data limit. Explanations of their internal machinery are produced using local interpretability techniques and are faithful and consistent with domain expert rationales. This language-centric framework makes accurate property predictions accessible to people without artificial-intelligence expertise.

Comlek Y., Pham T.D., Snurr R.Q., Chen W.

AbstractData-driven materials design often encounters challenges where systems possess qualitative (categorical) information. Specifically, representing Metal-organic frameworks (MOFs) through different building blocks poses a challenge for designers to incorporate qualitative information into design optimization, and leads to a combinatorial challenge, with large number of MOFs that could be explored. In this work, we integrated Latent Variable Gaussian Process (LVGP) and Multi-Objective Batch-Bayesian Optimization (MOBBO) to identify top-performing MOFs adaptively, autonomously, and efficiently. We showcased that our method (i) requires no specific physical descriptors and only uses building blocks that construct the MOFs for global optimization through qualitative representations, (ii) is application and property independent, and (iii) provides an interpretable model of building blocks with physical justification. By searching only ~1% of the design space, LVGP-MOBBO identified all MOFs on the Pareto front and 97% of the 50 top-performing designs for the CO2 working capacity and CO2/N2 selectivity properties.

Zheng Z., Zhang O., Borgs C., Chayes J.T., Yaghi O.M.

Korolev V., Mitrofanov A.

While artificial intelligence is promoting remarkable progress in natural sciences, its broader societal implications are mostly disregarded. In this study, we evaluate the environmental impacts of deep learning in materials science through extensive benchmarking. Specifically, a diverse set of neural networks is trained for a particular supervised learning task to assess greenhouse gas (GHG) emissions during training and inference phases. A chronological perspective shows diminishing returns, which are expressed in a 28% decrease in mean absolute error and nearly a 15000% increase in carbon footprint of model training in 2016–2022. By utilizing up-to-date graphics processing units, it is possible to partially offset the immense growth in GHG emissions. Nonetheless, the practice to employ energy-efficient hardware is unaddressed by the materials informatics community, as it follows from the literature analysis in the field. Based on our findings, we encourage researchers to report GHG emissions on par with standard performance metrics.

Chien A.A., Lin L., Nguyen H., Rao V., Sharma T., Wijayawardana R.

Intergovernmental Panel on Climate Change (IPCC)

The Working Group II contribution to the Sixth Assessment Report of the Intergovernmental Panel on Climate Change (IPCC) provides a comprehensive assessment of the scientific literature relevant to climate change impacts, adaptation and vulnerability. The report recognizes the interactions of climate, ecosystems and biodiversity, and human societies, and integrates across the natural, ecological, social and economic sciences. It emphasizes how efforts in adaptation and in reducing greenhouse gas emissions can come together in a process called climate resilient development, which enables a liveable future for biodiversity and humankind. The IPCC is the leading body for assessing climate change science. IPCC reports are produced in comprehensive, objective and transparent ways, ensuring they reflect the full range of views in the scientific literature. Novel elements include focused topical assessments, and an atlas presenting observed climate change impacts and future risks from global to regional scales. Available as Open Access on Cambridge Core.

Korovin A.N., Humonen I.S., Samtsevich A.I., Eremin R.A., Vasilev A.I., Lazarev V.D., Budennyy S.A.

The discovery of new catalysts is one of the significant topics of computational chemistry as it has the potential to accelerate the adoption of renewable energy sources. Recently developed deep learning approaches such as graph neural networks (GNNs) open new opportunity to significantly extend scope for modelling novel high-performance catalysts. Nevertheless, the graph representation of particular crystal structure is not a straightforward task due to the ambiguous connectivity schemes and numerous embeddings of nodes and edges. Here we present embedding improvement for GNN that has been modified by Voronoi tesselation and is able to predict the energy of catalytic systems within Open Catalyst Project dataset. Enrichment of the graph was calculated via Voronoi tessellation and the corresponding contact solid angles and types (direct or indirect) were considered as features of edges and Voronoi volumes were used as node characteristics. The auxiliary approach was enriching node representation by intrinsic atomic properties (electronegativity, period and group position). Proposed modifications allowed us to improve the mean absolute error of the original model and the final error equals to 651 meV per atom on the Open Catalyst Project dataset and 6 meV per atom on the intermetallics dataset. Also, by consideration of additional dataset, we show that a sensible choice of data can decrease the error to values above physically-based 20 meV per atom threshold.

Queen O., McCarver G.A., Thatigotla S., Abolins B.P., Brown C.L., Maroulas V., Vogiatzis K.D.

AbstractThe prediction of a variety of polymer properties from their monomer composition has been a challenge for material informatics, and their development can lead to a more effective exploration of the material space. In this work, PolymerGNN, a multitask machine learning architecture that relies on polymeric features and graph neural networks has been developed towards this goal. PolymerGNN provides accurate estimates for polymer properties based on a database of complex and heterogeneous polyesters (linear/branched, homopolymers/copolymers) with experimentally refined properties. In PolymerGNN, each polyester is represented as a set of monomer units, which are introduced as molecular graphs. A virtual screening of a large, computationally generated database with materials of variable composition was performed, a task that demonstrates the applicability of the PolymerGNN on future studies that target the exploration of the polymer space. Finally, a discussion on the explainability of the models is provided.

Mourino B., Jablonka K.M., Ortega‐Guerrero A., Smit B.

AbstractCovalent organic frameworks (COFs) stand out as prospective organic‐based photocatalysts given their intriguing optoelectronic properties, such as visible light absorption and high charge‐carrier mobility. The “Clean, Uniform, Refined with Automatic Tracking from Experimental Database” (CURATED) COFs is a database of reported experimental COFs that until now remained mostly unexplored for photocatalysis. In this study, the CURATED COFs database is screened for discovering potential photocatalysts using a set of DFT‐based descriptors that cost‐effectively assesses visible light absorption, preliminary thermodynamic feasibility of the desired pair of redox reactions, charge separation, and charge‐carrier mobility. The workflow can shortlist 13 COFs as prospective candidates for water splitting, and identify materials (Nx‐COF (x = 0–3)) that have been reported as candidates for hydrogen evolution reaction. Overall, the strategy addresses the challenge of exploring a large number of COFs by directing future research toward a selective group of COFs, while providing valuable insights into the structural design for achieving a desired photocatalytic process.

Zhang Z., Tang H., Wang M., Lyu B., Jiang Z., Jiang J.