Публикации

The discovery of new chemical transformations is central to advancing modern chemistry, yet conventional approaches often require months or years of extensive experimental screening. Here, we present a machine-learning-assisted and expert-guided pipeline for reaction discovery applied to the search for atom-economic cycloaddition reactions. Candidate reactions were generated from publicly available quantum chemical data, filtered through unsupervised machine learning, and clustered to reduce redundancy. A digital co-expert then enabled rapid prioritization, after which human expertise provided final selection and experimental validation. This hybrid workflow is fully compatible with current laboratory infrastructure and addresses the most time-consuming stage of reaction discovery, accelerating the expert screening bottleneck by approximately 180-fold (from > 1200 days to 7 days). Within ∼1 week, two novel cycloaddition reactions were identified and experimentally confirmed, yielding previously undescribed products. While fully autonomous robotic platforms represent a long-term vision, their high cost and limited availability restrict immediate application. In contrast, our approach demonstrates the practicality of human-AI collaboration for reaction discovery, combining computational screening, machine learning and expert knowledge to efficiently expand the accessible chemical space.

Light offers a unique means of controlling matter with high precision, yet the development of robust photoresponsive transition-metal complexes remains a challenge. Here we report a self-tuning photochromic system based on a diarylethene-derived bis-NHC-palladium complex. The trans-anti complex ( 1oo) undergoes efficient stepwise photocyclization as well as unprecedented light-induced trans/cis isomerization at the metal center. Isolation and crystallographic characterization of the cis-anti isomer (2oo) reveal a thermodynamically more stable structure with enhanced photochromic performance and reversible multistate switching. Thermal studies uncover interconversion with additional rotamer, establishing a dynamic equilibrium among several photoactive palladium species. Spectroscopic and computational investigations elucidate the electronic transitions that drive both diarylethene cyclization and Pd─NHC geometric rearrangements. We demonstrate that the catalytic activity in the Suzuki-Miyaura coupling reaction can be reversibly switched by light, with the photocyclized catalyst forms showing negligible catalytic activity, while the open forms achieve high efficiency. This establishes a direct link between photoisomerization and predicted catalytic performance. Pre-catalyst evolution demonstrates that the geometry of the complex controls the balance between nanoparticle-mediated and homogeneous reactivity, delineating a novel strategy for adaptive catalysis.

Single-atom catalysts (SACs) represent a pinnacle of atomic efficiency and catalytic precision. Their remarkable activity and selectivity arise from isolated, low-coordinate metal centers that engage directly in bond-forming events. However, under realistic reaction conditions, SACs are far from static. Increasing evidence reveals that single atoms undergo dynamic evolution over the reaction time. In this perspective, we challenge the conventional dichotomy that views SACs and nanoparticles (NPs) as fundamentally distinct catalytic systems. We propose that NPs, rather than acting as parallel or cooperative catalysts, may function as catalytic poisonants for SACs by trapping active metal atoms. This transformation results in loss of activity, reduced selectivity, and degradation of the catalytic system. Drawing on mechanistic studies, thermodynamic data, and experimental observations across diverse reaction classes, including hydrogenation, oxidation, and cross-coupling, we show that the aggregation of SACs into NPs is not merely a side process but rather a limitation to their stability and utility. We further outline thermodynamic and kinetic strategies to suppress this deactivation pathway and propose design principles that elevate NP suppression from a synthetic challenge to a foundational criterion in catalyst development. This perspective reframes the SAC–NP relationship as a dynamic continuum and emphasizes the importance of stabilizing isolated active sites in next-generation catalytic technologies.

Accumulation of furanic aldehydes, including furfural (FF) and 5-(hydroxymethyl)furfural (HMF), poses a major bottleneck in some microbial conversions of lignocellulosic biomass to bioethanol. These compounds significantly impair microbial activity in both biofuel fermentation and industrial wastewater treatment systems. In this study, a microbial consortium composed of Rhodococcus erythropolis Ac-858, Rhodococcus fascians Ac-1462, and Pseudomonas veronii B-877 was developed and evaluated for its capacity to biodegrade FF and HMF under model bioreactor conditions. The consortium demonstrated complete degradation of FF (4 g/L) within 48 hours under aerobic conditions and substantial conversion of HMF to non-toxic intermediates. Distinct metabolic pathways were observed depending on aeration intensity: FF and HMF were reduced to furfuryl alcohol and 2,5-bis (hydroxymethyl)furan (BHMF) under aerobic conditions, while oxidation under oxygen limitation produced furoic acid and 2,5-furandicarboxylic acid (FDCA). The synergistic action between P. veronii (reductive) and Rhodococcus spp. (oxidative) was confirmed. Scanning electron microscopy and fractal analysis revealed significant morphological stress responses to furanic aldehydes, with species-specific adaptation patterns. Phytotoxicity tests with Lepidium sativum (watercress) and Lemna spp. (duckweed) showed that the treated culture fluid, after appropriate dilution, met safety thresholds for environmental discharge. This study introduces a biologically based strategy for efficient removal of furanic inhibitors in bioethanol production and industrial effluents, with potential scalability and compliance with discharge regulations. The findings offer a promising route toward improving the environmental sustainability and economic viability of biorefinery and wastewater treatment technologies.

Catalytic systems derived from Au( I) and Cu(I) precatalysts bearing phosphine and N-heterocyclic carbene (NHC) ligands are traditionally considered homogeneous. However, we demonstrate that under catalytically relevant conditions these systems undergo rapid and reversible metal–ligand bond cleavage, generating complex cocktails of molecular complexes, clusters, and nanoparticulate species. Using a combination of TEM analyses and poisoning experiments, we reveal that the identity of the dominant active species is not intrinsic to the metal/ligand pair but is critically reaction-dependent. For example, ligand-free nanoparticulate Cu species govern the Chan–Evans–Lam coupling, while molecular copper complexes dominate the Cu-AAC click reaction. In Au catalysis, ligandless nanoparticles are prevalent in A3-coupling and alkyne hydration, whereas in hydroamination the IMes ligand plays a striking promoting role within the cocktail, outperforming both phosphine-based and ligand-free systems. Inspired by this insight, we developed a simple, solvent- and silver-free Au/IMes protocol for alkyne hydroamination using bench-stable precursors. This study establishes the "cocktail of catalysts" paradigm as a fundamental concept in Au and Cu catalysis and highlights the need to reconsider traditional ligand design strategies when nanoparticulate species dynamically contribute to catalysis.

The unique phenomenon of high morphological diversity of quaternary phosphonium salts (QPSs) has been observed via electron and optical microscopy. The molecular structure of the QPSs, which differ by one methylene group, was shown to be reflected in the microstructure of the crystallized droplets. Here, we describe experimental datasets of scanning electron and optical microscopy images at different magnifications, illustrating the versatile microstructures of 19 homologous QPSs. The unique patterns that appear in the microscopy images of the QPS are related to the molecular structure. The described datasets of microscopy images are made openly available for scientific purpose and include hierarchical morphological patterns and fractal elements. Importantly, the datasets are suitable for both directions of machine learning exploration: recognizing molecular formulas from microscopy images and, conversely, predicting morphological patterns from molecular structures. This bidirectionality establishes a benchmark for bridging the molecule–morphology gap and advancing data-driven materials design.

In this research, a novel eco-friendly method for developing antibacterial sol–gel materials with microbial porous-forming templates is proposed. This approach compliments traditional surfactants with sustainable microbial templates, which are accessible, inexpensive, morphologically reproducible, and do not possess a toxic effect on the environment. Here, we synthesized a series of sol–gel materials using yeast and bacterial cells as templates. The best material based on the encapsulated bacteria Paracoccus yeei VKM B-3302 in methyltriethoxysilane and tetraethoxysilane in a ratio of 50/50 vol% after heat treatment at 800 °C was established after studying the parameters of release of the well-known cationic antiseptic octenidine dihydrochloride. It was shown that the release of octenidine from the formed silicon-containing materials occurred in two phases: rapid local release and subsequent prolonged action. This biphasic release enables a dual-function effect: rapid initial antiseptic action for immediate microbial decontamination, followed by sustained release for long-term microbial protection. The antibacterial properties were evaluated by inhibiting the growth of the phytopathogen Rhodococcus fascians VKM Ac-1462. The diameter of the inhibition zone was 12 mm and decreased to 11 mm within 25 days. The most effective material was loaded into the silicone sealant. A mixture containing 3–10% antibacterial material maintained the inhibition zone above 10 mm for 25 days, which emphasizes the high potential of the obtained material for medical applications such as implant treatments, antiseptic wound surfaces and industrial applications including medical filters and membranes. This study demonstrates the use of whole microbial cells as biotemplates for fabricating environmentally friendly sol–gel matrices with tunable porosity for the controlled release of antiseptics.

Cyanoarene photocatalysts such as 3DPAFIPN are widely employed in visible-light-driven transformations, yet their intrinsic structural dynamics under irradiation remain poorly understood. Here we reveal that 3DPAFIPN undergoes light-induced reconfiguration into distinct cyclized and radical substitution products, with the parent species ultimately decomposing under reaction conditions. Using a combination of Photo-Chem-ESI-MS and TLC-mapping, we directly monitored the evolution of catalytic forms and identified the cyclized derivatives as the true active species in thiol–yne–ene coupling. The major cyclized product was isolated and structurally confirmed by single-crystal X-ray diffraction, providing unambiguous experimental evidence for light-driven photocatalyst reconfiguration. Compared with the precursor, the cyclized catalysts display enhanced stability against cyanide substitution by radical intermediates and deliver higher efficiency and selectivity. This study establishes a generalizable workflow for dissecting dynamic photocatalyst behavior and extends the ReAct-Light concept, highlighting light-triggered structural evolution as a new design principle for adaptive and robust photocatalytic systems.

Dearomative addition reactions of indoles represent a prominent strategy for the synthesis of practically valuable indolines. However, implementing this approach typically requires prefunctionalization of the indole precursor with specific groups (e.g., alkynes, alkenes, or ketones) to facilitate the subsequent cyclization-dearomatization step. In this work, we report a three-component, thiol-yne-mediated dearomative vinylation of indoles that circumvents the need for preinstalling reactive fragments into the indole scaffold. The developed photoredox-catalyzed thiol-yne-heteroarene reaction employs Boc-protected indole-3-carboxylates as effective terminating agents for the radical cascade.

Hidden hardware degradation can silently undermine the reliability of scientific experiments across disciplines. Instruments that appear to function normally may, in fact, produce systematically distorted results for months or years, affecting research precision and data reproducibility. In this study, we identify and analyze a previously unrecognized source of experimental error: gradual and uneven aging of high-power light-emitting diodes (LEDs) used in photochemical equipment. Although the efficiency loss of LEDs is known in engineering, their unpredictable and uneven degradation has not been recognized as a cause of error in chemical experiments. To detect and quantify this effect, we developed a multidisciplinary diagnostic framework combining seven complementary methods: three quantitative measurements (photodiode readout, graphite calorimetry, and ferrioxalate actinometry), two reaction-based methods (Paternò–Büchi cycloaddition, photocatalytic aryl bromide reduction), and two visual assays (phenothiazine photochromism, and a nickel–thiolate 'photoclock' system). Together, these approaches revealed gradual declines in LED performance, leading to significant nonuniformity in light output and reaction yields. These findings demonstrate that unnoticed hardware aging can compromise reproducibility even in well-controlled laboratories, emphasizing the need for regular diagnostics in modern photochemical and automated systems. In addition to photochemistry, this framework exemplifies a general strategy for improving the precision, traceability, and long-term reliability of experimental measurements.