The Molecular Biological Mechanism of Hydrogen Therapy and Its Application in Spinal Cord Injury.

Hydrogen, which is a novel biomedical molecule, is currently the subject of extensive research involving animal experiments and in vitro cell experiments, and it is gradually being applied in clinical settings. Hydrogen has been proven to possess anti-inflammatory, selective antioxidant, and antiapoptotic effects, thus exhibiting considerable protective effects in various diseases. In recent years, several studies have provided preliminary evidence for the protective effects of hydrogen on spinal cord injury (SCI). This paper provides a comprehensive review of the potential molecular biology mechanisms of hydrogen therapy and its application in treating SCI, with an aim to better explore the medical value of hydrogen and provide new avenues for the adjuvant treatment of SCI.

15N Hyperpolarization of Metronidazole Antibiotic in Aqueous Media Using Phase-Separated Signal Amplification by Reversible Exchange with Parahydrogen.

Metronidazole is a prospective hyperpolarized MRI contrast agent with potential hypoxia sensing utility for applications in cancer, stroke, neurodegenerative diseases, etc. We demonstrate a pilot procedure for production of ∼30 mM hyperpolarized [15N3]metronidazole in aqueous media by using a phase-separated SABRE-SHEATH hyperpolarization method, with nitrogen-15 polarization exceeding 2.2% on all three 15N sites achieved in less than 2 min. The 15N polarization T1 of ∼12 min is reported for the 15NO2 group at the clinically relevant field of 1.4 T in the aqueous phase, demonstrating a remarkably long lifetime of the hyperpolarized state. The produced aqueous solution of [15N3]metronidazole that contained only ∼100 µM of residual Ir was deemed biocompatible via validation through the MTT colorimetric test for assessing cell metabolic activity using human embryotic kidney HEK293T cells. This low-cost and ultrafast hyperpolarization procedure represents a major advance for the production of a biocompatible HP [15N3]metronidazole (and potentially other hyperpolarized drugs) formulation for MRI sensing applications.

Computation-Driven Rational Design of Self-Assembled Short Peptides for Catalytic Hydrogen Production.

Self-assembling peptides represent a captivating area of study in nanotechnology and biomaterials. This interest is largely driven by their unique properties and the vast application potential across various fields such as catalytic functions. However, design complexities, including high-dimensional sequence space and structural diversity, pose significant challenges in the study of such systems. In this work, we explored the possibility of self-assembled peptides to catalyze the hydrolysis of hydrosilane for hydrogen production using ab initio calculations and carried out wet-lab experiments to confirm the feasibility of these catalytic reactions under ambient conditions. Further, we delved into the nuanced interplay between sequence, structural conformation, and catalytic activity by combining modeling with experimental techniques such as transmission electron microscopy and nuclear magnetic resonance and proposed a dual mode of the microstructure of the catalytic center. Our results reveal that although research in this area is still at an early stage, the development of self-assembled peptide catalysts for hydrogen production has the potential to provide a more sustainable and efficient alternative to conventional hydrogen production methods. In addition, this work also demonstrates that a computation-driven rational design supplemented by experimental validation is an effective protocol for conducting research on functional self-assembled peptides.

A LAMP-based hydrogen ion selective electrochemical sensor for highly sensitive detection of Mycoplasma pneumoniae.

A concise and rapid detection method for Mycoplasma pneumoniae is urgently required due to its severe impact on human health. To meet such a need, this study proposed and constructed an innovative point-of-care testing (POCT) platform that consists of a hydrogen ion-selective loop-mediated isothermal amplification (H+-LAMP) sensor and an electrochemical detection device. The H+-LAMP sensor successfully integrated the working and reference electrodes and converted the H+ generated during the LAMP process into an electrochemical signal. High sensitivity and stability for pathogen detection were also achieved by treating the working electrode with an electrodeposited polyaniline solid contact layer and by using an ion-selective membrane. As a result, the sensor shows a sensitivity of 68.26 mV per pH, a response time of less than 2 s, and a potential drift of less than 5 mV within one hour, which well meets the urgent need. The results also demonstrated that the detection limit for Mycoplasma pneumoniae was lowered to 1 copy per µL, the nucleic acid extraction and detection process could be completed in 30 minutes, and the impact of interfering ions on the sensor was negligible. Validation with 20 clinical samples yielded satisfactory results. More importantly, the storage lifespan of such an electrochemical sensor is over seven days, which is a great advantage for on-site pathogen detection. Therefore, the hydrogen ion-selective sensor constructed in this investigation is particularly suitable as a core component for instant pathogen detection platforms.

Enhancing the catalytic H2 production performance of magnetic Ni-Fe2O3-C catalyst in biomass steam gasification using electromagnetic induction heating.

In this study, a novel magnetic Ni-Fe2O3-C catalyst combined with electromagnetic induction heating in biomass steam gasification was proposed to enhance H2 production. Better catalytic performance for H2 production was observed with the Ni-Fe2O3-C catalyst under induction heating, resulting in an increase in H2 yield from 735.1 to 2271.2 mL/g-biomass (a 209.1 % enhancement). SEM, TGA and XRD analysis demonstrated a significant decrease in coking deposition, caking, and particle agglomeration of the Ni-Fe2O3-C catalyst under induction heating, while maintaining more active sites. Importantly, the benefits of induction heating were also applicable to different magnetic catalysts like Ni-Al2O3-C, Ni-ZrO2-C, and Ni-MgO-C. Experimental results revealed a logarithmic correlation between the increase in H2 yields due to induction heating and the magnetic saturation (Ms) of the catalysts. The Ni-Fe2O3-C catalyst, with a high Ms of 50.9 emu/g, showed the highest catalytic activity for H2 production under induction heating in this study.

Prediction of Individual Gas Yields of Supercritical Water Gasification of Lignocellulosic Biomass by Machine Learning Models.

Supercritical water gasification (SCWG) of lignocellulosic biomass is a promising pathway for the production of hydrogen. However, SCWG is a complex thermochemical process, the modeling of which is challenging via conventional methodologies. Therefore, eight machine learning models (linear regression (LR), Gaussian process regression (GPR), artificial neural network (ANN), support vector machine (SVM), decision tree (DT), random forest (RF), extreme gradient boosting (XGB), and categorical boosting regressor (CatBoost)) with particle swarm optimization (PSO) and a genetic algorithm (GA) optimizer were developed and evaluated for prediction of H2, CO, CO2, and CH4 gas yields from SCWG of lignocellulosic biomass. A total of 12 input features of SCWG process conditions (temperature, time, concentration, pressure) and biomass properties (C, H, N, S, VM, moisture, ash, real feed) were utilized for the prediction of gas yields using 166 data points. Among machine learning models, boosting ensemble tree models such as XGB and CatBoost demonstrated the highest power for the prediction of gas yields. PSO-optimized XGB was the best performing model for H2 yield with a test R2 of 0.84 and PSO-optimized CatBoost was best for prediction of yields of CH4, CO, and CO2, with test R2 values of 0.83, 0.94, and 0.92, respectively. The effectiveness of the PSO optimizer in improving the prediction ability of the unoptimized machine learning model was higher compared to the GA optimizer for all gas yields. Feature analysis using Shapley additive explanation (SHAP) based on best performing models showed that (21.93%) temperature, (24.85%) C, (16.93%) ash, and (29.73%) C were the most dominant features for the prediction of H2, CH4, CO, and CO2 gas yields, respectively. Even though temperature was the most dominant feature, the cumulative feature importance of biomass characteristics variables (C, H, N, S, VM, moisture, ash, real feed) as a group was higher than that of the SCWG process condition variables (temperature, time, concentration, pressure) for the prediction of all gas yields. SHAP two-way analysis confirmed the strong interactive behavior of input features on the prediction of gas yields.

Discrimination of Aspartic and Isoaspartic Acid Residues in Peptides by Tandem Mass Spectrometry with Hydrogen Attachment Dissociation.

Long-lived proteins undergo chemical modifications that can cause age-related diseases. Among these chemical modifications, isomerization is the most difficult to identify. Isomerization often occurs at the aspartic acid (Asp) residues. In this study, we used tandem mass spectrometry equipped with a newly developed ion activation method, hydrogen attachment dissociation (HAD), to analyze peptides containing Asp isomers. Although HAD preferentially produces [cn + 2H]+ and [zm + 2H]+ via N-Cα bond cleavage, [cn + 58 + 2H]+ and [zm - 58 + 2H]+ originate from the fragmentation of the isoAsp residue. Notably, [cn + 58 + 2H]+ and [zm - 58 + 2H]+ could be used as diagnostic fragment ions for the isoAsp residue because these fragment ions did not originate from the Asp residue. The detailed fragmentation mechanism was investigated by computational analysis using density functional theory. According to the results, hydrogen attachment to the carbonyl oxygen in the isoAsp residue results in the Cα-Cß bond cleavage. The experimental and theoretical joint study indicates that the present method allows us to discriminate Asp and isoAsp residues, including site identification of the isoAsp residue. Moreover, we demonstrated that the molar ratio of peptide isomers in the mixture could be estimated from their fragment ion abundance. Therefore, tandem mass spectrometry with HAD is a useful method for the rapid discrimination and semiquantitative analysis of peptides containing isoAsp residues.

Tandem hydrogenation/hydrogenolysis of furfural to 2-methylfuran over multifunctional metallic Cu nanoparticles supported ZIF-8 catalyst.

Catalytic transfer hydrogenation (CTH), that employs protic solvents as hydrogen sources to alleviate the use of molecular hydrogen H2, has gained great attention. This work, reports multifunctional, metallic Cu nanoparticles supported ZIF-8 material for CTH of furfural to a highly valued fuel additive, 2-methylfuran (2-MF) using 2-propanol. Of all as-synthesized xCu(yM)/ZIF-8 catalysts with varied NaBH4 concentration (yM) and Cu loading (x), 11Cu(1.5 M)/ZIF-8 exhibited higher catalytic activity with > 99 % FAL conversion and 93.9 % 2-MF selectivity. This is ascribed to its high specific surface area, and existence of optimum amount of Lewis acid-base sites along with Cu0 species, which are responsible for hydrogenation of furfural to furfuryl alcohol and subsequent hydrogenolysis to produce 2-MF. The present work reports a highly efficient and stable, metal-MOF hybrid material for CTH of FAL to 2-MF, which is one among the best reports available in literature, therewith suggests a promising approach for bio-oil upgradation.

Influence of the Pd Oxidation State in PdNi Thin Films on Surface Acoustic Wave Hydrogen Sensing Performance.

PdNi alloy thin films demonstrate exceptional hydrogen sensing performance and exhibit significant potential for application in surface acoustic wave (SAW) hydrogen sensors. However, the long-term stability of SAW H2 sensors utilizing PdNi films as catalysts experiences a substantial decrease during operation. In this paper, X-ray photoelectron spectroscopy (XPS) is employed to investigate the failure mechanisms of PdNi thin films under operational conditions. The XPS analysis reveals that the formation of PdO species on PdNi thin films plays a crucial role in the failure of hydrogen sensing. Additionally, density functional theory (DFT) calculations indicate that hydrogen atoms encounter a diffusion energy barrier during the penetration process from the PdNiOx surface to the subsurface region. The identification of PdNi film failure mechanisms through XPS and DFT offers valuable insights into the development of gas sensors with enhanced long-term stability. Guided by these mechanisms, we propose a method to restore the hydrogen sensing response time and magnitude to a certain extent by reducing the partially oxidized surface of the PdNi alloy under a hydrogen atmosphere at 70 °C, thereby restoring Pd to its metallic state with zero valence.

Innovative Thermocatalytic H2 Sensor with Double-Sided Pd Nanocluster Films on an Ultrathin Mica Substrate.

Hydrogen (H2) is crucial in the future global energy landscape due to its eco-friendly properties, but its flammability requires precise monitoring. This study introduces an innovative thermocatalytic H2 sensor utilizing ultrathin mica sheets as substrates, coated on both sides with Pd nanocluster (NC) films. The ultrathin mica substrate ensures robustness and flexibility, enabling the sensor to withstand high temperatures and mechanical deformation. Additionally, it simplifies the fabrication process by eliminating the need for complex microelectro-mechanical systems (MEMS) technology. Constructed through cluster beam deposition, the sensor exhibits exceptional characteristics, including a wide concentration range (from 500 ppm to 4%), rapid response and recovery times (3.1 and 2.4 s for 1% H2), good selectivity, high stability, and repeatability. The operating temperature can be as low as 40 °C, achieving remarkably low power consumption. The study explores the impact of double-sided versus single-sided catalytic layers, revealing significantly higher sensitivity and response with the double-sided configuration due to the increased catalytic surface area. Additionally, the research investigates the relationship between the deposition amount of Pd NCs and the sensor's sensitivity, identifying an optimal value that maximizes performance without excessive use of Pd. The sensor's innovative design and excellent performance position it as a promising candidate for meeting the demands of a hydrogen-based energy economy.

Critical Sensing Modalities for Hydrogen: Technical Needs and Status of the Field to Support a Changing Energy Landscape.

Decarbonization of the energy system is a key aspect of the energy transition. Energy storage in the form of chemical bonds has long been viewed as an optimal scheme for energy conversion. With advances in systems engineering, hydrogen has the potential to become a low cost, low emission, energy carrier. However, hydrogen is difficult to contain, it exhibits a low flammability limit (>40000 ppm or 4%), low ignition energy (0.02 mJ), and it is a short-lived climate forcer. Beyond commercially available sensors to ensure safety through spot checks in enclosed environments, new sensors are necessary to support the development of low emission infrastructure for production, transmission, storage, and end use. Efficient scalable broad area hydrogen monitoring motivates lowering the detection limit below that (10 ppm) of best in class commercial technologies. In this perspective, we evaluate recent advances in hydrogen gas sensing to highlight technologies that may find broad utility in the hydrogen sector. It is clear in the near term that a sensor technology suite is required to meet the variable constraints (e.g., power, size/weight, connectivity, cost) that characterize the breadth of the application space, ranging from industrial complexes to remote pipelines. This perspective is not intended to be another standard hydrogen sensor review, but rather provide a critical evaluation of technologies with detection limits preferably below 1 ppm and low power requirements. Given projections for rapid market growth, promising techniques will also be amenable to rapid development in technical readiness for commercial deployment. As such, methods that do not meet these requirements will not be considered in depth.

Comparison of carrier gases for the separation and quantification of mineral oil hydrocarbon (MOH) fractions using online coupled high performance liquid chromatography-gas chromatography-flame ionisation detection.

On-line coupled high performance liquid chromatography-gas chromatography-flame ionisation detection (HPLC-GC-FID) was used to compare the effect of hydrogen, helium and nitrogen as carrier gases on the chromatographic characteristics for the quantification of mineral oil hydrocarbon (MOH) traces in food related matrices. After optimisation of chromatographic parameters nitrogen carrier gas exhibited characteristics equivalent to hydrogen and helium regarding requirements set by current guidelines and standardisation such as linear range, quantification limit and carry over. Though nitrogen expectedly led to greater peak widths, all required separations of standard compounds were sufficient and humps of saturated mineral oil hydrocarbons (MOSH) and aromatic mineral oil hydrocarbons (MOAH) were appropriate to enable quantitation similar to situations where hydrogen or helium had been used. Slightly increased peak widths of individual hump components did not affect shapes and widths of the MOSH and MOAH humps were not significantly affected by the use of nitrogen as carrier gas. Notably, nitrogen carrier gas led to less solvent peak tailing and smaller baseline offset. Overall, nitrogen may be regarded as viable alternative to hydrogen or helium and may even extend the range of quantifiable compounds to highly volatile hydrocarbon eluting directly after the solvent peak.

Conjugated Polymer/Recombinant Escherichia coli Biohybrid Systems for Photobiocatalytic Hydrogen Production.

Biohybrid photocatalysts are composite materials that combine the efficient light-absorbing properties of synthetic materials with the highly evolved metabolic pathways and self-repair mechanisms of biological systems. Here, we show the potential of conjugated polymers as photosensitizers in biohybrid systems by combining a series of polymer nanoparticles with engineered Escherichia coli cells. Under simulated solar light irradiation, the biohybrid system consisting of fluorene/dibenzo [b,d]thiophene sulfone copolymer (LP41) and recombinant E. coli (i.e., a LP41/HydA BL21 biohybrid) shows a sacrificial hydrogen evolution rate of 3.442 mmol g-1 h-1 (normalized to polymer amount). It is over 30 times higher than the polymer photocatalyst alone (0.105 mmol g-1 h-1), while no detectable hydrogen was generated from the E. coli cells alone, demonstrating the strong synergy between the polymer nanoparticles and bacterial cells. The differences in the physical interactions between synthetic materials and microorganisms, as well as redox energy level alignment, elucidate the trends in photochemical activity. Our results suggest that organic semiconductors may offer advantages, such as solution processability, low toxicity, and more tunable surface interactions with the biological components over inorganic materials.

Degradation of pharmaceuticals and other emerging pollutants employing bi-metal catalysts/magnesium and/or (green) hydrogen in aqueous solution.

Contaminations by pharmaceuticals, personal care products, and other emerging pollutants in water resources have become a seriously burgeoning issue of global concern in the first third of the twenty-first century. As societal reliance on pharmaceuticals continues to escalate, the inadvertent introduction of these substances into water reservoirs poses a consequential environmental threat. Therefore, the aim of this study was to investigate reductive degradation, particularly, catalytic hydrogenation regarding model pollutants such as diclofenac (DCF), ibuprofen (IBP), 17α-ethinylestradiol (EE2), or bisphenol-A (BPA), respectively,  in aqueous solutions at lab scale. Iron bimetals (zero valent iron, ZVI, and copper, Cu, or nickel, Ni) as well as zero valent magnesium (Mg, ZVM) in combination with  rhodium, Rh, or palladium, Pd, as hydrogenation catalysts (HK), were investigated. Studies were executed through various short-term batch experiments, with multiple sample collections, over a total range of 120 min. The results indicated that DCF was attenuated at over 90 % when exposed to Fe-Cu or a Fe-Ni bimetal (applied as a single model pollutant). However, when DCF was part of a mixture alongside with IBP, EE2, and BPA, the attenuation efficacy decreased to 79 % with Fe-Cu and 23 % with Fe-Ni. Conversely, both IBP and BPA exhibit notably low attenuation levels with both bimetals, less than 50 %, both deployed as single substances or in mixtures. No reaction (degradation) products could be identified employing LC-MS, but sometimes a release of the parent pollutant when applying an acetic acid buffer could be noted to a certain extent, suggesting adsorption processes on corrosion products such as iron hydroxide and/or oxides. Surprisingly, Mg in combination with Rh (Rh-HK) or Pd (Pd-HK) showed a significantly rapid decrease in the concentrations of DCF, EE2, and BPA, in part up to approximately 100 %, that is, within a few minutes only in part due to hydrogenation degradation reactions (related reaction products could actually be identified by LC-MS; adsorption processes were not observed here). Moreover, kinetic modeling of the DCF degradation with Mg-Rh-HK was conducted at different temperatures (15 °C, 20 °C, 25 °C, 35 °C) and varied initial concentrations (2.5 mg/L, 5.0 mg/L, 7.5 mg/L, 10.0 mg/L). The outcomes prove that the degradation of DCF at the Rh-HK's surface followed a modified first-order kinetics, most probably by catalytic hydrodehalogenation and subsequent hydrogenation of the aromatic moieties (molecular hydrogen was provided by the corrosion of Mg). From the determined reaction rate constants at four different temperatures, the activation energy was estimated to be 59.6 kJ/mol by means of the Arrhenius equation what is in good agreement with similar results reported in the literature. This coupled hydrodehalogenation and hydrogenation approach may be upscaled into a new promising technical process for comprehensively removing such pharmaceuticals and similar pollutants in sewage plants in a single step, furthermore, even in combination with adsorption by activated carbon and/or ozonation which have already been established at some sewage plants in Switzerland and Germany recently.

The refinery of the future.

Fossil fuels-coal, oil and gas-supply most of the world's energy and also form the basis of many products essential for everyday life. Their use is the largest contributor to the carbon dioxide emissions that drive global climate change, prompting joint efforts to find renewable alternatives that might enable a carbon-neutral society by as early as 2050. There are clear paths for renewable electricity to replace fossil-fuel-based energy, but the transport fuels and chemicals produced in oil refineries will still be needed. We can attempt to close the carbon cycle associated with their use by electrifying refinery processes and by changing the raw materials that go into a refinery from fossils fuels to carbon dioxide for making hydrocarbon fuels and to agricultural and municipal waste for making chemicals and polymers. We argue that, with sufficient long-term commitment and support, the science and technology for such a completely fossil-free refinery, delivering the products required after 2050 (less fuels, more chemicals), could be developed. This future refinery will require substantially larger areas and greater mineral resources than is the case at present and critically depends on the capacity to generate large amounts of renewable energy for hydrogen production and carbon dioxide capture.

In-silico-assisted derivatization of triarylboranes for the catalytic reductive functionalization of aniline-derived amino acids and peptides with H2.

Cheminformatics-based machine learning (ML) has been employed to determine optimal reaction conditions, including catalyst structures, in the field of synthetic chemistry. However, such ML-focused strategies have remained largely unexplored in the context of catalytic molecular transformations using Lewis-acidic main-group elements, probably due to the absence of a candidate library and effective guidelines (parameters) for the prediction of the activity of main-group elements. Here, the construction of a triarylborane library and its application to an ML-assisted approach for the catalytic reductive alkylation of aniline-derived amino acids and C-terminal-protected peptides with aldehydes and H2 is reported. A combined theoretical and experimental approach identified the optimal borane, i.e., B(2,3,5,6-Cl4-C6H)(2,6-F2-3,5-(CF3)2-C6H)2, which exhibits remarkable functional-group compatibility toward aniline derivatives in the presence of 4-methyltetrahydropyran. The present catalytic system generates H2O as the sole byproduct.

Spectroscopic analysis of the sum-frequency response of the carbon-hydrogen stretching modes in collagen type I.

We studied the origin of the vibrational signatures in the sum-frequency generation (SFG) spectrum of fibrillar collagen type I in the carbon-hydrogen stretching regime. For this purpose, we developed an all-reflective, laser-scanning SFG microscope with minimum chromatic aberrations and excellent retention of the polarization state of the incident beams. We performed detailed SFG measurements of aligned collagen fibers obtained from rat tail tendon, enabling the characterization of the magnitude and polarization-orientation dependence of individual tensor elements Xijk2 of collagen's nonlinear susceptibility. Using the three-dimensional atomic positions derived from published crystallographic data of collagen type I, we simulated its Xijk2 elements for the methylene stretching vibration and compared the predicted response with the experimental results. Our analysis revealed that the carbon-hydrogen stretching range of the SFG spectrum is dominated by symmetric stretching modes of methylene bridge groups on the pyrrolidine rings of the proline and hydroxyproline residues, giving rise to a dominant peak near 2942 cm-1 and a shoulder at 2917 cm-1. Weak asymmetric stretches of the methylene bridge group of glycine are observed in the region near 2870 cm-1, whereas asymmetric CH2-stretching modes on the pyrrolidine rings are found in the 2980 to 3030 cm-1 range. These findings help predict the protein's nonlinear optical properties from its crystal structure, thus establishing a connection between the protein structure and SFG spectroscopic measurements.

Effects of hydrogen permeation on the mechanical characteristics of electroless nickel-plated free-cutting steel for application to the hydrogen valves of hydrogen fuel cell electric vehicles.

Electroless nickel plating is a suitable technology for the hydrogen industry because electroless nickel can be mass-produced at a low cost. Investigating in a complex environment where hydrogen permeation and friction/wear work simultaneously is necessary to apply it to hydrogen valves for hydrogen fuel cell vehicles. In this research, the effects of hydrogen permeation on the mechanical characteristics of electroless nickel-plated free-cutting steel (SUM 24L) were investigated. Due to the inherent characteristics of electroless nickel plating, the damage (cracks and delamination of grain) and micro-particles by hydrogen permeation were clearly observed at the grain boundaries and triple junctions. In particular, the cracks grew from grain boundary toward the intergranualr. This is because the grain boundaries and triple junctions are hydrogen permeation pathways and increasing area of the hydrogen partial pressure. As a result, its surface roughness increased by a maximum of two times, and its hardness and adhesion strength decreased by hydrogen permeation. In particular, hydrogen permeation increased the friction coefficient of the electroless nickel-plated layer, and the damage caused by adhesive wear was significantly greater, increasing the wear depth by up to 5.7 times. This is believed to be due to the decreasing in wear resistance of the electroless nickel plating layer damaged by hydrogen permeation. Nevertheless, the Vickers hardness and the friction coefficient of the electroless nickel plating layer were improved by about 3 and 5.6 times, respectively, compared with those of the free-cutting steel. In particular, the electroless nickel-plated specimens with hydrogen embrittlement exhibited significantly better mechanical characteristics and wear resistance than the free-cutting steel.

Leveraging Language Model Multitasking To Predict C-H Borylation Selectivity.

C-H borylation is a high-value transformation in the synthesis of lead candidates for the pharmaceutical industry because a wide array of downstream coupling reactions is available. However, predicting its regioselectivity, especially in drug-like molecules that may contain multiple heterocycles, is not a trivial task. Using a data set of borylation reactions from Reaxys, we explored how a language model originally trained on USPTO_500_MT, a broad-scope set of patent data, can be used to predict the C-H borylation reaction product in different modes: product generation and site reactivity classification. Our fine-tuned T5Chem multitask language model can generate the correct product in 79% of cases. It can also classify the reactive aromatic C-H bonds with 95% accuracy and 88% positive predictive value, exceeding purpose-developed graph-based neural networks.

First-principles calculation on electronic properties of hydrogen evolution reaction of Ni-based electrode surfaces with different monatomic doping.

At present, the hydrogen evolution reaction (HER) of Ni-based electrode has an important influence on water electrolysis hydrogen production technology, involving complex electrochemical process of electrode. In this project, Materials Studio (MS) software was used to design and construct Ni-based electrode surface (NES) models with monatomic Mo, Co, Fe, Cr doping, and the NES models attached 1 H atom and 2H atoms were denoted as the NES-H models and NES-2H model, respectively. Then the first-principles calculation was carried out. The results showed that the doping of different atoms can effectively change the work function of the pure Ni. In the charge transfer process of the four NES-2H models, the distance between the two H atoms is most affected by Mo doping, and they leave the Ni electrode surface as a single H ion, respectively, while the effect on Co, Fe and Cr doping is relatively consistent, and they leave the Ni electrode surface with H2 molecules, respectively. The doping of four single atoms changes the distance of valence band (VB) top and conduction band (CB) bottom from Fermi level in NES, NES-H and NES-2H models, and affects the HER, in which Mo doping has the greatest effect. The TDOS of the above models is mainly derived from the PDOS of the d orbitals of the doped atoms and Ni atoms. The results will provide a theoretical basis for the research and development of Ni-based electrode materials in HER.