Theoretical Design Of Single-Atom Alloy Catalysts For Selective Hydrogenation Reactions

In the modern chemical industry,selective hydrogenation plays an indispensable role,widely applied in the production of intermediates,fine chemicals,pharmaceuticals,and other fields.It not only enhances product quality,reduces byproduct formation and production costs but also contributes to environmental protection and resource efficiency,crucial for driving the advancement of the chemical industry.Single-atom alloy（SAA）catalysts have shown promising prospects in selective hydrogenation due to their excellent hydrogen dissociation and spill-over abilities,uniform distribution of active sites,and 100%atomic utilization rate.However,with numerous metal combinations for SAAs,potential combinations reach into the hundreds or thousands.Traditional trial-and-error experiments and Density Functional Theory（DFT）calculations,while widely used,are costly and inefficient,limiting rapid screening of high-performance SAAs.Furthermore,incomplete understanding at the atomic/molecular level of SAA host-guest metal interactions and catalytic mechanisms constrains precise catalyst design and performance enhancement.Therefore,the rapid screening and rational design of single-atom alloy catalysts to enhance the performance of selective hydrogenation reactions targeting specific unsaturated functional groups emerge as a crucial topic with significant scientific value and practical application prospects in the modern chemical field.Building upon these challenges,this thesis focuses on the rational design of high-performance single-atom alloy catalysts for selective hydrogenation,guiding the directed transformation of unsaturated organic compounds and finely tuning around the mechanism of host-guest metal interactions in single-atom alloys.By combining density functional theory（DFT）calculations and machine learning（ML）techniques,we have achieved rapid screening of single-atom alloy catalysts with selective reactivity towards specific functional groups.Utilizing microkinetic simulation methods,we have investigated reaction rates and product selectivity under realistic reaction conditions,elucidating the mechanisms by which catalyst performance is influenced by reaction conditions.Employing a research strategy integrating in-situ experiments and theoretical calculations,we have revealed the mechanism of interaction between host and guest metals in single-atom alloy catalysis,as well as the relationship between the structure of intrinsic active sites and their performance.This comprehensive multiscale research approach offers efficient solutions and robust theoretical underpinnings for the deliberate engineering of high-performance SAA catalysts tailored for selective hydrogenation.The specific research includes:1.Machine-learning-assisted catalytic performance predictions of single-atom alloys for acetylene semi-hydrogenationBy integrating Density Functional Theory（DFT）calculations and Machine Learning（ML）predictions,we systematically investigated the catalytic performance of 70 single-atom alloy surfaces in the semi-hydrogenation of acetylene.The results revealed that single-atom alloys exhibit strong adsorption of acetylene and weaker adsorption of ethylene,showcasing their potential as excellent catalysts for acetylene semi-hydrogenation.Utilizing DFT-calculated C₂H_n hydrogenation barriers as a ML dataset,we constructed a ML model incorporating the physical and chemical properties of elements.The training results of the ML model demonstrated outstanding predictive performance with a Gradient Boosting Regression algorithm（R²=0.99,RMSE=0.02 e V）.Using the DFT/ML method,we predicted the hydrogenation energy barriers of C₂H_n on the surfaces of 70 SAAs.As a result,we identified five single-atom alloy catalysts（Pd₁Cu,Pt₁Cu,Cr₁Ag,Mn₁Ag,and V₁Au）with potential for excellent performance in catalyzing the semi-hydrogenation reaction of acetylene.,Validation of ML predictions showed errors less than 0.07 e V,indicating the excellent guiding role of ML-assisted methods in the rational prediction of transition state energy barriers.The effective screening method using ML algorithms in this study will facilitate the design of novel selective hydrogenation catalysts.2.Designing Efficient Single-Atom Alloy Catalysts for Selective C=O Hydrogenation:A First-Principles,Active Learning and Microkinetic StudyBuilding upon the aforementioned ML framework,we systematically investigated the adsorption preferences of crotonaldehyde on various single-atom alloy surfaces using DFT calculations and active learning methods.Through a combination of DFT calculations,ML predictions,and microkinetic simulations,we identified highly efficient single-atom alloy catalysts for the selective hydrogenation of crotonaldehyde.The training results of the ML model showed that a Gaussian Process Regression algorithm with six feature variables exhibited the best predictive accuracy.After three rounds of active learning iteration,with an uncertainty of 0.20 e V and an error of 0.10 e V,the specific adsorption of C=O bonds on Ti₁Au single-atom alloy was accurately predicted.ML feature engineering analysis and electronic structure calculations revealed that the electronegativity and coordination number of single-atom metals are key factors influencing the adsorption energies of crotonaldehyde in bothπ_CO andπ_CC modes.DFT calculations indicated that the reaction mechanism of crotonaldehyde on the Ti₁Au single-atom alloy exhibited lower C=O hydrogenation barriers（less than 0.60 e V）on Ti₁Au（320）and Ti₁Au（111）surfaces.Microkinetic simulations further confirmed close to 100%selectivity for crotonol within the temperature range of 373 to 553 K,highlighting the promising application potential of Ti₁Au single-atom alloy in selective hydrogenation of unsaturated aldehydes.This work demonstrates an effective approach combining DFT calculations,active learning,and microkinetic simulations,aiming to determine the optimal composition of single-atom alloy catalysts for the selective hydrogenation of specific unsaturated functional groups,effectively overcoming the limitations of individual methods.By integrating theoretical calculations and data-driven approaches,this study paves the way for efficient and targeted exploration in catalyst design.3.Rational design and precise synthesis of single-atom alloy catalysts for the selective hydrogenation of nitroarenesBased on prior research on the catalytic performance of different single-atom host-guest metal combinations,this study employs a combined approach of density functional theory（DFT）calculations and experimental methods to investigate the interaction mechanisms of 15 single-atom alloy host-guest metals and their impact on the selective hydrogenation performance of 4-nitrostyrene.Results from DFT calculations reveal that single-atom alloys with strong host-guest metal interactions facilitate N-O₁ bond cleavage,while the reaction energy barrier for hydrogenation is primarily influenced by the host metal.By delving into the intricacies of the reaction mechanism,we uncovered that the energy barriers linked to N-O₁ bond cleavage,C₈H₇NOH hydrogenation,and ethylene hydrogenation play crucial roles in shaping the catalytic efficacy in the selective hydrogenation of 4-nitrostyrene.A comprehensive analysis of these three key steps’energy barriers confirmed the outstanding activity and selectivity of Ru₁Ni and Ir₁Ni single-atom alloys.Among them,Ir₁Ni single-atom catalyst exhibited the best performance,making it the most promising candidate for catalysis.Moreover,we meticulously synthesize a range of Ni-based single-atom alloys using the wet impregnation technique,followed by a thorough evaluation of their effectiveness in catalyzing the selective hydrogenation of4-nitrostyrene to 4-aminostyrene.As anticipated,Ir₁Ni single-atom alloy demonstrates extraordinary catalytic performance（yield>96%）.In-situ FT-IR experiments and DFT calculations further validate the unique host-guest metal interactions at the Ir-Ni interface of Ir₁Ni single-atom alloy,endowing it with excellent 4-NS selective hydrogenation capability.This work provides a theoretical basis for understanding the mechanism of host-guest metal interactions in single-atom alloys,as well as for the rational design and precise synthesis of single-atom catalysts with specific selective hydrogenation properties.