For the most up-to-date list of publications, please refer to Google Scholar.
- Atomic structure generation from reconstructing structural fingerprintsFung, V., Jia, S., Zhang, J., Bi, S., Yin, J., and Ganesh, P.arxiv (2022)
Data-driven machine learning methods have the potential to dramatically accelerate the rate of materials design over conventional human-guided approaches. These methods would help identify or, in the case of generative models, even create novel crystal structures of materials with a set of specified functional properties to then be synthesized or isolated in the laboratory. For crystal structure generation, a key bottleneck lies in developing suitable atomic structure fingerprints or representations for the machine learning model, analogous to the graph-based or SMILES representations used in molecular generation. However, finding data-efficient representations that are invariant to translations, rotations, and permutations, while remaining invertible to the Cartesian atomic coordinates remains an ongoing challenge. Here, we propose an alternative approach to this problem by taking existing non-invertible representations with the desired invariances and developing an algorithm to reconstruct the atomic coordinates through gradient-based optimization using automatic differentiation. This can then be coupled to a generative machine learning model which generates new materials within the representation space, rather than in the data-inefficient Cartesian space. In this work, we implement this end-to-end structure generation approach using atom-centered symmetry functions as the representation and conditional variational autoencoders as the generative model. We are able to successfully generate novel and valid atomic structures of sub-nanometer Pt nanoparticles as a proof of concept. Furthermore, this method can be readily extended to any suitable structural representation, thereby providing a powerful, generalizable framework towards structure-based generation.
- CO2-Assisted Oxidative Dehydrogenation of Propane over VOx/In2O3 Catalysts: Interplay between Redox Property and Acid-Base InteractionsJiang, X., Lis, B., Purdy, S., Paladugu, S., Fung, V., Quan, W., Bao, Z., Yang, W., He, Y., Sumpter, B., and others, .ACS Catalysis 12, 11239—11252, (2022)
In this work, a series of VOx-loaded In2O3 catalysts were prepared, and their catalytic performance was evaluated for CO2-assisted oxidative dehydrogenation of propane (CO2-ODHP) and compared with In2O3 alone. The optimal composition is obtained on 3.4V/In2O3 (surface V density of 3.4V nm-2), which exhibited not only a higher C3H6 selectivity than other V/In catalysts and In2O3 under isoconversion conditions but also an improved reaction stability. To elucidate the catalyst structure-activity relationship, the VOx/In2O3 catalysts were characterized by chemisorption [NH3-temperature-programmed desorption (TPD), NH3-diffuse reflectance infrared Fourier transform spectroscopy (DRIFTS), CO2-TPD, and CO2-DRIFTS], H2-temperature-programmed reduction (TPR), in situ Raman spectroscopy, UV-vis diffuse reflectance spectroscopy, near-ambient pressure X-ray photoelectron spectroscopy, X-ray absorption spectroscopy, and further examined using density functional theory. The In-O-V structure and the extent of oligomerization, which play a crucial role in improving selectivity and stability, were identified in the VOx/In2O3 catalysts. In particular, the presence of surface VOx (i) inhibits the deep reduction of In2O3, thereby preserving the activity, (ii) neutralizes the excess basicity on In2O3, thus suppressing propane dry reforming and achieving a higher propylene selectivity, and (iii) introduces additional redox sites that participate in the dehydrogenation reaction by utilizing CO2 as a soft oxidant. The present work provides insights into developing selective, stable, and robust metal-oxide catalysts for CO2-ODHP by controlling the conversion of reagents via desired pathways through the interplay between acid-base interactions and redox properties.
- Hydrogen spillover and its relation to hydrogenation: observations on structurally defined single-atom sitesHulsey, M., Fung, V., Hou, X., Wu, J., and Yan, N.Angewandte Chemie International Edition 12, 11239—11252, (2022)
Hydrogen spillover, involving the transfer of H atoms from metal sites onto the catalyst support, is ubiquitous in chemical processes such as catalytic hydrogenation and hydrogen storage. Atomic level information concerning the kinetics of this process, the structural evolution of catalysts during hydrogen spillover, as well as the nature of participation of the spilled over H in catalysis, remain vastly lacking. Here, we provide insights to those questions with a solubilized polyoxometalate-supported single-atom catalyst which allows for the use of characterization techniques generally inaccessible to the study of heterogeneous catalysts. Hydrogenation kinetics together with poisoning studies further reveal that hydrogen spillover can be either detrimental or beneficial for catalysis, the direction and magnitude of which depends mostly on the nature of the reducible functional group. Similar trends were observed on one of the most prototypical hydrogen spillover catalysts-Pt/WO3.
- Measuring and directing charge transfer in heterogenous catalystsZachman, M., Fung, V., Polo-Garzon, F., Cao, S., Moon, J., Huang, Z., Jiang, D., Wu, Z., and Chi, M.Nature Communications 13, 1—8, (2022)
Precise control of charge transfer between catalyst nanoparticles and supports presents a unique opportunity to enhance the stability, activity, and selectivity of heterogeneous catalysts. While charge transfer is tunable using the atomic structure and chemistry of the catalyst-support interface, direct experimental evidence is missing for three-dimensional catalyst nanoparticles, primarily due to the lack of a high-resolution method that can probe and correlate both the charge distribution and atomic structure of catalyst/support interfaces in these structures. We demonstrate a robust scanning transmission electron microscopy (STEM) method that simultaneously visualizes the atomic-scale structure and sub-nanometer-scale charge distribution in heterogeneous catalysts using a model Au-catalyst/SrTiO3-support system. Using this method, we further reveal the atomic-scale mechanisms responsible for the highly active perimeter sites and demonstrate that the charge transfer behavior can be readily controlled using post-synthesis treatments. This methodology provides a blueprint for better understanding the role of charge transfer in catalyst stability and performance and facilitates the future development of highly active advanced catalysts.
- Physically Informed Machine Learning Prediction of Electronic Density of StatesFung, V., Ganesh, P., and Sumpter, B.Chemistry of Materials 34, 4848—4855, (2022)
The electronic structure of a material, such as its density of states (DOS), provides key insights into its physical and functional properties and serves as a valuable source of high-quality features for many materials screening and discovery workflows. However, the computational cost of calculating the DOS, most commonly with density functional theory (DFT), becomes prohibitive for meeting high-fidelity or high-throughput requirements, necessitating a cheaper but sufficiently accurate surrogate. To fulfill this demand, we develop a general machine learning method based on graph neural networks for predicting the DOS purely from atomic positions, six orders of magnitude faster than DFT. This approach can effectively use large materials databases and be applied generally across the entire periodic table to materials classes of arbitrary compositional and structural diversity. We furthermore devise a highly adaptable scheme for physically informed learning which encourages the DOS prediction to favor physically reasonable solutions defined by any set of desired constraints. This functionality provides a means for ensuring that the predicted DOS is reliable enough to be used as an input to downstream materials screening workflows to predict more complex functional properties, which rely on accurate physical features.
- High-throughput predictions of metal-organic framework electronic properties: theoretical challenges, graph neural networks, and data explorationRosen, A., Fung, V., Huck, P., O'Donnell, C., Horton, M., Truhlar, D., Persson, K., Notestein, J., and Snurr, R.npj Computational Materials 8, 1—10, (2022)
With the goal of accelerating the design and discovery of metal-organic frameworks (MOFs) for electronic, optoelectronic, and energy storage applications, we present a dataset of predicted electronic structure properties for thousands of MOFs carried out using multiple density functional approximations. Compared to more accurate hybrid functionals, we find that the widely used PBE generalized gradient approximation (GGA) functional severely underpredicts MOF band gaps in a largely systematic manner for semi-conductors and insulators without magnetic character. However, an even larger and less predictable disparity in the band gap prediction is present for MOFs with open-shell 3d transition metal cations. With regards to partial atomic charges, we find that different density functional approximations predict similar charges overall, although hybrid functionals tend to shift electron density away from the metal centers and onto the ligand environments compared to the GGA point of reference. Much more significant differences in partial atomic charges are observed when comparing different charge partitioning schemes. We conclude by using the dataset of computed MOF properties to train machine-learning models that can rapidly predict MOF band gaps for all four density functional approximations considered in this work, paving the way for future high-throughput screening studies. To encourage exploration and reuse of the theoretical calculations presented in this work, the curated data is made publicly available via an interactive and user-friendly web application on the Materials Project.
- Revealing the interplay between intelligent behavior and surface reconstruction of non-precious metal doped SrTiO3 catalysts during methane combustionBao, Z., Fung, V., Moon, J., Hood, Z., Rochow, M., Kammert, J., Polo-Garzon, F., and Wu, Z.Catalysis Today 8, 1—10, (2022)
The impact of surface reconstruction of a model perovskite, SrTiO3 (STO), on CH4 activation for combustion and oxidative coupling was previously revealed that the reaction rate was proportional to the creation of Sr-terminated step sites. Doped perovskites (SrTi1-xMxO3, M=metal dopant) present yet another form of reconstruction throughout the surface and the bulk, where the metal dopant can migrate in and out of the perovskite lattice, also known as "intelligent behavior". In this work, understanding the interplay between perovskite surface reconstruction (surface termination) and the "intelligent behavior" is tackled for the first time, and the catalytic consequences are probed with CH4 combustion as a model reaction. A set of experimental techniques, including XRD, Raman spectroscopy, X-ray adsorption spectroscopy, kinetic measurements, as well as DFT calculations were used to understand the catalytic behavior of the reconstructed surfaces of Ni and Cu-doped STO for methane combustion. We found that during methane oxidation, the diffusion of Ni and Cu into the lattice due to the "intelligent behavior" is accompanied by Sr enrichment on the surface of the perovskite. This Sr-enrichment process is reversible when Cu or Ni species exsolute as clusters/nanoparticles upon H2 treatment. Such a surface reconstruction is found to greatly impact the catalytic activity of doped perovskites towards methane combustion.
- Inverse design of two-dimensional materials with invertible neural networksFung, V., Zhang, J., Hu, G., Ganesh, P., and Sumpter, B.npj Computational Materials 7, 1—9, (2021)
The ability to readily design novel materials with chosen functional properties on-demand represents a next frontier in materials discovery. However, thoroughly and efficiently sampling the entire design space in a computationally tractable manner remains a highly challenging task. To tackle this problem, we propose an inverse design framework (MatDesINNe) utilizing invertible neural networks which can map both forward and reverse processes between the design space and target property. This approach can be used to generate materials candidates for a designated property, thereby satisfying the highly sought-after goal of inverse design. We then apply this framework to the task of band gap engineering in two-dimensional materials, starting with MoS2. Within the design space encompassing six degrees of freedom in applied tensile, compressive and shear strain plus an external electric field, we show the framework can generate novel, high fidelity, and diverse candidates with near-chemical accuracy. We extend this generative capability further to provide insights regarding metal-insulator transition in MoS2 which are important for memristive neuromorphic applications, among others. This approach is general and can be directly extended to other materials and their corresponding design spaces and target properties.
- Single-Atom High-Temperature Catalysis on a Rh1O5 Cluster for Production of Syngas from MethaneTang, Y., Fung, V., Zhang, X., Li, Y., Nguyen, L., Sakata, T., Higashi, K., Jiang, D., and Tao, F.Journal of the American Chemical Society 143, 16566—16579, (2021)
Single-atom catalysts are a relatively new type of catalyst active for numerous reactions but mainly for chemical transformations performed at low or intermediate temperatures. Here we report that singly dispersed Rh1O5 clusters on TiO2 can catalyze the partial oxidation of methane (POM) at high temperatures with a selectivity of 97% for producing syngas (CO + H2) and high activity with a long catalytic durability at 650C. The long durability results from the substitution of a Ti atom of the TiO2 surface lattice by Rh1, which forms a singly dispersed Rh1 atom coordinating with five oxygen atoms (Rh1O5) and an undercoordinated environment but with nearly saturated bonding with oxygen atoms. Computational studies show the back-donation of electrons from the dz2 orbital of the singly dispersed Rh1 atom to the unoccupied orbital of adsorbed CHn (n > 1) results in the charge depletion of the Rh1 atom and a strong binding of CHn to Rh1. This strong binding decreases the barrier for activating C-H, thus leading to high activity of Rh1/TiO2. A cationic Rh1 single atom anchored on TiO2 exhibits a weak binding to atomic carbon, in contrast to the strong binding of the metallic Rh surface to atomic carbon. The weak binding of atomic carbon to Rh1 atoms and the spatial isolation of Rh1 on TiO2 prevent atomic carbon from coupling on Rh1/TiO2 to form carbon layers, making Rh1/TiO2 resistant to carbon deposition than supported metal catalysts for POM. The highly active, selective, and durable high-temperature single-atom catalysis performed at 650C demonstrates an avenue of application of single-atom catalysis to chemical transformations at high temperatures.
- Benchmarking graph neural networks for materials chemistryFung, V., Zhang, J., Juarez, E., and Sumpter, B.npj Computational Materials 7, 1—8, (2021)
Graph neural networks (GNNs) have received intense interest as a rapidly expanding class of machine learning models remarkably well-suited for materials applications. To date, a number of successful GNNs have been proposed and demonstrated for systems ranging from crystal stability to electronic property prediction and to surface chemistry and heterogeneous catalysis. However, a consistent benchmark of these models remains lacking, hindering the development and consistent evaluation of new models in the materials field. Here, we present a workflow and testing platform, MatDeepLearn, for quickly and reproducibly assessing and comparing GNNs and other machine learning models. We use this platform to optimize and evaluate a selection of top performing GNNs on several representative datasets in computational materials chemistry. From our investigations we note the importance of hyperparameter selection and find roughly similar performances for the top models once optimized. We identify several strengths in GNNs over conventional models in cases with compositionally diverse datasets and in its overall flexibility with respect to inputs, due to learned rather than defined representations. Meanwhile several weaknesses of GNNs are also observed including high data requirements, and suggestions for further improvement for applications in materials chemistry are discussed.
- New insights into the bulk and surface defect structures of ceria nanocrystals from neutron scattering studyLuo, S., Li, M., Fung, V., Sumpter, B., Liu, J., Wu, Z., and Page, K.Chemistry of Materials 33, 3959—3970, (2021)
Neutron diffraction and pair distribution function studies coupled with Raman spectroscopy have successfully unraveled the detailed oxygen defect structures of ceria nanocubes and nanorods. Two types of defect sites are revealed for the ceria nanocrystals: surface and bulk defects. It is proposed that the surface oxygen defects in both types of CeO2 nanocrystals are predominantly the partially reduced Ce3O5+x, with the bulk defect structures dominated by interstitial Frenkel-type oxygen vacancies. Ceria nanorods possess much higher concentration of surface oxygen defects relative to the nanocubes, albeit with only slightly higher concentration of bulk Frenkel-type oxygen vacancies. Upon annealing the nanorod sample at 600C under vacuum (~10-4 to 10-5 mbar), a partially reduced ceria phase with long-range oxygen vacancy ordering (Ce3O5+x) has been observed experimentally for the first time. This intriguing observation that surface defect phases can take on ordered defect sublattices under certain conditions is of great value in understanding the temperature-dependent catalytic performance of ceria nanocrystals. Furthermore, a drastic decrease of the surface vacancies in the ceria nanocrystals is observed upon exposure to SO2, especially for the nanorods, a likely origin for the sulfur poisoning effect on ceria-based materials. This study suggests that tailoring surface morphology is a promising strategy to control defect properties of ceria nanomaterials. It also provides fundamental insights to stabilize surface oxygen defects in CeO2 nanocrystals to achieve high redox performance under corrosive environments such as under SO2/SOx exposure.
- Work Function Engineering of 2D Materials: The Role of Polar Edge ReconstructionsHu, G., Fung, V., Huang, J., and Ganesh, P.The Journal of Physical Chemistry Letters 12, 2320—2326, (2021)
2D materials have attracted tremendous interest as functional materials because of their diverse and tunable properties, especially at their edges. A material's work function is a critical parameter in many electronic devices; however, a fundamental understanding and a path toward large alterations of the work function in 2D materials still remain elusive. Here, we report the first evidence for anisotropy of the work function in 2D MoS2 from first-principles calculations. We also demonstrate large work-function tunability (in the range of 3.45-6.29 eV) choosing the 2H phase of MoS2 as a model system by sampling various edge configurations. We furthermore reveal the origin of this work function anisotropy and tunability by extending the existing work function relation to the local dipole moment at surfaces of 3D materials to those at edges in 2D materials. We then use machine-learning approaches to correlate work function with edge structures. These results pave the way for intrinsic edge engineering for electronic and catalytic applications.
- Oxidative dehydrogenation of propane to propylene with soft oxidants via heterogeneous catalysisJiang, X., Sharma, L., Fung, V., Park, S., Jones, C., Sumpter, B., Baltrusaitis, J., and Wu, Z.ACS Catalysis 11, 2182—2234, (2021)
Oxidative dehydrogenation of propane to propylene can be achieved using conventional, oxygen-assisted dehydrogenation of propane (O2-ODHP) or via the use of soft oxidants, such as CO2, N2O, S-containing compounds, and halogens/halides. The major roles of soft oxidants include inhibiting overoxidation and improving propylene selectivity, which are considered to be current challenges in O2-assisted dehydrogenation. For both CO2- and N2O-ODHP reactions, significant efforts have been devoted to developing redox-active (e.g., chromium, vanadate, iron, etc.), nonredox-type main group metal oxide (e.g., group IIIA, gallium), and other transition metal/metal oxide catalysts (e.g., molybdenum, palladium platinum, rhodium, ruthenium, etc.), as well as zeolite-based catalysts with adjustable acidibase properties, unique pore structures, and topologies. Metal sulfides have shown promising performance in DHP, whereas the development of suitable catalysts has lagged for SO2- or S-assisted ODHP. Recently, significant efforts have been focused on homogeneous and heterogeneous ODHP using halogens (e.g., Br2, I2, Cl2, etc.) and hydrogen halides (e.g., HCl and HBr) for the development of facile processes for C3H6 synthesis. This Review aims to provide a critical, comprehensive review of recent advances in oxidative dehydrogenation of propane with these soft oxidants, particularly highlighting the current state of understanding of the following factors: (i) relationships between composition, structure, and catalytic performance, (ii) effects of the support, acidity, and promoters, (iii) reaction pathway and mechanistic insights, and (iv) the various roles of soft oxidants. Theoretical and computational insights toward understanding reaction mechanisms and catalyst design principles are also covered. Future research opportunities are discussed in terms of catalyst design and synthesis, deactivation and regeneration, reaction mechanisms, and alternative approaches.
- Machine learned features from density of states for accurate adsorption energy predictionFung, V., Hu, G., Ganesh, P., and Sumpter, B.Nature Communications 12, 1—11, (2021)
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.
- Nickel-Platinum Nanoparticles as Peroxidase Mimics with a Record High Catalytic EfficiencyXi, Z., Wei, K., Wang, Q., Kim, M., Sun, S., Fung, V., and Xia, X.Journal of the American Chemical Society 143, 2660—2664, (2021)
While nanoscale mimics of peroxidase have been extensively developed over the past decade or so, their catalytic efficiency as a key parameter has not been substantially improved in recent years. Herein, we report a class of highly efficient peroxidase mimic-nickel-platinum nanoparticles (Ni-Pt NPs) that consist of nickel-rich cores and platinum-rich shells. The Ni-Pt NPs exhibit a record high catalytic efficiency with a catalytic constant (Kcat) as high as 4.5 x 107 s-1, which is ~46- and 104-fold greater than the Kcat values of conventional Pt nanoparticles and natural peroxidases, respectively. Density functional theory calculations reveal that the unique surface structure of Ni-Pt NPs weakens the adsorption of key intermediates during catalysis, which boosts the catalytic efficiency. The Ni-Pt NPs were applied to an immunoassay of a carcinoembryonic antigen that achieved an ultralow detection limit of 1.1 pg/mL, hundreds of times lower than that of the conventional enzyme-based assay.
- In Situ Strong Metal--Support Interaction (SMSI) Affects Catalytic Alcohol ConversionPolo-Garzon, F., Blum, T., Bao, Z., Wang, K., Fung, V., Huang, Z., Bickel, E., Jiang, D., Chi, M., and Wu, Z.ACS Catalysis 11, 1938—1945, (2021)
Strong metal-support interactions (SMSIs) and catalyst deactivation have been heavily researched for decades by the catalysis community. The promotion of SMSIs in supported metal oxides is commonly associated with H2 treatment at high temperature (>500C), and catalyst deactivation is commonly attributed to sintering, leaching of the active metal, and overoxidation of the metal, as well as strong adsorption of reaction intermediates. Alcohols can reduce metal oxides, and thus we hypothesized that catalytic conversion of alcohols can promote SMSIs in situ. In this work we show, via IR spectroscopy of CO adsorption and electron energy loss spectroscopy (EELS), that during 2-propanol conversion over Pd/TiO2 coverage of Pd sites occurs due to SMSIs at low reaction temperatures (as low as ~190C). The emergence of SMSIs during the reaction (in situ) explains the apparent catalyst deactivation when the reaction temperature is varied. A steady-state isotopic transient kinetic analysis (SSITKA) shows that the intrinsic reactivity of the catalytic sites does not change with temperature when SMSI is promoted in situ; rather, the number of available active sites changes (when a TiOx layer migrates over Pd NPs). SMSI generated during the reaction fully reverses upon exposure to O2 at room temperature for ~15 h, which may have made their identification elusive up to now.
- Control of single-ligand chemistry on thiolated Au25 nanoclustersCao, Y., Fung, V., Yao, Q., Chen, T., Zang, S., Jiang, D., and Xie, J.Nature Communications 11, 1—7, (2020)
Diverse methods have been developed to tailor the number of metal atoms in metal nanoclusters, but control of surface ligand number at a given cluster size is rare. Here we demonstrate that reversible addition and elimination of a single surface thiolate ligand (-SR) on gold nanoclusters can be realized, opening the door to precision ligand engineering on atomically precise nanoclusters. We find that oxidative etching of [Au25SR18]- nanoclusters adds an excess thiolate ligand and generates a new species, [Au25SR19]0. The addition reaction can be reversed by CO reduction of [Au25SR19]0, leading back to [Au25SR18]- and eliminating precisely one surface ligand. Intriguingly, we show that the ligand shell of Au25 nanoclusters becomes more fragile and rigid after ligand addition. This reversible addition/elimination reaction of a single surface ligand on gold nanoclusters shows potential to precisely control the number of surface ligands and to explore new ligand space at each nuclearity.
- Descriptors for hydrogen evolution on single atom catalysts in nitrogen-doped grapheneFung, V., Hu, G., Wu, Z., and Jiang, D.The Journal of Physical Chemistry C 124, 19571—19578, (2020)
Single-atom catalysts (SACs) are a new research frontier in electrocatalysis such as in the hydrogen evolution reaction (HER). Recent theoretical and experimental studies have demonstrated that certain M-N-C (metal-nitrogen-carbon) based SACs exhibit excellent performance for HER. Here we report a new approach to tune HER activity for SACs by changing the size and dimensionality of the carbon substrate while maintaining the same coordination environment. We screen the 3d, 4d, and 5d transition metal SACs in N-doped 2D graphene and nanographenes of several sizes for HER using first-principles density functional theory (DFT). Nanographenes containing V, Rh, and Ir are predicted to have significantly enhanced HER activity compared to their 2D graphene counterparts. We turn to machine learning to accurately predict the free energy of hydrogen adsorption (GH) based on various descriptors and compressed sensing to identify key descriptors for activity, which can be used to further screen for additional candidates.
- Alcohol-induced low-temperature blockage of supported-metal catalysts for enhanced catalysisPolo-Garzon, F., Blum, T., Fung, V., Bao, Z., Chen, H., Huang, Z., Mahurin, S., Dai, S., Chi, M., and Wu, Z.ACS Catalysis 10, 8515—8523, (2020)
The partial or complete blockage of active sites of metal nanoparticles (NPs) on supported-metal catalysts has been of interest for tuning the stability, selectivity, and rate of reactions. Here, we show that Au-sites in Au/TiO2 surprisingly become blocked upon treatment in common alcohols (2-propanol and methanol), with 2-propanol causing a greater extent of blockage. Nearly 95% of Au-sites are covered after treatment in 2-propanol at room temperature, followed by desorption at 150C. Infrared spectroscopy of CO adsorption unambiguously confirms the occurrence of this phenomenon. Electron energy loss spectroscopy (EELS), temperature-programmed desorption (TPD), Raman spectroscopy, and DFT simulations suggest that the formation of carbon deposits from 2-propanol decomposition and/or the migration of a TiOx layer over the supported NPs may be responsible for the blockage of Au-sites. Nearly full coverage of Au NPs after treatment in 2-propanol led to negligible activity for catalytic CO oxidation, whereas partial retraction of the overlayer led to enhanced activity with time-on-stream, suggesting a self-activating catalytic performance.
- Stable Surface Terminations of a Perovskite Oxyhydride from First-PrinciplesWang, K., Fung, V., Wu, Z., and Jiang, D.The Journal of Physical Chemistry C 124, 18557—18563, (2020)
Successful synthesis of some perovskite oxyhydrides and their unique catalytic properties have recently attracted researchers' attention. However, their surface structure remains unclear. Here we identify stable surface terminations of a prototypical perovskite oxyhydride, BaTiO2.5H0.5, under catalytically relevant temperatures and pressures by using first-principles thermodynamics based on density functional theory. Five low-index facets, including (100), (010), (210), (011), and (211), and their various terminations for a total of 47 different surfaces have been examined for relative stability at different temperatures (700, 500, 300 K) and gas environments (10-15 < PO2 , 1 atm, 10-15 , PH2 , 100 atm). The most stable ones are found to be (010)-Ba2O2, (210)-Ti2O2, and (211)-Ba2O4H surface terminations. These polar surfaces are stabilized by charge compensation. This work provides important insights into the stable surfaces of perovskite oxyhydrides for future studies of their catalytic properties.
- Hydrogen in NanocatalysisFung, V., Hu, G., Wu, Z., and Jiang, D.The Journal of Physical Chemistry Letters 11, 7049—7057, (2020)
Hydrogen is ubiquitous in catalysis. It is involved in many important reactions such as water splitting, N2 reduction, CO2 reduction, and alkane activation. In this Perspective, we focus on the hydrogen atom and follow its electron as it interacts with a catalyst or behaves as part of a catalyst from a computational point of view. We present recent examples in both nanocluster and solid catalysts to elucidate the parameters governing the strength of the hydrogen-surface interactions based on site geometry and electronic structure. We further show the interesting behavior of hydride in nanometal and oxides for catalysis. The key take-home messages are: (1) the in-the-middle electronegativity and small size of hydrogen give it great versatility in interacting with active sites on nanoparticles and solid surfaces; (2) the strength of hydrogen binding to an active site on a surface is an important descriptor of the chemical and catalytic properties of the surface; (3) the energetics of the hydrogen binding is closely related to the electronic structure of the catalyst; (4) hydrides in nanoclusters and oxides and on surfaces offer unique reactivity for reduction reactions.
- Understanding the conversion of ethanol to propene on In2O3 from first principlesHuang, R., Fung, V., Wu, Z., and Jiang, D.Catalysis Today 350, 19—24, (2020)
It is highly desirable to convert bioethanol to value-added chemicals. As such, conversion of ethanol to propene (ETP) is attractive because propene is an important raw material for the production of plastics. In2O3 has shown promising catalytic performance for ETP conversion. However, the underlying mechanisms remain elusive. In this work, we use density functional theory (DFT) to investigate ETP reaction pathways on the In2O3 (110) surface. We find that the ETP reactions proceed through three major stages: ethanol to acetaldehyde, acetaldehyde to acetone, and acetone to propene. The ethanol-to-acetaldehyde step is kinetically facile. Comparing the two pathways from acetaldehyde to acetone, we show that the aldol reaction pathway via direct coupling of two acetaldehyde is more favorable than the acetate-ketonization pathway. The acetone-to-propene process is found to be the rate-limiting step of the overall reaction. This work provides a detailed mechanistic view of the ETP chemistry on In2O3(110) that paves the way for further exploration of effects such as surface termination, surface doping, and co-feeding of H2 and H2O on selectivity and catalyst stability.
- Harnessing strong metal--support interactions via a reverse routeWu, P., Tan, S., Moon, J., Yan, Z., Fung, V., Li, N., Yang, S., Cheng, Y., Abney, C., Wu, Z., and others, .Nature Communications 11, 1—10, (2020)
Engineering strong metal-support interactions (SMSI) is an effective strategy for tuning structures and performances of supported metal catalysts but induces poor exposure of active sites. Here, we demonstrate a strong metal-support interaction via a reverse route (SMSIR) by starting from the final morphology of SMSI (fully-encapsulated core-shell structure) to obtain the intermediate state with desirable exposure of metal sites. Using core-shell nanoparticles (NPs) as a building block, the Pd-FeOx NPs are transformed into a porous yolk-shell structure along with the formation of SMSIR upon treatment under a reductive atmosphere. The final structure, denoted as Pd-Fe3O4-H, exhibits excellent catalytic performance in semi-hydrogenation of acetylene with 100% conversion and 85.1% selectivity to ethylene at 80C. Detailed electron microscopic and spectroscopic experiments coupled with computational modeling demonstrate that the compelling performance stems from the SMSIR, favoring the formation of surface hydrogen on Pd instead of hydride.
- Predicting synthesizable multi-functional edge reconstructions in two-dimensional transition metal dichalcogenidesHu, G., Fung, V., Sang, X., Unocic, R., and Ganesh, P.npj Computational Materials 6, 1—9, (2020)
Two-dimensional (2D) transition metal dichalcogenides (TMDCs) have attracted tremendous interest as functional materials due to their exceptionally diverse and tunable properties, especially in their edges. In addition to the conventional armchair and zigzag edges common to hexagonal 2D materials, more complex edge reconstructions can be realized through careful control over the synthesis conditions. However, the whole family of synthesizable, reconstructed edges remains poorly studied. Here, we develop a computational approach integrating ensemble-generation, force-relaxation, and electronic-structure calculations to systematically and efficiently discover additional reconstructed edges and screen their functional properties. Using MoS2 as a model system, we screened hundreds of edge-reconstruction to discover over 160 reconstructed edges to be more stable than the conventional ones. More excitingly, we discovered nine new synthesizable reconstructred edges with record thermodynamic stability, in addition to successfully reproducing three recently synthesized edges. We also find our predicted reconstructed edges to have multi-functional properties-they show near optimal hydrogen evolution activity over the conventional edges, ideal for catalyzing hydrogen-evolution reaction (HER) and also exhibit half-metallicity with a broad variation in magnetic moments, making them uniquely suitable for nanospintronic applications. Our work reveals the existence of a wide family of synthesizable, reconstructed edges in 2D TMDCs and opens a new materials-by-design paradigm of 'intrinsic' edge engineering multifunctionality in 2D materials.
- The interplay between surface facet and reconstruction on isopropanol conversion over SrTiO3 nanocrystalsBao, Z., Fung, V., Polo-Garzon, F., Hood, Z., Cao, S., Chi, M., Bai, L., Jiang, D., and Wu, Z.Journal of Catalysis 384, 49—60, (2020)
Strontium titanate (SrTiO3) is an extensively investigated perovskite for various applications due to its optical, electrical and chemical properties. To gain an in-depth understanding of the active sites involved in heterogeneous catalysis over the broadly used SrTiO3 (STO), we studied a model reaction, isopropanol conversion, on three differently shape-controlled nanocrystals: cube, truncated cube and dodecahedra. SEM, XRD and XPS confirmed the morphology, phase and composition of STO shapes. Low energy ion scattering (LEIS) revealed the occurrence of surface reconstruction over STO shapes during O2 pretreatment at different temperatures. Based on the catalytic activities, scanning transmission electron microscopy images and density functional theory calculations, the step sites on STO derived from surface reconstruction were proposed to be the active sites for isopropanol conversion. This was further confirmed by steady state isotopic kinetic analysis (SSITKA) which demonstrated similar intrinsic turnover frequencies (TOFs) for the differently reconstructed STO shapes. It is concluded that the crystal facets impose an indirect effect on the catalysis of STO via controlling the degrees of surface reconstruction: the less stable STO (1 1 0) facet (dodecahedra) leads to more step sites after reconstruction and hence higher overall reaction rate than the more stable (1 0 0) facet (cube). This work highlights the important interplay between the crystal facet and surface reconstruction in controlling the nature and density of active sties and thus catalysis over complex oxides.
- Nature of reactive hydrogen for ammonia synthesis over a Ru/C12A7 electride catalystKammert, J., Moon, J., Cheng, Y., Daemen, L., Irle, S., Fung, V., Liu, J., Page, K., Ma, X., Phaneuf, V., and others, .Journal of the American Chemical Society 142, 7655—7667, (2020)
Recently, there have been renewed interests in exploring new catalysts for ammonia synthesis under mild conditions. Electride-based catalysts are among the emerging ones. Ruthenium particles supported on an electride composed of a mixture of calcium and aluminum oxides (C12A7) have attracted great attention for ammonia synthesis due to their facile ability in activating N2 under ambient pressure. However, the exact nature of the reactive hydrogen species and the role of electride support still remain elusive for this catalytic system. In this work, we report for the first time that the surface-adsorbed hydrogen, rather than the hydride encaged in the C12A7 electride, plays a major role in ammonia synthesis over the Ru/C12A7 electride catalyst with the aid of in situ neutron scattering techniques. Combining in situ neutron diffraction, inelastic neutron spectroscopy, density functional theory (DFT) calculation, and temperature-programmed reactions, the results provide direct evidence for not only the presence of encaged hydrides during ammonia synthesis but also the strong thermal and chemical stability of the hydride species in the Ru/C12A7 electride. Steady state isotopic transient kinetic analysis (SSITKA) of ammonia synthesis showed that the coverage of reactive intermediates increased significantly when the Ru particles were promoted by the electride form (coverage up to 84%) of the C12A7 support rather than the oxide form (coverage up to 15%). Such a drastic change in the intermediate coverage on the Ru surface is attributed to the positive role of electride support where the H2 poisoning effect is absent during ammonia synthesis over Ru. The finding of this work has significant implications for understanding catalysis by electride-based materials for ammonia synthesis and hydrogenation reactions in general.
- Radical chemistry and reaction mechanisms of propane oxidative dehydrogenation over hexagonal boron nitride catalystsZhang, X., You, R., Wei, Z., Jiang, X., Yang, J., Pan, Y., Wu, P., Jia, Q., Bao, Z., Bai, L., and others, .Angewandte Chemie International Edition 59, 8042—8046, (2020)
Although hexagonal boron nitride (h-BN) has recently been identified as a highly efficient catalyst for the oxidative dehydrogenation of propane (ODHP) reaction, the reaction mechanisms, especially regarding radical chemistry of this system, remain elusive. Now, the first direct experimental evidence of gas-phase methyl radicals (CH3.) in the ODHP reaction over boron-based catalysts is achieved by using online synchrotron vacuum ultraviolet photoionization mass spectroscopy (SVUV-PIMS), which uncovers the existence of gas-phase radical pathways. Combined with density functional theory (DFT) calculations, the results demonstrate that propene is mainly generated on the catalyst surface from the C-H activation of propane, while C2 and C1 products can be formed via both surface-mediated and gas-phase pathways. These observations provide new insights towards understanding the ODHP reaction mechanisms over boron-based catalysts.
- A new trick for an old support: Stabilizing gold single atoms on LaFeO3 perovskiteTian, C., Zhang, H., Zhu, X., Lin, B., Liu, X., Chen, H., Zhang, Y., Mullins, D., Abney, C., Shakouri, M., and others, .Applied Catalysis B: Environmental 261, 118178, (2020)
Single-atom catalysts (SACs) have shown great potential for achieving superior catalytic activity due to maximizing metal efficiency. The key obstacle in developing SACs lies in the availability of supports that can stabilize SACs. Here we report the first successful development of single gold (Au) atom catalysts supported on high-surface-area hierarchical perovskite oxides. The resulting Au single-atoms are extremely stable at calcination temperatures up to 700C in air and under reaction conditions. A high catalytic activity for CO oxidation and distinct self-activating property were also achieved. Furthermore, evidenced by theoretical calculations and experimental studies including X-ray absorption fine structures and in situ Fourier-transform infrared spectra, the surface Au active sites are confirmed to be predominately positively charged. This work provides a generalizable approach to fabricating highly stable Au single-atom catalysts with tunable catalytic performance, and we anticipate that this discovery will facilitate new possibilities for the development of single atom catalysts.
- Electronic band contraction induced low temperature methane activation on metal alloysFung, V., Hu, G., and Sumpter, B.Journal of Materials Chemistry A 8, 6057—6066, (2020)
The catalytic conversion of methane under mild conditions is an appealing approach to selectively produce value-added products from natural gas. Catalysts which can chemisorb methane can potentially overcome challenges associated with its high stability and achieve facile activation. Although transition metals can activate C-H bonds, chemisorption and low-temperature conversion remain elusive on these surfaces. The broad electronic bands of metals can only weakly interact with the methane orbitals, in contrast to specific transition metal oxide and supported metal cluster surfaces which are now recognized to form methane sigma-complexes. Here, we report methane chemisorption can, remarkably, occur on metal surfaces via electronic band contraction and localization from metal alloying. From a broad screening including single atom and intermetallic alloys in various substrates, we find early transition metals as promising metal solutes for methane chemisorption and low-temperature activation. These findings demonstrate a combinatorial diversity of possible candidates in earth abundant metal alloys with this attractive catalytic behavior.
- Perovskite-supported Pt single atoms for methane activationWan, Q., Fung, V., Lin, S., Wu, Z., and Jiang, D.Journal of Materials Chemistry A 8, 4362—4368, (2020)
ABO3 perovskites are increasingly being explored as catalysts, but it is unclear how they behave as supports for single atoms and how the subsequent single-atom catalysts can be employed for important reactions such as methane activation. Here we examine the stability of Pt single atoms (Pt1) on the commonly exposed (100) surfaces of SrBO3 perovskites (B = 3d transition metals) and their methane-adsorption properties by first principles density functional theory. We find that the stability and charge state of Pt1 on the SrBO3(100) surfaces are termination-sensitive. Due to polar compensation, Pt1 is negatively charged on the A termination but positively charged on the B termination. This charge state greatly impacts methane adsorption: negatively charged Pt1 on the A-termination chemisorbs methane (in some cases, dissociatively), but positively charged Pt1 on the B-termination adsorbs methane physically. Analysis of the density of states of the negatively charged Pt1 reveals that its sp states are key to methane chemisorption and C-H activation. Our work shows that polar compensation on the perovskite surfaces can be used to tune the charge state of a single atom for methane chemisorption and C-H activation.
- Real time monitoring of the dynamic intracluster diffusion of single gold atoms into silver nanoclustersZheng, K., Fung, V., Yuan, X., Jiang, D., and Xie, J.Journal of the American Chemical Society 141, 18977—18983, (2019)
Alloying metal materials with heterometal atoms is an efficient way to diversify the function of materials, but in-depth understanding of the dynamic heterometallic diffusion inside the alloying materials is rather limited, especially at the atomic level. Here, we report the real-time monitoring of the dynamic diffusion process of a single gold (Au) atom into an atomically precise silver nanocluster (Ag NC), Ag25(MHA)18 (MHA = 6-mercaptohexanoic acid), by using in situ UV-vis absorption spectroscopy in combination with mass and tandem mass spectrometry. We found that the Au heteroatom first replaces the Ag atom at the surface Ag2(MHA)3 motifs of Ag25(MHA)18. After that, the Au atom diffuses into the surface layer of the icosahedral Ag13 kernel and finally occupies the center of the alloy NCs to form the thermodynamically stable Au@Ag24(MHA)18 product. Density functional theory (DFT) calculations reveal that the key thermodynamic driving force is the preference of the Au heteroatom for the central site of alloy NCs. The real-time monitoring approach developed in this study could also be extended to other metal alloy systems to reveal the reaction dynamics of intracluster diffusion of heteroatoms, as well as the formation mechanisms of metal alloy nanomaterials.
- Methane Chemisorption on Oxide-Supported Pt Single AtomFung, V., Hu, G., Tao, F., and Jiang, D.ChemPhysChem 20, 2217—2220, (2019)
Methane chemisorption has been recently demonstrated on the rutile IrO2(110) surface. However, it remains unclear how the general requirements are for methane chemisorption or complexation with a single atom on an oxide surface. By exploring methane adsorption on Pt1 substitutionally doped on many rutile-type oxides using hybrid density functional theory, we show that the occupancy of the Pt dz2 orbital is the key to methane chemisorption. Pt single atom on the semiconducting or wide-gap oxides such as TiO2 and GeO2 strongly chemisorbs methane, because the empty Pt dz2 orbital is located in the gap and can effectively accept sigma-electron donation from the methane C-H bond. In contrast, Pt single atom on metallic oxides such as IrO2 and RuO2 does not chemisorb methane, because the Pt dz2 orbital strongly mixes with the support-oxide electronic states and become more occupied, losing its ability to chemisorb methane. This study sheds further light on the impact of the interaction between a Pt single atom and the oxide support on methane adsorption.
- Superior electrocatalytic hydrogen evolution at engineered non-stoichiometric two-dimensional transition metal dichalcogenide edgesHu, G., Fung, V., Sang, X., Unocic, R., and Ganesh, P.Journal of Materials Chemistry A 7, 18357—18364, (2019)
Two-dimensional transition metal dichalcogenide (TMDC) edges show activity for the catalytic hydrogen evolution reaction (HER), but further improvements require extrinsic doping, usually performed in an Edisonian manner. Herein we investigate if tuning the non-stoichiometric degree of the edges itself can improve HER activities. Using first-principles density functional theory (DFT), we study six non-stoichiometric MoSe2 edges that have been recently synthesized under a scanning transmission electron microscope (STEM). We find that non-stoichiometric edges can have near optimal HER activity over conventional stoichiometric edges. More excitingly, we find a strong linear correlation between Bader charges on H and the Gibbs free energy of hydrogen adsorption (GH) at these edges, providing a design principle for discovering better HER catalytic edges. In general, HER activity is not only influenced by the formation of H-Se/Mo chemical bonds as previously thought, but also by geometric reconstructions and charge redistribution. Our predictions open the door for engineering non-stoichiometric TMDC edges for superior HER activity.
- Density-Functional Tight-Binding for Platinum Clusters and Bulk: Electronic vs Repulsive ParametersLee, K., Van Vuong, Q., Fung, V., Jiang, D., and Irle, S.MRS Advances 4, 1821—1832, (2019)
We present a general purpose Pt-Pt density-functional tight-binding (DFTB) parameter for Pt clusters as well as bulk, using a genetic algorithm (GA) to automatize the parameterization effort. First we quantify the improvement possible by only optimizing the repulsive potential alone, and second we investigate the effect of improving the electronic parameter as well. During both parameterization efforts we employed our own training set and test sets, with one set containing ~20,000 spin-polarized DFT structures. We analyze the performance of our two DFTB Pt-Pt parameter sets against density functional theory (DFT) as well as an earlier DFTB Pt-Pt parameters. Our study sheds light on the role of both repulsive and electronic parameters with regards to DFTB performance.
- Elucidation of the reaction mechanism for high-temperature water gas shift over an industrial-type copper-chromium-iron oxide catalystPolo-Garzon, F., Fung, V., Nguyen, L., Tang, Y., Tao, F., Cheng, Y., Daemen, L., Ramirez-Cuesta, A., Foo, G., Zhu, M., and others, .Journal of the American Chemical Society 141, 7990—7999, (2019)
The water gas shift (WGS) reaction is of paramount importance for the chemical industry, as it constitutes, coupled with methane reforming, the main industrial route to produce hydrogen. Copper-chromium-iron oxide-based catalysts have been widely used for the high-temperature WGS reaction industrially. The WGS reaction mechanism by the CuCrFeOx catalyst has been debated for years, mainly between a "redox" mechanism involving the participation of atomic oxygen from the catalyst and an "associative" mechanism proceeding via a surface formate-like intermediate. In the present work, advanced in situ characterization techniques (infrared spectroscopy, temperature-programmed surface reaction (TPSR), near-ambient pressure XPS (NAP-XPS), and inelastic neutron scattering (INS)) were applied to determine the nature of the catalyst surface and identify surface intermediate species under WGS reaction conditions. The surface of the CuCrFeOx catalyst is found to be dynamic and becomes partially reduced under WGS reaction conditions, forming metallic Cu nanoparticles on Fe3O4. Neither in situ IR not INS spectroscopy detect the presence of surface formate species during WGS. TPSR experiments demonstrate that the evolution of CO2 and H2 from the CO/H2O reactants follows different kinetics than the evolution of CO2 and H2 from HCOOH decomposition (molecule mimicking the associative mechanism). Steady-state isotopic transient kinetic analysis (SSITKA) (CO + H216O -> CO + H218O) exhibited significant 16O/18O scrambling, characteristic of a redox mechanism. Computed activation energies for elementary steps for the redox and associative mechanism by density functional theory (DFT) simulations indicate that the redox mechanism is favored over the associative mechanism. The combined spectroscopic, computational, and kinetic evidence in the present study finally resolves the WGS reaction mechanism on the industrial-type high-temperature CuCrFeOx catalyst that is shown to proceed via the redox mechanism.
- Promotion of catalytic selectivity on transition metal oxide through restructuring surface latticeLiu, J., Fung, V., Wang, Y., Du, K., Zhang, S., Nguyen, L., Tang, Y., Fan, J., Jiang, D., and Tao, F.Applied Catalysis B: Environmental 237, 957—969, (2018)
Pursuit of high catalytic selectivity is paramount in the design of catalysts for green chemical processes towards minimizing the production of undesired products. We demonstrated that catalytic selectivity for production of alkene through oxidative dehydrogenation of alkane on transition metal oxides can be promoted through tailoring the surface lattice of the oxide catalyst. Selectivity for production of ethylene through oxidative dehydrogenation (ODH) of ethane on Co3O4 nanocrystals can be substantially increased by 30%-35% via temperature-mediated reconstruction of surface lattice of Co3O4. Co3O4 nanocrystals formed at 800C leads to smooth, flat crystal plane with predominantly exposed (111) facet in contrast to high Miller index (311) facet of Co3O4 formed at <700C, revealed by environmental transmission electron microscopy. Isotope-labelled experiments suggest that the higher catalytic selectivity on the (111) facet results from the lower activity of its surface lattice oxygen atoms. Consistent with these experimental results, DFT calculations suggest low activity of surface lattice oxygen atoms and high activation barriers for adsorption and dissociation of C-H bond on the (111) surface in contrast to (311). Upon the activation of C-H on (311), the stronger binding of ethylene on more active, under-coordinated surface lattice oxygen atoms of (311) forms a robust "deprotonated ethylene glycol"-like intermediate on (311) with a rate-limiting desorption barrier to the formation of ethylene. Compared to (311), the kinetically favorable desorption of bound ethylene species from (111) surface well rationalized the higher selectivity for production of ethylene on (111) than (311). These findings demonstrate that temperature-mediated tailoring of the surface lattice for a transition metal oxide nanocatalyst is a promising approach in pursuing high selectivity in oxidative dehydrogenation of hydrocarbons.
- Insights into interfaces, stability, electronic properties, and catalytic activities of atomically precise metal nanoclusters from first principlesTang, Q., Hu, G., Fung, V., and Jiang, D.Accounts of Chemical Research 237, 957—969, (2018)
Atomically precise, ligand-protected metal nanoclusters are of great interest for their well-defined structures, intriguing physicochemical properties, and potential applications in catalysis, biology, and nanotechnology. Their structure precision provides many opportunities to correlate their geometries, stability, electronic properties, and catalytic activities by closely integrating theory and experiment. In this Account, we highlight recent theoretical advances from our efforts to understand the metal-ligand interfaces, the energy landscape, the electronic structure and optical absorption, and the catalytic applications of atomically precise metal nanoclusters. We mainly focus on gold nanoclusters. The bonding motifs and energetics at the gold-ligand interfaces are two main interests from a computational perspective. For the gold-thiolate interface, the -RS-Au-SR- staple motif is not always preferred; in fact, the bridging motif (-SR-) is preferred at the more open facets such as Au(100) and Au(110). This finding helps understand the diversity of the gold-thiolate motifs for different core geometries and sizes. A great similarity is demonstrated between gold-thiolate and gold-alkynyl interfaces, regarding formation of the staple-type motifs with PhC-C- as an example. In addition, N-heterocyclic carbenes (NHCs) without bulky groups also form the staple-type motif. Alkynyls and bulky NHCs have the strongest binding with the gold surface from comparing 27 ligands of six types, suggesting a potential to synthesize NHC-protected gold clusters. The energy landscape of nanosystems is usually complex, but experimental progress in synthesizing clusters of the same Au-S composition with different R groups and isomers of the same Aun(SR)m formula have made detailed theoretical analyses of energetic contributions possible. Ligand-ligand interactions turn out to play an important role in the cluster stability, while metastable isomers can be obtained via kinetic control. Although the superatom-complex theory is the starting point to understand the electronic structure of atomically precise gold clusters, other factors also greatly affect the orbital levels that manifest themselves in the experimental optical absorption spectra. For example, spin-orbit coupling needs to be included to reproduce the splitting of the HOMO-LUMO transition observed experimentally for Au25(SR)18-, the poster child of the family. In addition, doping can lead to structural changes and charge states that do not follow the superatomic electron count. Atomically precise metal nanoclusters are an ideal system for understanding nanocatalysis due to their well-defined structures. Active sites and catalytic mechanisms are explored for selective hydrogenation and hydrogen evolution on thiolate-protected gold nanoclusters with and without dopants. The behavior of H in nanogold is analyzed in detail, and the most promising site to attract H is found to be coordinately unsaturated Au atoms. Many insights have been gained from first-principles studies of atomically precise, ligand-protected gold nanoclusters. Interesting and important questions remaining to be addressed are pointed out in the end.
- New bonding model of radical adsorbate on lattice oxygen of perovskitesFung, V., Wu, Z., and Jiang, D.The Journal of Physical Chemistry Letters 9, 6321—6325, (2018)
A new model of bonding between radical adsorbates and lattice oxygens is proposed that considers both the adsorbate-oxygen bonding and the weakening of the metal-oxygen bonds. Density functional calculations of SrMO3 perovskites for M being 3d, 4d, and 5d transition metals are used to correlate the bulk electronic structure with the surface-oxygen reactivity. Occupation of the metal-oxygen antibonding states, examined via the crystal orbital Hamilton population (COHP), is found to be a useful bulk descriptor that correlates with the vacancy formation energy of the lattice oxygen and its hydrogen adsorption energy. Analysis of density-of-states and COHP indicates that H adsorption energy is a combined result of formation of the O-H bond and the weakening of the surface metal-oxygen bond due to occupation of the metal-oxygen antibonding states by the electron from H. This insight will be useful in understanding the trends in surface reactivity of perovskites and transition-metal oxides in general.
- Golden single-atomic-site platinum electrocatalystsDuchesne, P., Li, Z., Deming, C., Fung, V., Zhao, X., Yuan, J., Regier, T., Aldalbahi, A., Almarhoon, Z., Chen, S., and others, .Nature Materials 17, 1033—1039, (2018)
Bimetallic nanoparticles with tailored structures constitute a desirable model system for catalysts, as crucial factors such as geometric and electronic effects can be readily controlled by tailoring the structure and alloy bonding of the catalytic site. Here we report a facile colloidal method to prepare a series of platinum-gold (PtAu) nanoparticles with tailored surface structures and particle diameters on the order of 7 nm. Samples with low Pt content, particularly Pt4Au96, exhibited unprecedented electrocatalytic activity for the oxidation of formic acid. A high forward current density of 3.77 A mgPt-1 was observed for Pt4Au96, a value two orders of magnitude greater than those observed for core-shell structured Pt78Au22 and a commercial Pt nanocatalyst. Extensive structural characterization and theoretical density functional theory simulations of the best-performing catalysts revealed densely packed single-atom Pt surface sites surrounded by Au atoms, which suggests that their superior catalytic activity and selectivity could be attributed to the unique structural and alloy-bonding properties of these single-atomic-site catalysts.
- Understanding the impact of surface reconstruction of perovskite catalysts on CH4 activation and combustionPolo-Garzon, F., Fung, V., Liu, X., Hood, Z., Bickel, E., Bai, L., Tian, H., Foo, G., Chi, M., Jiang, D., and others, .ACS Catalysis 8, 10306—10315, (2018)
Methane conversion has received renewed interest due to the rapid growth in production of shale gas. Methane combustion for power generation and transportation is one of the alternatives for methane utilization. However, complete conversion of methane is critical to minimize negative environmental effects from unburned methane, whose noxious effect is 25 times greater than that of CO2. Although perovskite catalysts have high thermal stability, their low activities for methane combustion prevent them from being utilized on a commercial basis. In this work, we show the impact from reconstruction of surface and subsurface monolayers of perovskite catalysts on methane combustion, using SrTiO3 (STO) as a model perovskite. Several STO samples obtained through different synthetic methods and subjected to different postsynthetic treatments were tested for methane combustion. Through top surface characterization, kinetic experiments (including isotope labeling experiments) and density functional theory calculations, it is shown that both surface segregation of Sr and creation of step surfaces of STO can impact the rate of methane combustion over an order of magnitude. This work highlights the role of surface reconstruction in tuning perovskite catalysts for methane activation.
- Low-temperature activation of methane on doped single atoms: descriptor and predictionFung, V., Tao, F., and Jiang, D.Physical Chemistry Chemical Physics 20, 22909—22914, (2018)
Catalytic transformation of methane under mild conditions remains a grand challenge. Fundamental understanding of C-H activation of methane is crucial for designing a catalyst for the utilization of methane at low temperature. Recent experiments show that strong methane chemisorption on oxides of precious metals leads to facile C-H activation. However, only a very few such oxides are capable (for example, IrO2 and PdO). Here we show for the first time that strong methane chemisorption and facile C-H activation can be accomplished by single transition-metal atoms on TiO2, some of which are even better than IrO2. Using methane adsorption energy as a descriptor, we screened over 30 transition-metal single atoms doped on TiO2 for the chemisorption of methane by replacing a surface Ti atom with a single atom of another transition metal. It is found that the adsorption energies of methane on a single atom of Pd, Rh, Os, Ir, and Pt doped on rutile TiO2(110) are greater than or similar to those on rutile IrO2(110), a benchmark for the chemisorption of methane on transition oxides. Electronic structure analysis uncovered orbital overlap and mixing between methane and the single atom, as well as significant localization of the charge between the molecule and the surface, demonstrating chemical bonding of CH4 to doped single atoms. Facile C-H dissociation has been found on the single-atom sites with the transition state energies lower than desorption energies. Our computational studies predict that Pd, Rh, Os, Ir, and Pt single atoms on rutile TiO2(110) can activate C-H of methane at a temperature lower than 25C.
- Synthesis of water-soluble [Au25(SR)18]- using a stoichiometric amount of NaBH4Chen, T., Fung, V., Yao, Q., Luo, Z., Jiang, D., and Xie, J.Journal of the American Chemical Society 140, 11370—11377, (2018)
Determination of the stoichiometry of reactions is a pivotal step for any chemical reactions toward a desirable product, which has been successfully achieved in organic synthesis. Here, we present the first precise determination of the stoichiometry for the reactions toward gold nanoparticle formation in the sodium borohydride reduction method. Leveraging on the real-time mass spectrometry technique, we have determined a precise balanced reaction, 32/x [Au(SR)]x + 8 e- = [Au25(SR)18]- + 7 [Au(SR)2]- (here SR denotes a thiolate ligand), toward a stoichiometric synthesis of water-soluble [Au25(SR)18]-, where 8 electrons (from reducing agents) are sufficient to react with every 32 Au atoms, leading to the formation of high-purity [Au25(SR)18]-. More interestingly, by real-time monitoring of the growth process of thiolate-protected Au nanoclusters, we have successfully identified an important yet missing byproduct, [Au(SR)2]-. This study not only provides a new method for Au nanocluster synthesis using only a stoichiometric amount of reducing agent in aqueous solutions (although the synthesis of organic-soluble Au nanoclusters might require a more delicate design of synthetic chemistry) but also promotes the mechanistic understandings of the Au nanocluster growth process.
- Trends of alkane activation on doped cobalt (II, III) oxide from first principlesFung, V., Tao, F., and Jiang, D.ChemCatChem 10, 244—249, (2018)
The surface doping of a metal oxide can tune its catalytic performance, but it remains unclear how the tuning depends on the dopant type and the surface facet. Herein we study doped Co3O4 (111) and (311) surface facets using first-principles density functional theory (DFT) to obtain general descriptors for oxygen reactivity (which include vacancy formation energy and hydrogen adsorption energy) and correlate them to ethane C-H activation energy as a measure of the catalytic performance. The periodic trends of the dopants are investigated for a total of 20 dopants, namely, the elements from K to Ge. We find strong linear correlations between the oxygen reactivity descriptors and the computed energy barriers. We also discover a strong surface facet sensitivity among certain dopants such that different surface orientations and sites lead to different or even the opposite dopant performance. This work provides a useful guide for dopant performance in ethane activation on the two very different Co3O4 surfaces.
- Revealing isoelectronic size conversion dynamics of metal nanoclusters by a noncrystallization approachYao, Q., Fung, V., Sun, C., Huang, S., Chen, T., Jiang, D., Lee, J., and Xie, J.Nature Communications 9, 1—11, (2018)
Atom-by-atom engineering of nanomaterials requires atomic-level knowledge of the size evolution mechanism of nanoparticles, which remains one of the greatest mysteries in nanochemistry. Here we reveal atomic-level dynamics of size evolution reaction of molecular-like nanoparticles, i.e., nanoclusters (NCs) by delicate mass spectrometry (MS) analyses. The model size-conversion reaction is [Au23(SR)16]- -> [Au25(SR)18]- (SR=thiolate ligand). We demonstrate that such isoelectronic (valence electron count is 8 in both NCs) size-conversion occurs by a surface-motif-exchange-induced symmetry-breaking core structure transformation mechanism, surfacing as a definitive reaction of [Au23(SR)16]- + 2[Au2(SR)3]- -> [Au25(SR)18]- + 2 [Au(SR)2]-. The detailed tandem MS analyses further suggest the bond susceptibility hierarchies in feed and final Au NCs, shedding mechanistic light on cluster reaction dynamics at atomic level. The MS-based mechanistic approach developed in this study also opens a complementary avenue to X-ray crystallography to reveal size evolution kinetics and dynamics.
- Single rhodium atoms anchored in micropores for efficient transformation of methane under mild conditionsTang, Y., Li, Y., Fung, V., Jiang, D., Huang, W., Zhang, S., Iwasawa, Y., Sakata, T., Nguyen, L., Zhang, X., Frenkel, A., and Tao, F.Nature Communications 9, 1—11, (2018)
Catalytic transformation of CH4 under a mild condition is significant for efficient utilization of shale gas under the circumstance of switching raw materials of chemical industries to shale gas. Here, we report the transformation of CH4 to acetic acid and methanol through coupling of CH4, CO and O2 on single-site Rh1O5 anchored in microporous aluminosilicates in solution at <150C. The activity of these singly dispersed precious metal sites for production of organic oxygenates can reach about 0.10 acetic acid molecules on a Rh1O5 site per second at 150C with a selectivity of ~70% for production of acetic acid. It is higher than the activity of free Rh cations by >1000 times. Computational studies suggest that the first C-H bond of CH4 is activated by Rh1O5 anchored on the wall of micropores of ZSM-5; the formed CH3 then couples with CO and OH, to produce acetic acid over a low activation barrier.
- Understanding Methanol Coupling on SrTiO3 from First PrinciplesHuang, R., Fung, V., Zhang, Y., Mullins, D., Wu, Z., and Jiang, D.The Journal of Physical Chemistry C 122, 7210—7216, (2018)
Perovskites are interesting materials for catalysis due to their great tunability. However, the correlation of many reaction processes to the termination of a perovskite surface is still unclear. In this study, we use the methanol coupling reaction on the SrTiO3(100) surface as a probe reaction to investigate direct C-C coupling from a computational perspective. We use density functional theory to assess methanol adsorption, C-H activation, and direct C-C coupling reactions on the SrTiO3(100) surface of different terminations. We find that, although methanol molecules dissociatively adsorb on both A and B terminations with similar strength, the dehydrogenation and C-C coupling reactions have significantly lower activation energies on the B termination than on the A termination. The predicted formation of methoxy and acetate on the SrTiO3(100) B termination can well explain the ambient-pressure XPS data of methanol on the single-crystal SrTiO3(100) surface at 250C. This work suggests that a choice of B termination of perovskites would be beneficial for the C-C coupling reaction of methanol.
- Precise control of alloying sites of bimetallic nanoclusters via surface motif exchange reactionYao, Q., Feng, Y., Fung, V., Yu, Y., Jiang, D., Yang, J., and Xie, J.Nature Communications 8, 1—11, (2017)
Precise control of alloying sites has long been a challenging pursuit, yet little has been achieved for the atomic-level manipulation of metallic nanomaterials. Here we describe utilization of a surface motif exchange (SME) reaction to selectively replace the surface motifs of parent [Ag44(SR)30]4-(SR=thiolate) nanoparticles (NPs), leading to bimetallic NPs with well-defined molecular formula and atomically-controlled alloying sites in protecting shell. A systematic mass (and tandem mass) spectrometry analysis suggests that the SME reaction is an atomically precise displacement of SR-Ag(I)-SR-protecting modules of Ag NPs by the incoming SR-Au(I)-SR modules, giving rise to a core-shell [Ag32@Au12(SR)30]4-. Theoretical calculation suggests that the thermodynamically less favorable core-shell Ag@Au nanostructure is kinetically stabilized by the intermediate Ag20 shell, preventing inward diffusion of the surface Au atoms. The delicate SME reaction opens a door to precisely control the alloying sites in the protecting shell of bimetallic NPs with broad utility.
- Exploring perovskites for methane activation from first principlesFung, V., Polo-Garzon, F., Wu, Z., and Jiang, D.Catalysis Science & Technology 8, 702—709, (2017)
The diversity of perovskites offers many opportunities for catalysis, but an overall trend has been elusive. Using density functional theory, we studied a large set of perovskites in the ABO3 formula via descriptors of oxygen reactivity such as vacancy formation energy, hydrogen adsorption energy, and the first C-H activation energy of methane. It was found that changing the identity of B within a period increases the oxygen reactivity from the early to late transition metals, while changing A within a group has a much smaller effect on oxygen reactivity. Within the same group, B in the 3d period has the most reactive lattice oxygen compared to the 4d or 5d period. Some perovskites display large differences in reactivity for different terminations. Further examination of the second C-H bond breaking on these perovskites revealed that larger A cations and non-transition metal B cations have higher activation energies, which is conducive to the formation of coupling products instead of oxidation to CO or CO2. Balance between the first C-H bond breaking and methyl desorption suggests a just right oxygen reactivity as described by the hydrogen adsorption energy. These insights may help in designing better perovskite catalysts for methane activation.
- Understanding seed-mediated growth of gold nanoclusters at molecular levelYao, Q., Yuan, X., Fung, V., Yu, Y., Leong, D., Jiang, D., and Xie, J.Nature Communications 8, 1—11, (2017)
The continuous development of total synthesis chemistry has allowed many organic and biomolecules to be produced with known synthetic history-that is, a complete set of step reactions in their synthetic routes. Here, we extend such molecular-level precise reaction routes to nanochemistry, particularly to a seed-mediated synthesis of inorganic nanoparticles. By systematically investigating the time-dependent abundance of 35 intermediate species in total, we map out relevant step reactions in a model size growth reaction from molecularly pure Au25 to Au44 nanoparticles. The size growth of Au nanoparticles involves two different size-evolution processes (monotonic LaMer growth and volcano-shaped aggregative growth), which are driven by a sequential 2-electron boosting of the valence electron count of Au nanoparticles. Such fundamental findings not only provide guiding principles to produce other sizes of Au nanoparticles (e.g., Au38), but also represent molecular-level insights on long-standing puzzles in nanochemistry, including LaMer growth, aggregative growth, and digestive ripening.
- Controlling reaction selectivity through the surface termination of perovskite catalystsPolo-Garzon, F., Yang, S., Fung, V., Foo, G., Bickel, E., Chisholm, M., Jiang, D., and Wu, Z.Angewandte Chemie International Edition 129, 9952—9956, (2017)
Although perovskites have been widely used in catalysis, tuning of their surface termination to control reaction selectivity has not been well established. In this study, we employed multiple surface-sensitive techniques to characterize the surface termination (one aspect of surface reconstruction) of SrTiO3 (STO) after thermal pretreatment (Sr enrichment) and chemical etching (Ti enrichment). We show, by using the conversion of 2-propanol as a probe reaction, that the surface termination of STO can be controlled to greatly tune catalytic acid/base properties and consequently the reaction selectivity over a wide range, which is not possible with single-metal oxides, either SrO or TiO2. Density functional theory (DFT) calculations explain well the selectivity tuning and reaction mechanism on STO with different surface termination. Similar catalytic tunability was also observed on BaZrO3, thus highlighting the generality of the findings of this study.
- General structure--reactivity relationship for oxygen on transition-metal oxidesFung, V., Tao, F., and Jiang, D.The Journal of Physical Chemistry Letters 8, 2206—2211, (2017)
Despite many recent developments in designing and screening catalysts for improved performance, transition-metal oxides continue to prove to be challenging due to the myriad inherent complexities of the systems and the possible morphologies that they can exhibit. Herein we propose a structural descriptor, the adjusted coordination number (ACN), which can generalize the reactivity of the oxygen sites over the many possible surface facets and defects of a transition-metal oxide. We demonstrate the strong correlation of this geometric descriptor with the electronic and energetic properties of the active sites on five facets of four transition-metal oxides. We then use the structural descriptor to predict C-H activation energies to show the great potential of using ACN for the prediction of catalytic performance. This study presents a first look into quantifying the relation between active site structure and reactivity of transition-metal-oxide catalysts.
- Acid--base reactivity of perovskite catalysts probed via conversion of 2-propanol over titanates and zirconatesFoo, G., Polo-Garzon, F., Fung, V., Jiang, D., Overbury, S., and Wu, Z.ACS Catalysis 7, 4423—4434, (2017)
Although perovskite catalysts are well-known for their excellent redox property, their acid-base reactivity remains largely unknown. To explore the potential of perovskites in acid-base catalysis, we made a comprehensive investigation in this work on the acid-base properties and reactivity of a series of selected perovskites, SrTiO3, BaTiO3, SrZrO3, and BaZrO3, via a combination of various approaches including adsorption microcalorimetry, in situ FTIR spectroscopy, steady state kinetic measurements, and density functional theory (DFT) modeling. The perovskite surfaces are shown to be dominated with intermediate and strong basic sites with the presence of some weak Lewis acid sites, due to the preferred exposure of SrO/BaO on the perovskite surfaces as evidenced by low energy ion scattering (LEIS) measurements. Using the conversion of 2-propanol as a probe reaction, we found that the reaction is more selective to dehydrogenation over dehydration due to the dominant surface basicity of the perovskites. Furthermore, the adsorption energy of 2-propanol (Hads,2-propanol) is found to be related to both a bulk property (tolerance factor) and the synergy between surface acid and base sites. The results from in situ FTIR and DFT calculations suggest that both dehydration and dehydrogenation reactions occur mainly through the E1cB pathway, which involves the deprotonation of the alcohol group to form a common alkoxy intermediate on the perovskite surfaces. The results obtained in this work pave a path for further exploration and understanding of acid-base catalysis over perovskite catalysts.
- Exploring structural diversity and fluxionality of Ptn(n=10-13) clusters from first-principlesFung, V., and Jiang, D.The Journal of Physical Chemistry C 121, 10796—10802, (2017)
Subnanometer transition-metal clusters have been shown to possess catalytic activity that is size-dependent. It has been suggested that the fluxionality of these small clusters may be closely related to their catalytic activity. Here we use basin-hopping global optimization with density functional theory (DFT) to study the energy landscape of Ptn(n=10-13) clusters. We analyze a large set of local minima obtained from the DFT-based global optimization. We find that Pt10 is unique with respect to the other studied sizes in its structural landscape, which shows a single, distinct structural motif corresponding to a tetrahedral global minimum. In contrast, Pt11-13 all display characteristics of high fluxionality with the presence of multiple significantly differing structural features in the low-energy region, as characterized by coordination number, interatomic distances, and shape. These observations demonstrate the structural diversity and fluxionality of the subnanometer Pt clusters that will have important implications for catalysis.
- Tuning catalytic selectivity of oxidative catalysis through deposition of nonmetallic atoms in surface lattice of metal oxideLiu, J., Zhang, S., Zhou, Y., Fung, V., Nguyen, L., Jiang, D., Shen, W., Fan, J., and Tao, F.ACS Catalysis 6, 4218—4228, (2016)
Catalytic selectivity for producing an ideal product is a key topic for chemical transformations through heterogeneous catalysis. Tuning catalytic selectivity by integrating the second metal to form an alloy has been well demonstrated in the literature. Here we report a method to tune catalytic selectivity in oxidative catalysis on another category of heterogeneous catalysts, transition-metal oxides. By choosing the oxidative dehydrogenation (ODH) of ethane to ethylene as a probe reaction, we demonstrated that doping nonmetallic atoms to the surface lattice of catalyst of a transition-metal oxide can enhance catalytic selectivity through suppression of complete oxidation of the reactant molecules. Catalysts of Co3O4 with doped silicon atoms (Six-Co3O4) maintaining the spinel structure of pure Co3O4 exhibit much higher selectivity for the production of ethylene through ODH of ethane in comparison to pure Co3O4 at 600C by 40%. The suppression of activity of surface lattice oxygen atoms was evidenced by the observation that the surface lattice oxygen atoms of Six-Co3O4 cannot exchange oxygen atoms with gas-phase oxygen at low temperatures while pure Co3O4 can. The difference in releasing surface lattice oxygen atoms and dissociating molecular oxygen between pure Co3O4 and Six-Co3O4 was supported by DFT calculations. The calculated activation barriers for dissociation of molecular O2 and energy barriers for hopping surface oxygen vacancies of Six-Co3O4 are obviously higher than those of pure Co3O4, respectively. These experimental exploration and computational studies established a correlation between increase of catalytic selectivity and suppression of the activity of surface lattice oxygen atoms/oxygen vacancies. This correlation suggests an approach for increasing the catalytic selectivity of oxidative catalysis through suppressing activity of surface lattice oxygen atoms/vacancies via doping atoms of a nonmetallic element. This new approach was further confirmed by the observed higher catalytic selectivity for production of ethylene on Ge0.2-Co3O4 in comparison to pure Co3O4.
- Understanding oxidative dehydrogenation of ethane on Co3O4 nanorods from density functional theoryFung, V., Tao, F., and Jiang, D.Catalysis Science & Technology 6, 6861—6869, (2016)
Co3O4 is a metal oxide catalyst with weak, tunable M-O bonds promising for catalysis. Here, density functional theory (DFT) is used to study the oxidative dehydrogenation (ODH) of ethane on Co3O4 nanorods based on the preferred surface orientation (111) from the experimental electron-microscopy image. The pathway and energetics of the full catalytic cycle including the first and second C-H bond cleavages, hydroxyl clustering, water formation, and oxygen-site regeneration are determined. We find that both lattice O and Co may participate as active sites in the dehydrogenation, with the lattice-O pathway being favored. We identify the best ethane ODH pathway based on the overall energy profiles of several routes. We identify that water formation from the lattice oxygen has the highest energy barrier and is likely a rate-determining step. This work of the complete catalytic cycle of ethane ODH will allow further study into tuning the surface chemistry of Co3O4 nanorods for high selectivity of alkane ODH reactions.
- Towards Efficient Uncertainty Estimation in Deep Learning for Robust Energy Prediction in Materials ChemistrySirui, B., Fung, V., and Zhang, J.International Conference on Learning Representations Workshop on Deep Learning for Simulation (2021)
- Efficient Inverse Learning for Materials Design and DiscoveryZhang, J., and Fung, V.International Conference on Learning Representations Workshop on Science and Engineering of Deep Learning (2021)