Computational chemistry provides powerful research tools for catalysis. It potentially allows us to study the structures of the catalytic sites and reaction mechanisms, which are difficult to observe only by experiment. This is particularly true for supported heterogeneous catalysts, of which reactivity and catalytic behavior are directly related to the presence of various functional groups and reactive ensembles on their surfaces. Such surface heterogeneities give rise to the formation of multifunctional reactive ensembles ready to convert substrate molecules to the desired products efficiently. At the same time, the presence of various reactive centers on the surface may contribute to undesirable conversion paths. Understanding the role of the multifunctional reaction environments established on the complex surfaces of supported heterogeneous catalysts is key to formulating design rules for achieving control over their activity and selectivity....

Since the industrial revolution in the 1760s, the CO2 concentration in the atmosphere has been rising incessantly driving global warming closer to the point of no return. The world requires urgent actions to not only reduce CO2 emissions but also capture the CO2 for utilization to mitigate the future environmental crisis. CO2 hydrogenation to CH3OH offers an alternative to produce a feasible and economic substitute for oil. This technology also resembles the nearly 100 years old CH3OH synthesis processes from syngas containing H2, CO, and CO2. The conventional Cu/ZnO/Al2O3 catalyst has also been applied for more than 50 years, and its high performance stems from synergies between Cu and ZnO. However, the true nature of the interfacial sites is still extensively debated. Moreover, lower temperature and higher pressure are thermodynamically favorable for maximum CO2 conversion and CH3OH selectivity according to Le Châtelier’s principle and beneficial in terms of energy consumption and catalyst stability against sintering. The limitation in the catalytic performance of Cu/ZnO/Al2O3 in such conditions demands the exploration of novel catalysts.

Part I of this dissertation is dedicated to gaining a deeper understanding of Cu-ZnO synergistic structure as well as other Cu-based catalysts. In Chapter 2, we proposed a greener synthesis route for Cu/ZnO catalysts via urea hydrolysis of acetate precursors that can achieve comparable activity to commercial Cu/ZnO/Al2O3 catalysts without producing wastewater. Co-precipitated Cu-Zn hydroxycarbonate mineral-like precursors are crucial for a high inter-dispersion between CuO and ZnO after calcination and providing Cu-ZnO interfacial sites for the reaction. In Chapter 3, the effects of key process conditions, namely temperature and pressure, on CO2 hydrogenation over a commercial Cu/ZnO/Al2O3 catalyst were investigated using a space-resolved study. The gradients of reactants/products concentration and catalyst bed temperature within the catalytic reactor can reveal the significant effect of temperature on the dominant reaction pathways. CH3OH is formed through direct CO2 hydrogenation at low temperatures, while CH3OH formation is mediated via CO which is formed by a reverse water–gas shift reaction at a high temperature. Although pressure did not influence the reaction pathway, higher pressure helped suppress CH3OH decomposition to CO. In Chapter 4, the decisive roles of peripheral promoters to Cu nanoparticles in promoting CH3OH selectivity were elucidated. The model Cu-based catalysts (Cu-M/SiO2, M = Zn, Ga, and In) were prepared via surface organometallic chemistry (SOMC). The M+ sites played important roles in stabilizing formate species spillovered from Cu and determining the reactivity of formate hydrogenation. Improving the spillover and tuning the reactivity of formate help suppress formate decomposition to CO over Cu and ultimately boost CH3OH selectivity.

Part II is dedicated to exploring the novel catalysts for low-temperature CO¬2 hydrogenation, as well as, gaining a deeper understanding of the state-of-the-art Re/TiO2 catalyst. In Chapter 5, the bifunctionality of Re supported on TiO2 was deciphered, where metallic Re functions as the H2 activator and cationic Re as the CO2 activator. Re/TiO2 suffers from additional CH4 formation, and the active intermediates and reaction pathways for CH3OH and CH4 were identified. Understanding the nature of active sites and reaction mechanisms over Re/TiO2 led to approaches for CH4 selectivity mitigation in Chapter 6. Exploring various transition metals under low-temperature conditions provided insights into the formate stabilization of the coinage metals (Cu, Ag, and Au). Since the balance between metallic and cationic Re limited the CH3OH selectivity of Re/TiO2, the addition of Ag complemented the role of cationic Re. A synergistic interplay between Ag and Re did not only improve CH3OH selectivity significantly by suppressing intermediates in the reaction pathways toward CH4 but also exhibited superior stability.

Finally, the dissertation conveys a message that obtaining the definitive synthesis of well-defined active sites, expansive structure-activity relationships, and comprehensive reaction mechanisms are the major prerequisites for the rational design of novel catalysts.

Many drugs cannot be made without homogeneous catalysis. To increase the yield of drug synthesis, the search for new catalysts continues. Computational catalysis is becoming a more prominent tool since it en- ables to screen many catalysts without performing (m)any laboratory experiments. In this research a com- putational workflow has been created to calculate both electronic (e.g. dipole, ionisation potential, nucle- ophilicity, etc.) and steric (bite angle, buried volume, cone angle, etc.) molecular descriptors. The structures were automatically created starting from the metal centre, some bidentate phosphorus ligands, auxiliary lig- ands, and the substrate. An in-house computational workflow, MACE, is used for for high-throughput gen- eration of structures from the starting bidentate phosphorus ligands, by generating stereo-isomers around to metal centre. Afterwards small substituents (H, CH3, Ph, etc.) are changed by ChemSpaX increasing the number of structures combinatorically. To reduce the computational cost of this workflow, it has been re- searched whether properties of the octahedral geometry could be predicted using properties from a simple model structure containing only the metal centre and the bidentate phosphorus ligand. This model structure did not show a correlation except for the electronic energy and the solvent accessible surface area, which are both primarily influenced by the number of electrons. Other correlations may be found if some other descriptors were calculated which where excluded now, like the HOMO-LUMO gap and the substrate bind- ing energy. The workflow could be extended to machine learning and improved by including symmetry and optical isomerism.

In this thesis, we have presented and investigated the possible strategies for modelling complex heterogeneous catalytic systems in operando regimes. By introducing modern computational approaches to sample the potential energy surface of the catalytic active site, we have attempted to account for the reactive conditions, solvent presence, additives inclusion, and structural dynamics of the active site. There is growing spectroscopic and theoretical evidence of the critical role of the active site dynamics for the catalytic performance, advocating for the active site representation as an ensemble of possible isomers. Challenged by the complexity of the reactive environment and common heterogeneous catalysts, we strongly believe that addressing these factors and incorporating them explicitly into the model description will contribute to a more realistic representation of the catalytic system.

The best available catalysts for the reduction of carbonyl groups in ketones are
currently based on rare metals such as platinum, osmium or ruthenium. The developmentof alternative catalysts made from more abundant metal such as manganese offers chances to increase sustainability and cost efficiency. Analysing and predicting chemical and physical properties with a computational would streamline the development process by allowing early elimination of options and by proposing new structures with a high chance of success. Density-functional theory (DFT) can predict structures and ligand properties at a high level. In this investigation, around 40 ligands were defined based on previously researched manganese compounds with known catalytic properties or previously investigated ruthenium complexes. Properties such as hydricity, CO frequencies and bond lengths were calculated and linked to known catalytic activity from the literature.No hard correlations between the investigated properties and reported catalytic activities have been found. Some links however do appear to exist such as a connection between high calculated hydricities and a higher reported catalytic activity.

Catalysts play an essential role in the daily lives of humans. These catalysts are used in many industries to make processes more energetically favourable. Climate change is pushing humanity towards the usage of more green energy and catalysts play an important role in this transition.For example, in the hydrogenation reaction used for the storage of H2, where the catalyst is involved in the storage and removal of H2 on a storage medium like CO2. The properties of the catalyst involved in this (de)hydrogenation reaction can affect the selectivity and yield of the reaction. Designing a catalyst that maximizes the property (yield for example) that we are interested in for a specific reaction, is an essential asset to tune catalyzed processes. Computational screening of many catalysts has attracted the attention of academia and industry due to constant developments in the field of computational chemistry.In these computational methods, predictive models together with DFT and/or DFTB methods can be used to correlate a set of reaction descriptors with catalyst properties. The model has a higher probability to find novel molecules with a high activity when more (reliable) training data is used and when the search spac eof the model is confined to a local chemical space. This means that newly added molecules for screening should be structurally closely related to the molecule that was used to build the model. Unfortunately, large data sets are not readily available for transition-metal containing complexes although these complexes are widely applied in the field of homogeneous catalysis. In this research a Python-based workflow, ChemSpaX, that is aimed at automating local chemical space exploration for any type of molecule is introduced. Thisworkflow enables the user to place fragments on molecules based on 3Dinformation, while staying close to the quality of the initial structure. This enables data-driven property calculations and prediction models, which could eventually be extended towards the automated design of new catalysts. Various representative applications of ChemSpaX are presented in which data-driven xTBand DFT property calculations are done. The found correlations between catalyst properties are shown and it is shown that ChemSpaX generates structures that have a reasonable quality for usage in data-driven prediction models for high-throughput screening.  

Recently machine learning (ML) has become increasingly popular, and has been shown to be a powerful predictive technique. The applications of ML cover a wide range of disciplines, including the natural sciences. Presently, the field of catalysis is still relatively unexposed to ML and other data-driven techniques. This can largely be attributed to the broad variety and complexity of catalytic data, which obstructs data unification into large structured databases. This is problematic because ML requires large amounts of information rich data to ensure its effectiveness. Additionally, applying ML techniques requires expertise and often coding experience. These requisites impede the adoption of ML by catalysis researchers, that are not necessarily programming experts. In this thesis a data management and analytics platform is developed to reduce this inaccessibility barrier. Our platform is designed to guide catalysis researchers through the ML workflow, and construct effective ML models. Throughout this process the platform supports several key functionalities, which include interaction with a database instance, data visualization, ML model construction, and ML model application. Furthermore, the usefulness of our platform is demonstrated in a case study, where ML models are built to predict the catalytic performance based on molecular descriptors of the catalyst. Due to a lack of suitable existing catalytic datasets, we construct and use an artificial dataset that mimics kinetic catalytic data. Artificially constructing the data allows us full control of its underlying mechanisms. This aspect is used to build dataset variations where we study the effect of database size and descriptor strength on the performance of ML models. The results of the case study quantify the effects of database size and descriptor strength, which are useful in identifying the objectives for future construction of databases from real catalytic data.

Catalysis is a critical enabling technology that directly contributes to higher standards of living. This is a remarkable feat for a technology that most consumers have no direct contact with and perhaps only vaguely know from their car’s catalytic converter. Even many technical users regard catalysts as ideal substances that promote a target transformation without being consumed in the reaction. Reality, however, is much more complex because catalysts can also produce undesirable side-products or stop working before the target reaction is complete. This dissertation explores such complexity.
The aim of this work was to study real catalytic systems under relevant reaction conditions. This challenge was approached from two different directions, and the dissertation is therefore divided into two parts. Part I describes our study of non-noble Mn complexes as catalysts for a variety of reduction chemistries. Particular attention was given to the study of deactivation phenomena to understand why some catalysts stop operating early in their lifetime. This work led to catalytic methods that enable improved operation at significantly lower catalyst loading. The second part of this dissertation focuses on the development and application of automation and data-rich experimentation methods for catalysis R&D. An automated reaction analysis platform was developed that simultaneously produces kinetic data and several representative data streams of operando spectroscopy. These tools were used to study and optimise the performance of a variety of chemistries, and are expected to provide the required dense data for contemporary data-driven methods.

Ceramic nanofiltration (NF) is a newly developed technology for municipal sewage reclamation for drinking water production, agricultural and industrial utilisation, but organic fouling limits it to pilot-scale application. Herein, Fenton-based cleaning catalysed by FeOCl pre-coat was performed on ceramic NF membrane to clean persistent gel-like fouling. FeOCl pre-coating on ceramic NF membrane was conducted by pressure-driven filtration. Gel-like fouling was simulated by Ca-alginate. The compact fouling layer was proposed to impede the diffusion of H2O2; therefore, NaCl was introduced to perform Na-Ca ion exchange to relax the fouling layer and improve the mass transfer in Fenton-based cleaning. The cleaning efficiency of Fenton cleaning and NaCl treatment were evaluated individually, and their synergistic effect was determined by a mixture of NaCl/H2O2. In pre-coating, the iron loading could be controlled by permeate flux. Moreover, the iron loading on the membrane had a linear relationship with FeOCl dosage. The pre-coated FeOCl layer slightly decreased membrane permeability (by 4%), presumably because the porous FeOCl pre-coat provided abundant channels for water permeation. NaCl pre-treatment displayed high efficiency in the removal of the reversible fouling layer, which contributed to 50% of flux decline. The FeOCl pre-coat catalysed Fenton-based cleaning, which achieved a complete flux recovery in 1 hour. The addition of tert-butanol (an ∙OH scavenger) in Fenton-based cleaning system decreased the cleaning efficiency, which confirmed the participation of ∙OH in Fenton-based cleaning. No remarkable synergistic effect was observed between NaCl induced fouling layer swelling and Fenton oxidation. It was likely due to the effect of turbulent flush on improving the diffusion of H2O2 in Fenton-based cleaning, which surpassed the effect of fouling layer swelling induced by NaCl treatment.

In this thesis we presented comprehensive studies on methane activation by binuclear Fe-oxo sites located in porous MOF and zeolite frameworks. Many of such studies have been previously performed, however the main focus was usually on the C-H bond activation of methane. Here we focused on the whole reaction mechanism including the activation of the Fe site and the overoxidation of methanol, as well as the effects of the porous framework.