ZeoSyn: A Comprehensive Zeolite Synthesis Dataset Enabling Machine-Learning Rationalization of Hydrothermal Parameters.

Zeolites, nanoporous aluminosilicates with well-defined porous structures, are versatile materials with applications in catalysis, gas separation, and ion exchange. Hydrothermal synthesis is widely used for zeolite production, offering control over composition, crystallinity, and pore size. However, the intricate interplay of synthesis parameters necessitates a comprehensive understanding of synthesis-structure relationships to optimize the synthesis process. Hitherto, public zeolite synthesis databases only contain a subset of parameters and are small in scale, comprising up to a few thousand synthesis routes. We present ZeoSyn, a dataset of 23,961 zeolite hydrothermal synthesis routes, encompassing 233 zeolite topologies and 921 organic structure-directing agents (OSDAs). Each synthesis route comprises comprehensive synthesis parameters: 1) gel composition, 2) reaction conditions, 3) OSDAs, and 4) zeolite products. Using ZeoSyn, we develop a machine learning classifier to predict the resultant zeolite given a synthesis route with >70% accuracy. We employ SHapley Additive exPlanations (SHAP) to uncover key synthesis parameters for >200 zeolite frameworks. We introduce an aggregation approach to extend SHAP to all building units. We demonstrate applications of this approach to phase-selective and intergrowth synthesis. This comprehensive analysis illuminates the synthesis parameters pivotal in driving zeolite crystallization, offering the potential to guide the synthesis of desired zeolites. The dataset is available at https://github.com/eltonpan/zeosyn_dataset.

One-Pot Synthesis of CHA/ERI-Type Zeolite Intergrowth from a Single Multiselective Organic Structure-Directing Agent.

We report the one-pot synthesis of a chabazite (CHA)/erionite (ERI)-type zeolite intergrowth structure characterized by adjustable extents of intergrowth enrichment and Si/Al molar ratios. This method utilizes readily synthesizable 6-azaspiro[5.6]dodecan-6-ium as the exclusive organic structure-directing agent (OSDA) within a potassium-dominant environment. High-throughput simulations were used to accurately determine the templating energy and molecular shape, facilitating the selection of an optimally biselective OSDA from among thousands of prospective candidates. The coexistence of the crystal phases, forming a distinct structure comprising disk-like CHA regions bridged by ERI-rich pillars, was corroborated via rigorous powder X-ray diffraction and integrated differential-phase contrast scanning transmission electron microscopy (iDPC S/TEM) analyses. iDPC S/TEM imaging further revealed the presence of single offretite layers dispersed within the ERI phase. The ratio of crystal phases between CHA and ERI in this type of intergrowth could be varied systematically by changing both the OSDA/Si and K/Si ratios. Two intergrown zeolite samples with different Si/Al molar ratios were tested for the selective catalytic reduction (SCR) of NOx with NH3, showing competitive catalytic performance and hydrothermal stability compared to that of the industry-standard commercial NH3-SCR catalyst, Cu-SSZ-13, prevalent in automotive applications. Collectively, this work underscores the potential of our approach for the synthesis and optimization of adjustable intergrown zeolite structures, offering competitive alternatives for key industrial processes.

Effect of Framework Composition and NH3 on the Diffusion of Cu+ in Cu-CHA Catalysts Predicted by Machine-Learning Accelerated Molecular Dynamics.

Cu-exchanged zeolites rely on mobile solvated Cu+ cations for their catalytic activity, but the role of the framework composition in transport is not fully understood. Ab initio molecular dynamics simulations can provide quantitative atomistic insight but are too computationally expensive to explore large length and time scales or diverse compositions. We report a machine-learning interatomic potential that accurately reproduces ab initio results and effectively generalizes to allow multinanosecond simulations of large supercells and diverse chemical compositions. Biased and unbiased simulations of [Cu(NH3)2]+ mobility show that aluminum pairing in eight-membered rings accelerates local hopping and demonstrate that increased NH3 concentration enhances long-range diffusion. The probability of finding two [Cu(NH3)2]+ complexes in the same cage, which is key for SCR-NOx reaction, increases with Cu content and Al content but does not correlate with the long-range mobility of Cu+. Supporting experimental evidence was obtained from reactivity tests of Cu-CHA catalysts with a controlled chemical composition.

Synergic Effect of Isolated Ce3+ and Ptδ+ Species in UiO-66(Ce) for Heterogeneous Catalysis.

In this work, we have synthesized through an efficient electrostatic deposition a Pt single-atom catalyst (SAC) supported on a Ce-MOF. The basic solution employed in the impregnation process favors the deprotonation of the hydroxyl groups allocated on the clusters that can easily interact with the cationic Pt species. The resulting material, denoted as Pt/UiO-66(Ce), shows an increment of Ce3+ content, as demonstrated by UV-vis and Ce L3-edge XANES spectroscopy. These Ce3+ species and their corresponding oxygen vacancies are able to accommodate very disperse Pt single sites. Moreover, Pt L3-edge XANES and CO-FTIR spectroscopy confirm the cationic nature of the supported Ptδ+ (2+ < δ < 4+). For comparison purpose, we have synthesized and characterized a well-known Pt single-site catalyst supported on nanocrystalline ceria, denoted as Pt/nCeO2. Since the simultaneous presence of Ce3+ and Ptδ+ on the MOF clusters were able to activate the oxygen molecules and the CO molecule, respectively, we tested Pt/UiO-66(Ce) for the CO oxidation reaction. Interestingly, this catalyst showed ∼six-fold increment in activity in comparison with the traditional Pt/nCeO2 material. Finally, the characterization after catalysis reveals that the Pt nature is preserved and that the activity is maintained during 14 h at 100 °C without any evidence of deactivation.

Approaching enzymatic catalysis with zeolites or how to select one reaction mechanism competing with others.

Approaching the level of molecular recognition of enzymes with solid catalysts is a challenging goal, achieved in this work for the competing transalkylation and disproportionation of diethylbenzene catalyzed by acid zeolites. The key diaryl intermediates for the two competing reactions only differ in the number of ethyl substituents in the aromatic rings, and therefore finding a selective zeolite able to recognize this subtle difference requires an accurate balance of the stabilization of reaction intermediates and transition states inside the zeolite microporous voids. In this work we present a computational methodology that, by combining a fast high-throughput screeening of all zeolite structures able to stabilize the key intermediates with a more computationally demanding mechanistic study only on the most promising candidates, guides the selection of the zeolite structures to be synthesized. The methodology presented is validated experimentally and allows to go beyond the conventional criteria of zeolite shape-selectivity.

Tunable CHA/AEI Zeolite Intergrowths with A Priori Biselective Organic Structure-Directing Agents: Controlling Enrichment and Implications for Selective Catalytic Reduction of NOx.

A novel ab initio methodology based on high-throughput simulations has permitted designing unique biselective organic structure-directing agents (OSDAs) that allow the efficient synthesis of CHA/AEI zeolite intergrowth materials with controlled phase compositions. Distinctive local crystallographic ordering of the CHA/AEI intergrowths was revealed at the nanoscale level using integrated differential phase contrast scanning transmission electron microscopy (iDPC STEM). These novel CHA/AEI materials have been tested for the selective catalytic reduction (SCR) of NOx, presenting an outstanding catalytic performance and hydrothermal stability, even surpassing the performance of the well-established commercial CHA-type catalyst. This methodology opens the possibility for synthetizing new zeolite intergrowths with more complex structures and unique catalytic properties.

Data-Driven Design of Biselective Templates for Intergrowth Zeolites.

Zeolites are inorganic materials with wide industrial applications due to their topological diversity. Tailoring confinement effects in zeolite pores, for instance by crystallizing intergrown frameworks, can improve their catalytic and transport properties, but controlling zeolite crystallization often relies on heuristics. In this work, we use computational simulations and data mining to design organic structure-directing agents (OSDAs) to favor the synthesis of intergrown zeolites. First, we propose design principles to identify OSDAs which are selective toward both end members of the disordered structure. Then, we mine a database of hundreds of thousands of zeolite-OSDA pairs and downselect OSDA candidates to synthesize known intergrowth zeolites such as CHA/AFX, MTT/TON, and BEC/ISV. The computationally designed OSDAs balance phase competition metrics and shape selectivity toward the frameworks, thus bypassing expensive dual-OSDA approaches typically used in the synthesis of intergrowths. Finally, we propose potential OSDAs to obtain hypothesized disordered frameworks such as AEI/SAV. This work may accelerate zeolite discovery through data-driven synthesis optimization and design.

A priori control of zeolite phase competition and intergrowth with high-throughput simulations.

Zeolites are versatile catalysts and molecular sieves with large topological diversity, but managing phase competition in zeolite synthesis is an empirical, labor-intensive task. In this work, we controlled phase selectivity in templated zeolite synthesis from first principles by combining high-throughput atomistic simulations, literature mining, human-computer interaction, synthesis, and characterization. Proposed binding metrics distilled from more than 586,000 zeolite-molecule simulations reproduced the extracted literature and rationalized framework competition in the design of organic structure-directing agents. Energetic, geometric, and electrostatic descriptors of template molecules were found to regulate synthetic accessibility windows and aluminum distributions in pure-phase zeolites. Furthermore, these parameters allowed us to realize an intergrowth zeolite through a single bi-selective template. The computation-first approach enables control of both zeolite synthesis and structure composition using a priori theoretical descriptors.

Tailoring Lewis/Brønsted acid properties of MOF nodes via hydrothermal and solvothermal synthesis: simple approach with exceptional catalytic implications.

The Lewis/Brønsted catalytic properties of the Metal-Organic Framework (MOF) nodes can be tuned by simply controlling the solvent employed in the synthetic procedure. In this work, we demonstrate that Hf-MOF-808 can be prepared from a material with a higher amount of Brønsted acid sites, via modulated hydrothermal synthesis, to a material with a higher proportion of unsaturated Hf Lewis acid sites, via modulated solvothermal synthesis. The Lewis/Brønsted acid properties of the resultant metallic clusters have been studied by different characterization techniques, including XAS, FTIR and NMR spectroscopies, combined with a DFT study. The different nature of the Hf-MOF-808 materials allows their application as selective catalysts in different target reactions requiring Lewis, Brønsted or Lewis-Brønsted acid pairs.

Design and Synthesis of the Active Site Environment in Zeolite Catalysts for Selectively Manipulating Mechanistic Pathways.

By combining kinetics and theoretical calculations, we show here the benefits of going beyond the concept of static localized and defined active sites on solid catalysts, into a system that globally and dynamically considers the active site located in an environment that involves a scaffold structure particularly suited for a target reaction. We demonstrate that such a system is able to direct the reaction through a preferred mechanism when two of them are competing. This is illustrated here for an industrially relevant reaction, the diethylbenzene-benzene transalkylation. The zeolite catalyst (ITQ-27) optimizes location, density, and environment of acid sites to drive the reaction through the preselected and preferred diaryl-mediated mechanism, instead of the alkyl transfer pathway. This is achieved by minimizing the activation energy of the selected pathway through weak interactions, much in the way that it occurs in enzymatic catalysts. We show that ITQ-27 outperforms previously reported zeolites for the DEB-Bz transalkylation and, more specifically, industrially relevant zeolites such as faujasite, beta, and mordenite.

Discovering Relationships between OSDAs and Zeolites through Data Mining and Generative Neural Networks.

Organic structure directing agents (OSDAs) play a crucial role in the synthesis of micro- and mesoporous materials especially in the case of zeolites. Despite the wide use of OSDAs, their interaction with zeolite frameworks is poorly understood, with researchers relying on synthesis heuristics or computationally expensive techniques to predict whether an organic molecule can act as an OSDA for a certain zeolite. In this paper, we undertake a data-driven approach to unearth generalized OSDA-zeolite relationships using a comprehensive database comprising of 5,663 synthesis routes for porous materials. To generate this comprehensive database, we use natural language processing and text mining techniques to extract OSDAs, zeolite phases, and gel chemistry from the scientific literature published between 1966 and 2020. Through structural featurization of the OSDAs using weighted holistic invariant molecular (WHIM) descriptors, we relate OSDAs described in the literature to different types of cage-based, small-pore zeolites. Lastly, we adapt a generative neural network capable of suggesting new molecules as potential OSDAs for a given zeolite structure and gel chemistry. We apply this model to CHA and SFW zeolites generating several alternative OSDA candidates to those currently used in practice. These molecules are further vetted with molecular mechanics simulations to show the model generates physically meaningful predictions. Our model can automatically explore the OSDA space, reducing the amount of simulation or experimentation needed to find new OSDA candidates.

Metalloenzyme-Inspired Ce-MOF Catalyst for Oxidative Halogenation Reactions.

The structure of UiO-66(Ce) is formed by CeO2-x defective nanoclusters connected by terephthalate ligands. The initial presence of accessible Ce3+ sites in the as-synthesized UiO-66(Ce) has been determined by X-ray photoelectron spectroscopy (XPS) and Fourier transform infrared (FTIR)-CO analyses. Moreover, linear scan voltammetric measurements reveal a reversible Ce4+/Ce3+ interconversion within the UiO-66(Ce) material, while nanocrystalline ceria shows an irreversible voltammetric response. This suggests that terephthalic acid ligands facilitate charge transfer between subnanometric metallic nodes, explaining the higher oxidase-like activity of UiO-66(Ce) compared to nanoceria for the mild oxidation of organic dyes under aerobic conditions. Based on these results, we propose the use of Ce-based metal-organic frameworks (MOFs) as efficient catalysts for the halogenation of activated arenes, as 1,3,5-trimethoxybenzene (TMB), using oxygen as a green oxidant. Kinetic studies demonstrate that UiO-66(Ce) is at least three times more active than nanoceria under the same reaction conditions. In addition, the UiO-66(Ce) catalyst shows an excellent stability and can be reused after proper washing treatments. Finally, a general mechanism for the oxidative halogenation reaction is proposed when using Ce-MOF as a catalyst, which mimics the mechanistic pathway described for metalloenzymes. The superb control in the generation of subnanometric CeO2-x defective clusters connected by adequate organic ligands in MOFs offers exciting opportunities in the design of Ce-based redox catalysts.

Single-Site vs. Cluster Catalysis in High Temperature Oxidations.

The behavior of single Pt atoms and small Pt clusters was investigated for high-temperature oxidations. The high stability of these molecular sites in CHA is a key to intrinsic structure-performance descriptions of elemental steps such as O2 dissociation, and subsequent oxidation catalysis. Subtle changes in the atomic structure of Pt are responsible for drastic changes in performance driven by specific gas/metal/support interactions. Whereas single Pt atoms and Pt clusters (> ca. 1 nm) are unable to activate, scramble, and desorb two O2 molecules at moderate T (200 °C), clusters <1 nm do so catalytically, but undergo oxidative fragmentation. Oxidation of alkanes at high T is attributed to stable single Pt atoms, and the C-H cleavage is inferred to be rate-determining and less sensitive to changes in metal nuclearity compared to its effect on O2 scrambling. In contrast, when combustion involves CO, catalysis is dominated by metal clusters, not single Pt atoms.

Coordinatively Unsaturated Hf-MOF-808 Prepared via Hydrothermal Synthesis as a Bifunctional Catalyst for the Tandem N-Alkylation of Amines with Benzyl Alcohol.

The modulated hydrothermal (MHT) synthesis of an active and selective Hf-MOF-808 material for the N-alkylation reaction of aniline with benzyl alcohol under base-free mild reaction conditions is reported. Through kinetic experiments and isotopically labeled NMR spectroscopy studies, we have demonstrated that the reaction mechanism occurs via borrowing hydrogen (BH) pathway, in which the alcohol dehydrogenation is the limiting step. The high concentration of defective -OH groups generated on the metallic nodes through MHT synthesis enhances the alcohol activation, while the unsaturated Hf4+, which acts as a Lewis acid site, is able to borrow the hydrogen from the methylene position of benzyl alcohol. This fact makes this material at least 14 times more active for the N-alkylation reaction than the material obtained via solvothermal synthesis. The methodology described in this work could be applied to a wide range of aniline and benzyl alcohol derivates, showing in all cases high selectivity toward the corresponding N-benzylaniline product. Finally, Hf-MOF-808, which acts as a true heterogeneous catalyst, can be reused in at least four consecutive runs without any activity loss.

Unusually Low Heat of Adsorption of CO2 on AlPO and SAPO Molecular Sieves.

The capture of CO2 from post-combustion streams or from other mixtures, such as natural gas, is an effective way of reducing CO2 emissions, which contribute to the greenhouse effect in the atmosphere. One of the developing technologies for this purpose is physisorption on selective solid adsorbents. The ideal adsorbents are selective toward CO2, have a large adsorption capacity at atmospheric pressure and are easily regenerated, resulting in high working capacity. Therefore, adsorbents combining molecular sieving properties and low heats of adsorption of CO2 are of clear interest as they will provide high selectivities and regenerabilities in CO2 separation process. Here we report that some aluminophosphate (AlPO) and silicoaluminophosphate (SAPO) materials with LTA, CHA and AFI structures present lower heats of adsorption of CO2 (13-25 kJ/mol) than their structurally analogous zeolites at comparable framework charges. In some cases, their heats of adsorption are even lower than those of pure silica composition (20-25 kJ/mol). This could mean a great improvement in the regeneration process compared to the most frequently used zeolitic adsorbents for this application while maintaining most of their adsorption capacity, if materials with the right stability and pore size and topology are found.

Impact of Zeolite Framework Composition and Flexibility on Methanol-To-Olefins Selectivity: Confinement or Diffusion?

The methanol-to-olefins reaction catalyzed by small-pore cage-based acid zeolites and zeotypes produces a mixture of short chain olefins, whose selectivity to ethene, propene and butene varies with the cavity architecture and with the framework composition. The product distribution of aluminosilicates and silicoaluminophosphates with the CHA and AEI structures (H-SSZ-13, H-SAPO-34, H-SSZ-39 and H-SAPO-18) has been experimentally determined, and the impact of acidity and framework flexibility on the stability of the key cationic intermediates involved in the mechanism and on the diffusion of the olefin products through the 8r windows of the catalysts has been evaluated by means of periodic DFT calculations and ab initio molecular dynamics simulations. The preferential stabilization by confinement of fully methylated hydrocarbon pool intermediates favoring the paring pathway is the main factor controlling the final olefin product distribution.

Selective active site placement in Lewis acid zeolites and implications for catalysis of oxygenated compounds.

The selective incorporation of isolated framework Lewis acid sites at specific crystallographic positions in high-silica zeolites was achieved by applying a rationalized post-synthetic grafting methodology. The removal of framework Ge atoms from a Ge-BEC zeolite with low concentrations of Ge in the framework (Si/Ge ∼ 150) followed by grafting allows the synthesis of Sn-BEC zeolites with Sn atoms positionally biased into the double-4-ring (D4R) crystallographic positions of the BEC framework. Spectroscopic characterization using solid-state nuclear magnetic resonance (NMR) coupled with theoretical calculations revealed that Sn atoms preferentially form open Sn sites in the D4R of Sn-BEC. This observation was supported by IR spectra of adsorbed deuterated acetonitrile (CD3CN), a known titrant of Sn sites in zeolites. The catalytic implications of selective incorporation of open Sn sites in Sn-BEC were probed using the Meerwein-Ponndorf-Verley-Oppenauer (MPVO) reaction. Although the MPVO turnover rates normalized by the total number of open Sn sites were comparable on Sn-BEC and a conventional Sn-Beta catalyst synthesized in fluoride media (Sn-Beta(F)), Sn-BEC demonstrated higher per gram reaction rates because of its larger fraction of open sites compared to Sn-Beta(F). These results highlight the advantage of placing active sites in targeted locations within a zeolite structure. The methodology presented here to selectively place catalytic active sites via sacrificial heteroatoms, such as Ge, can be generalized for the design of many other tetrahedrally-coordinated metal-containing zeolites.

Conceptual similarities between zeolites and artificial enzymes.

By using a Diels-Alder (DA) reaction as a base case, we show that a pure silica zeolite acting as an entropy-trapping scaffold can be synthesized with an organic structure directing agent (OSDA) analogue of the transition state (TS) of the DA reaction. A cavity stabilization of the TS is observed with the corresponding decrease in the activation energy of the reaction. A lower enthalpy of activation and a larger decrease in entropy are obtained with the zeolite synthesized with the analogue of the DA TS when compared with other zeolitic structures. Those differences are maintained, while catalytic activity is increased, when active sites are introduced in the zeolite. The catalytic zeolitic system synthesized with the OSDA analogue of the TS shows conceptual similarities with "de novo design" of an artificial enzyme to perform DA reactions, in where a suitable scaffold of existing proteins is chosen, and computationally designed active sites able to catalyze the cycloaddition reaction are introduced.

Machine Learning Applied to Zeolite Synthesis: The Missing Link for Realizing High-Throughput Discovery.

Zeolites are microporous crystalline materials with well-defined cavities and pores, which can be prepared under different pore topologies and chemical compositions. Their preparation is typically defined by multiple interconnected variables (e.g., reagent sources, molar ratios, aging treatments, reaction time and temperature, among others), but unfortunately their distinctive influence, particularly on the nucleation and crystallization processes, is still far from being understood. Thus, the discovery and/or optimization of specific zeolites is closely related to the exploration of the parametric space through trial-and-error methods, generally by studying the influence of each parameter individually. In the past decade, machine learning (ML) methods have rapidly evolved to address complex problems involving highly nonlinear or massively combinatorial processes that conventional approaches cannot solve. Considering the vast and interconnected multiparametric space in zeolite synthesis, coupled with our poor understanding of the mechanisms involved in their nucleation and crystallization, the use of ML is especially timely for improving zeolite synthesis. Indeed, the complex space of zeolite synthesis requires drawing inferences from incomplete and imperfect information, for which ML methods are very well-suited to replace the intuition-based approaches traditionally used to guide experimentation. In this Account, we contend that both existing and new ML approaches can provide the "missing link" needed to complete the traditional zeolite synthesis workflow used in our quest to rationalize zeolite synthesis. Within this context, we have made important efforts on developing ML tools in different critical areas, such as (1) data-mining tools to process the large amount of data generated using high-throughput platforms; (2) novel complex algorithms to predict the formation of energetically stable hypothetical zeolites and guide the synthesis of new zeolite structures; (3) new "ab initio" organic structure directing agent predictions to direct the synthesis of hypothetical or known zeolites; (4) an automated tool for nonsupervised data extraction and classification from published research articles. ML has already revolutionized many areas in materials science by enhancing our ability to map intricate behavior to process variables, especially in the absence of well-understood mechanisms. Undoubtedly, ML is a burgeoning field with many future opportunities for further breakthroughs to advance the design of molecular sieves. For this reason, this Account includes an outlook of future research directions based on current challenges and opportunities. We envision this Account will become a hallmark reference for both well-established and new researchers in the field of zeolite synthesis.

A Machine Learning Approach to Zeolite Synthesis Enabled by Automatic Literature Data Extraction.

Zeolites are porous, aluminosilicate materials with many industrial and "green" applications. Despite their industrial relevance, many aspects of zeolite synthesis remain poorly understood requiring costly trial and error synthesis. In this paper, we create natural language processing techniques and text markup parsing tools to automatically extract synthesis information and trends from zeolite journal articles. We further engineer a data set of germanium-containing zeolites to test the accuracy of the extracted data and to discover potential opportunities for zeolites containing germanium. We also create a regression model for a zeolite's framework density from the synthesis conditions. This model has a cross-validated root mean squared error of 0.98 T/1000 Å3, and many of the model decision boundaries correspond to known synthesis heuristics in germanium-containing zeolites. We propose that this automatic data extraction can be applied to many different problems in zeolite synthesis and enable novel zeolite morphologies.