Skip to main content

Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

  • Article
  • Published:

A comprehensive mapping of zeolite–template chemical space

Nature Computational Science volume 5, pages 661–674 (2025)Cite this article

Subjects

A preprint version of the article is available at ChemRxiv.

Abstract

Zeolites are industrially important catalysts and adsorbents, typically synthesized using specific molecules known as organic structure-directing agents (OSDAs). The templating effect of the OSDAs is pivotal in determining the zeolite polymorph formed and its physicochemical properties. However, de novo design of selective OSDAs is challenging because of the diversity and size of the zeolite–OSDA chemical space. Here we present ZeoBind, a computational workflow powered by machine learning that enables an exhaustive exploration of the OSDA space. We design predictive tasks that capture zeolite–molecule matching, train predictive models for these tasks on hundreds of thousands of datapoints and curate a library of 2.3 million synthetically accessible, hypothetical OSDA-like molecules enumerated from commercially available precursors. We use ZeoBind to screen nearly 500 million zeolite–molecule pairs and identified and experimentally validated two new OSDAs that template zeolites with novel compositions. The scale of the OSDA library, along with the open-access tools and data, has the potential to accelerate OSDA design for zeolite synthesis.

Access through your institution
Buy or subscribe

This is a preview of subscription content, access via your institution

Access options

Access through your institution

Buy this article

  • Purchase on SpringerLink
  • Instant access to full article PDF
Buy now

Prices may be subject to local taxes which are calculated during checkout

Fig. 1: Schematic of ZeoBind.
Fig. 2: Analysis of hypothetical OSDA library.
Fig. 3: ML model performance.
Fig. 4: OSDA screening for ERI.
Fig. 5: OSDA screening for CHA.

Similar content being viewed by others

Data availability

All training and validation data, experimental data, hypothetical molecule data and predictions made over the entire known zeolite–hypothetical molecule space are available at Materials Data Facility61. Source data are provided with this paper.

Code availability

The code for developing the ML models, analyzing predictions and screening is available via GitHub at https://github.com/learningmatter-mit/zeobind and via Zenodo at https://doi.org/10.5281/zenodo.15425842 (ref. 62). The Lipari color scheme from Crameri et al.63 was used for the figures.

References

  1. Pérez-Botella, E., Valencia, S. & Rey, F. Zeolites in adsorption processes: state of the art and future prospects. Chem. Rev. 122, 17647–17695 (2022).

    Article  Google Scholar 

  2. Li, Y., Li, L. & Yu, J. Applications of zeolites in sustainable chemistry. Chem 3, 928–949 (2017).

    Article  Google Scholar 

  3. Lobo, R. F., Zones, S. I. & Davis, M. E. in Inclusion Chemistry with Zeolites: Nanoscale Materials by Design (eds Herron, N. & Corbin D. R.) 47–78 (Kluwer Academic, 1995).

  4. Moliner, M., Rey, F. & Corma, A. Towards the rational design of efficient organic structure-directing agents for zeolite synthesis. Angew. Chem. Int. Ed. 52, 13880–13889 (2013).

    Article  Google Scholar 

  5. Li, J., Gao, M., Yan, W. & Yu, J. Regulation of the Si/Al ratios and Al distributions of zeolites and their impact on properties. Chem. Sci. 14, 1935–1959 (2023).

    Article  Google Scholar 

  6. Le, T. T., Chawla, A. & Rimer, J. D. Impact of acid site speciation and spatial gradients on zeolite catalysis. J. Catal. 391, 56–68 (2020).

    Article  Google Scholar 

  7. Hoffman, A. J. et al. Rigid arrangements of ionic charge in zeolite frameworks conferred by specific aluminum distributions preferentially stabilize alkanol dehydration transition states. Angew. Chem. Int. Ed. 59, 18686–18694 (2020).

    Article  Google Scholar 

  8. Pan, E. et al. ZeoSyn: a comprehensive zeolite synthesis dataset enabling machine-learning rationalization of hydrothermal parameters. ACS Cent. Sci. 10, 729–743 (2024).

    Article  Google Scholar 

  9. Schwalbe-Koda, D. et al. A priori control of zeolite phase competition and intergrowth with high-throughput simulations. Science 374, 308–315 (2021).

    Article  Google Scholar 

  10. Simancas, R. et al. Recent progress in the improvement of hydrothermal stability of zeolites. Chem. Sci. 12, 7677–7695 (2021).

    Article  Google Scholar 

  11. Sastre, G. & Daeyaert, F. (eds) AI-Guided Design and Property Prediction for Zeolites and Nanoporous Materials (Wiley, 2023).

  12. Ito, S., Muraoka, K. & Nakayama, A. De novo design of organic structure-directing agents for zeolites using a general-purpose large language model. Chem. Mater. 37, 2447–2456 (2025).

    Article  Google Scholar 

  13. Daeyaert, F., Ye, F. & Deem, M. W. Machine-learning approach to the design of OSDAs for zeolite beta. Proc. Natl Acad. Sci. USA 116, 3413–3418 (2019).

    Article  Google Scholar 

  14. Muraoka, K., Chaikittisilp, W. & Okubo, T. Multi-objective: de novo molecular design of organic structure-directing agents for zeolites using nature-inspired ant colony optimization. Chem. Sci. 11, 8214–8223 (2020).

    Article  Google Scholar 

  15. Schmidt, J. E., Deem, M. W., Lew, C. & Davis, T. M. Computationally-guided synthesis of the 8-ring zeolite AEI. Top. Catal. 58, 410–415 (2015).

    Article  Google Scholar 

  16. Jensen, Z. et al. Discovering relationships between OSDAs and zeolites through data mining and generative neural networks. ACS Cent. Sci. 7, 858–867 (2021).

    Article  Google Scholar 

  17. Schwalbe-Koda, D. et al. Repurposing templates for zeolite synthesis from simulations and data mining. Chem. Mater. 34, 5366–5376 (2022).

    Article  Google Scholar 

  18. Schwalbe-Koda, D. First-Principles Control of Zeolite Synthesis, Transformations, and Intergrowth (Massachusetts Institute of Technology, 2023).

  19. Burton, A. Recent trends in the synthesis of high-silica zeolites. Catal. Rev. Sci. Eng 60, 132–175 (2018).

    Article  Google Scholar 

  20. Bemis, G. W. & Murcko, M. A. The properties of known drugs. 1. Molecular frameworks. J. Med. Chem. 39, 2887–2893 (1996).

    Article  Google Scholar 

  21. Lundberg, S. & Lee, S.-I. A unified approach to interpreting model predictions. In Proc. 31st International Conference on NIPS 4768–4777 (Curran Associates, 2017).

  22. Goldman, S., Das, R., Yang, K. K. & Coley, C. W. Machine learning modeling of family wide enzyme-substrate specificity screens. PLoS Comput. Biol. 18, e1009853 (2022).

    Article  Google Scholar 

  23. Lee, H., Shin, J. & Hong, S. B. Tetraethylammonium-mediated zeolite synthesis via a multiple inorganic cation approach. ACS Mater. Lett. 3, 308–312 (2021).

    Article  Google Scholar 

  24. Moliner, M., Martínez, C. & Corma, A. Synthesis strategies for preparing useful small pore zeolites and zeotypes for gas separations and catalysis. Chem. Mater. 26, 246–258 (2014).

    Article  Google Scholar 

  25. Dusselier, M. & Davis, M. E. Small-pore zeolites: synthesis and catalysis. Chem. Rev. 118, 5265–5329 (2018).

    Article  Google Scholar 

  26. Miller, M. A. et al. in Studies in Surface Science and Catalysis Vol. 170, 487–492 (Elsevier, 2007).

  27. Martín, N. et al. Cage-based small-pore catalysts for NH3-SCR prepared by combining bulky organic structure directing agents with modified zeolites as reagents. Appl. Catal. B 217, 125–136 (2017).

    Article  Google Scholar 

  28. Lee, J. H. et al. Synthesis and characterization of ERI-type UZM-12 zeolites and their methanol-to-olefin performance. J. Am. Chem. Soc. 132, 12971–12982 (2010).

    Article  Google Scholar 

  29. Xie, D. & Lew, C. Method for preparing zeolite SSZ-98. US20170O88432A1 (2017).

  30. Vennestrøm, P. N. R. et al. Advances and perspectives from a decade of collaborative efforts on zeolites for selective catalytic reduction of NOx. Microporous Mesoporous Mater. 358, 112336 (2023).

    Article  Google Scholar 

  31. Shan, Y. et al. Selective catalytic reduction of NOx with NH3: opportunities and challenges of Cu-based small-pore zeolites. Natl Sci. Rev. 8, nwab010 (2021).

    Article  Google Scholar 

  32. Zones, S. I. Conversion of faujasites to high-silica chabazite SSZ-13 in the presence of N,N,N-trimethyl-1-adamantammonium iodide. J. Chem. Soc. Faraday Trans. 87, 3709 (1991).

    Article  Google Scholar 

  33. Cao, G. et al. Synthesis of chabazite-containing molecular sieves and their use in the conversion of oxygenates to olefins. US7094389B2 (2006).

  34. Shan, Y. et al. A comparative study of the activity and hydrothermal stability of Al-rich Cu-SSZ-39 and Cu-SSZ-13. Appl. Catal. B 264, 118511 (2020).

    Article  Google Scholar 

  35. Bello, E. et al. NH3-SCR catalysts for heavy-duty diesel vehicles: preparation of CHA-type zeolites with low-cost templates. Appl. Catal. B 303, 120928 (2022).

    Article  Google Scholar 

  36. Yang, S., Turrina, A., Youngner, L., & Gilleland, D. A method of synthesizing a low SAR chabazite zeolite and the zeolite obtained thereby. GB2609550A (2023).

  37. Xie, L. et al. Excellent performance of one-pot synthesized Cu-SSZ-13 catalyst for the selective catalytic reduction of NOx with NH3. Environ. Sci. Technol. 48, 566–572 (2014).

    Article  Google Scholar 

  38. Wang, J. et al. One-pot synthesis of Na+-free Cu-SSZ-13 and its application in the NH3–SCR reaction. Chem. Commun. 57, 4898–4901 (2021).

    Article  Google Scholar 

  39. Moini, A., Kunkes, E. L., Thomas, J. C., Vattipalli, V. & Castellano, C. R. Zeolite structure synthesized using mixtures of organic structure directing agents. EP4065513A1 (2022).

  40. Perez-Ramirez, J. et al. Evolution of isomorphously substituted iron zeolites during activation: comparison of Fe-beta and Fe-ZSM-5. J. Catal. 232, 318–334 (2005).

    Article  Google Scholar 

  41. Lin, Q.-F. et al. A stable aluminosilicate zeolite with intersecting three-dimensional extra-large pores. Science 374, 1605–1608 (2021).

    Article  Google Scholar 

  42. Lu, P. et al. A stable zeolite with atomically ordered and interconnected mesopore channel. Nature 636, 368–373 (2024).

    Article  Google Scholar 

  43. Hoffman, A. J., Xie, M. & Gómez-Bombarelli, R. Learning descriptors to predict organic structure-directing agent applicability in zeolite synthesis. Microporous Mesoporous Mater. 386, 113467 (2025).

    Article  Google Scholar 

  44. Pan, E. et al. A chemically-guided generative diffusion model for materials synthesis planning. In NeurIPS AI4Mat Workshop https://openreview.net/forum?id=hy39qxU6CQ (2024).

  45. Gao, W. & Coley, C. W. The synthesizability of molecules proposed by generative models. J. Chem. Inf. Model. 60, 5714–5723 (2020).

    Article  Google Scholar 

  46. Ferri, P. et al. Approaching enzymatic catalysis with zeolites or how to select one reaction mechanism competing with others. Nat. Commun. 14, 2878 (2023).

    Article  Google Scholar 

  47. RDKit: open-source cheminformatics (version 2021.03.4). http://www.rdkit.org (2021).

  48. Tosco, P., Stiefl, N. & Landrum, G. Bringing the MMFF force field to the RDKit: implementation and validation. J. Cheminformatics 6, 37 (2014).

    Article  Google Scholar 

  49. Schwalbe-Koda, D. & Gómez-Bombarelli, R. Supramolecular recognition in crystalline nanocavities through Monte Carlo and Voronoi network algorithms. J. Phys. Chem. C 125, 3009–3017 (2021).

    Article  Google Scholar 

  50. Mayo, S. L., Olafson, B. D. & Goddard, W. A. DREIDING: a generic force field for molecular simulations. J. Phys. Chem. 94, 8897–8909 (1990).

    Article  Google Scholar 

  51. Gale, J. D. & Rohl, A. L. The General Utility Lattice Program (GULP). Mol. Simul. 29, 291–341 (2003).

    Article  Google Scholar 

  52. Schwalbe-Koda, D. learningmatter-mit/gulpy: GULPy 1.0. Zenodo https://doi.org/10.5281/zenodo.5260056 (2021).

  53. Schwalbe-Koda, D. & Gómez-Bombarelli, R. Benchmarking binding energy calculations for organic structure-directing agents in pure-silica zeolites. J. Chem. Phys. 154, 174109 (2021).

    Article  Google Scholar 

  54. Todeschini, R. & Gramatica, P. SD-modelling and prediction by WHIM descriptors. Part 5. Theory development and chemical meaning of WHIM descriptors. Mol. Informatics 16, 113–119 (1997).

    Google Scholar 

  55. Consonni, V., Todeschini, R. & Pavan, M. Structure/response correlations and similarity/diversity analysis by GETAWAY descriptors. 1. Theory of the novel 3D molecular descriptors. J. Chem. Inf. Comput. Sci. 42, 682–692 (2002).

    Article  Google Scholar 

  56. Consonni, V., Todeschini, R., Pavan, M. & Gramatica, P. Structure/response correlations and similarity/diversity analysis by GETAWAY descriptors. 2. Application of the novel 3D molecular descriptors to QSAR/QSPR studies. J. Chem. Inf. Comput. Sci. 42, 693–705 (2002).

    Article  Google Scholar 

  57. Willems, T. F., Rycroft, C. H., Kazi, M., Meza, J. C. & Haranczyk, M. Algorithms and tools for high-throughput geometry-based analysis of crystalline porous materials. Microporous Mesoporous Mater. 149, 134–141 (2012).

    Article  Google Scholar 

  58. Jones, A. J., Ostrouchov, C., Haranczyk, M. & Iglesia, E. From rays to structures: representation and selection of void structures in zeolites using stochastic methods. Microporous Mesoporous Mater. 181, 208–216 (2013).

    Article  Google Scholar 

  59. Chen, T. & Guestrin, C. XGBoost: a scalable tree boosting system. In Proc. 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining 785–794 (Association for Computing Machinery, 2016).

  60. Dewancker, I. et al. Evaluation system for a Bayesian optimization service. Preprint at http://arxiv.org/abs/1605.06170 (2016).

  61. Xie, M. et al. An exhaustive mapping of zeolite–template chemical space. Materials Data Facility https://doi.org/10.18126/TRPY-ER69 (2025).

  62. Xie, M. & Gomez-Bombarelli, R. learningmatter-mit/zeobind: initial release. Zenodo https://doi.org/10.5281/zenodo.15425842 (2025).

  63. Crameri, F., Shephard, G. E. & Heron, P. J. The misuse of colour in science communication. Nat. Commun. 11, 5444 (2020).

    Article  Google Scholar 

  64. Brand, A., Allen, L., Altman, M., Hlava, M. & Scott, J. Beyond authorship: attribution, contribution, collaboration, and credit. Learn. Publ. 28, 151–155 (2015).

    Article  Google Scholar 

Download references

Acknowledgements

M.M., C.P., E.B.-J. and Y.M.S.-E. acknowledge financial support by the Spanish Government through PID2021-122755OB-I00 (funded by MCIN/AEI/10.13039/501100011033) and TED2021-130739B–I00 (funded by MCIN/AEI/10.13039/501100011033/EU/PRTR), by the Generalitat Valenciana through the Prometeo Program (CIPROM/2023/34) and CSIC through the I-link+ Program (ILINK24035). M.M., C.P., E.B.-J. and Y.M.S.-E. also thank the Severo Ochoa financial support by the Spanish Ministry of Science and Innovation (CEX2021-001230-S/funding by MCIN/AEI/10.13039/501100011033). M.X. acknowledges funding from the Agency of Science, Technology and Research (A*STAR) scholarship. D.S.-K. acknowledges funding from the MIT Energy Fellowship. E.B.-J. and Y.M.S.-E. acknowledge the Spanish Government for an FPI scholarship (PRE2019-088360) and a Severo Ochoa FPI scholarship (PRE2020-092319), respectively. R.G.-B. and A.H. acknowledge funding from MIT Deshpande Center, Alfred P. Sloan Research Fellowship (FG-2023-20026) and MISTI Inditex Fund. The Electron Microscopy Service of the UPV is also acknowledged for their help in sample characterization. Computations were executed on MIT Engaging and Supercloud clusters.

Author information

Authors and Affiliations

  1. Department of Chemical Engineering, Massachusetts Institute of Technology, Cambridge, MA, USA

    Mingrou Xie

  2. Department of Materials Science and Engineering, Massachusetts Institute of Technology, Cambridge, MA, USA

    Daniel Schwalbe-Koda, Alexander Hoffman, Omar Santiago-Reyes & Rafael Gómez-Bombarelli

  3. Department of Materials Science and Engineering, University of California, Los Angeles, Los Angeles, CA, USA

    Daniel Schwalbe-Koda

  4. Instituto de Tecnología Química, Universitat Politècnica de València – Consejo Superior de Investigaciones Científicas, Valencia, Spain

    Yolanda Marcela Semanate-Esquivel, Estefanía Bello-Jurado, Cecilia Paris & Manuel Moliner

  5. Division of Chemistry and Chemical Engineering, California Institute of Technology, Pasadena, CA, USA

    Omar Santiago-Reyes

Authors
  1. Mingrou Xie
    View author publications

    Search author on:PubMed Google Scholar

  2. Daniel Schwalbe-Koda
    View author publications

    Search author on:PubMed Google Scholar

  3. Yolanda Marcela Semanate-Esquivel
    View author publications

    Search author on:PubMed Google Scholar

  4. Estefanía Bello-Jurado
    View author publications

    Search author on:PubMed Google Scholar

  5. Alexander Hoffman
    View author publications

    Search author on:PubMed Google Scholar

  6. Omar Santiago-Reyes
    View author publications

    Search author on:PubMed Google Scholar

  7. Cecilia Paris
    View author publications

    Search author on:PubMed Google Scholar

  8. Manuel Moliner
    View author publications

    Search author on:PubMed Google Scholar

  9. Rafael Gómez-Bombarelli
    View author publications

    Search author on:PubMed Google Scholar

Contributions

Following the CRediT taxonomy64, D.S.-K. and R.G.-B. conceptualized the project. D.S.-K. developed the software and methodology for the training data and conducted initial studies. D.S.-K., O.S.-R. and A.H. curated the hypothetical molecule library. M.X. curated the datasets, developed the ML models and screening workflow, and carried out the high-throughput screening and data analysis. C.P. synthesized the organic molecules used as OSDAs for the zeolite preparations. Y.M.S.-E. and E.B.-J. synthesized and characterized the zeolitic materials. E.B.-J. performed the catalytic tests. M.X. created the visuals and wrote the original draft of the paper. All authors reviewed and edited the paper. M.M. and R.G.-B. directed and supervised the work.

Corresponding authors

Correspondence to Manuel Moliner or Rafael Gómez-Bombarelli.

Ethics declarations

Competing interests

The authors declare no competing interests.

Peer review

Peer review information

Nature Computational Science thanks Hilal Daglar, Ben Slater and the other, anonymous, reviewer(s) for their contribution to the peer review of this work. Peer reviewer reports are available. Primary Handling Editor: Kaitlin McCardle, in collaboration with the Nature Computational Science team.

Additional information

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary information

Supplementary Information

Supplementary Figs. 1–34, Tables 1–6 and Sections 1–7.

Peer Review File

Supplementary Data 1

Mapping between the actual loading (of molecules) per unit cell and the loading class in the multiclassification task.

Supplementary Data 2

Performance metrics for ensembling methods for binding energy (BE), labeled (b: ensembling method, e: ensembling method), ordered by RMSE. Where the ensembling method is a number, it indicates that it is a single model prediction rather than an ensemble prediction. Rows containing single model predictions for both tasks are highlighted in blue. The selected ensembling method is highlighted in green. RMSE in kJ mol−1 Si; MSE in (kJ mol−1 Si)2; MAE, mean absolute error in kJ mol−1 Si.

Supplementary Data 3

Frequency of frameworks predicted as the best framework by BE for a monoquaternary hypothetical molecule.

Supplementary Data 4

Frequency of frameworks predicted as the second-best framework by BE for a monoquaternary hypothetical molecule. Frequency of frameworks predicted as the best framework by BE for a diquaternary hypothetical molecule.

Supplementary Data 5

Frequency of frameworks predicted as the best framework by BE for a diquaternary hypothetical molecule.

Supplementary Data 6

Frequency of frameworks predicted as the second-best framework by BE for a diquaternary hypothetical molecule.

Source data

Source Data Fig. 2

Source data for plots in Fig. 2.

Source Data Fig. 3

Source data for plots in Fig. 3.

Source Data Fig. 4

Source data for plots in Fig. 4.

Source Data Fig. 5

Source data for plots in Fig. 5.

Rights and permissions

Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Xie, M., Schwalbe-Koda, D., Semanate-Esquivel, Y.M. et al. A comprehensive mapping of zeolite–template chemical space. Nat Comput Sci 5, 661–674 (2025). https://doi.org/10.1038/s43588-025-00842-5

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue date:

  • DOI: https://doi.org/10.1038/s43588-025-00842-5

This article is cited by

Search

Advanced search

Quick links

Nature Briefing

Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.

Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing