Skip to main content

Thank you for visiting 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.

Machine learning reveals how complex molecules bind to catalyst surfaces

A machine learning method is developed and used to predict the adsorption configurations and energies of complex molecules at the surfaces of transition metals and alloys. This method will be useful for investigating complex reaction networks at complex catalyst materials to understand and improve the performance of heterogeneous catalysts.

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

Access options

Buy article

Get time limited or full article access on ReadCube.


All prices are NET prices.

Fig. 1: Schematic illustration of the WWL-GPR method.


  1. Nørskov, J. K., Abild-Pedersen, F., Studt, F. & Bligaard, T. Density functional theory in surface chemistry and catalysis. Proc. Natl Acad. Sci. USA 108, 937–943 (2011). A review article that presents the role of QM calculations in catalysis research.

    Article  Google Scholar 

  2. Kitchin, J. R. Machine learning in catalysis. Nat. Catal. 1, 230–232 (2018). A review article that presents the role of machine learning methods in catalysis research.

    Article  Google Scholar 

  3. Togninalli, M., Ghisu, E., Llinares-López, F., Rieck, B. & Borgwardt, K. Wasserstein Weisfeiler-Lehman graph kernels. Adv. Neural Inf. Process. Syst. 578, 6439–6449 (2019). This paper proposes the Wasserstein Weisfeiler–Lehman graph kernel used in our work.

    Google Scholar 

  4. Chanussot, L. et al. Open Catalyst 2020 (OC20) dataset and community challenges. ACS Catal. 11, 6059–6072 (2021). This paper presents open datasets and challenges for the machine learning community in catalysis research.

  5. Xu, W., Andersen, M. & Reuter, K. Data-driven descriptor engineering and refined scaling relations for predicting transition metal oxide reactivity. ACS Catal 11, 734–742 (2020). This paper reports a machine learning approach for predicting the adsorption energies of atoms and small molecules at metal oxides.

    Article  Google Scholar 

Download references

Additional information

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

This is a summary of: Xu, W. et al. Predicting binding motifs of complex adsorbates using machine learning with a physics-inspired graph representation. Nat. Comput. Sci. (2022).

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Machine learning reveals how complex molecules bind to catalyst surfaces. Nat Comput Sci 2, 477–478 (2022).

Download citation

  • Published:

  • Issue Date:

  • DOI:


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