Pith. sign in

REVIEW 3 major objections 4 minor 69 references

Machine Learning Accelerated Descriptor Design for Catalyst Discovery in CO$_2$ to Methanol Conversion

T0 review · 3 major / 4 minor · reviewed 2026-08-11 · deepseek-v4-flash

Pith's one-line read The paper proposes a new catalyst descriptor—the distribution of adsorption energies across facets and sites—and uses it to flag ZnRh and ZnPt3 as promising, untested candidates for CO2-to-methanol conversion.

desk verdict A sensible, honestly limited screening study whose two candidate predictions rest on a clustering step that the paper never checks against its own MLFF error bars. read the letter →

arxiv 2412.13838 v5 pith:CELBWFJQ submitted 2024-12-18 physics.chem-ph cond-mat.mtrl-sciphysics.comp-ph

classification physics.chem-phcond-mat.mtrl-sciphysics.comp-ph
keywords CO2hydrogenationmethanolsynthesisadsorptionenergydistributionmachine-learnedforcefieldscatalystdiscoveryhierarchicalclusteringWassersteindistancebimetallicalloys
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

This paper tries to establish a fast, general route to catalyst discovery: characterize a material not by one binding energy but by the whole distribution of adsorption energies of key reaction intermediates across all its surface facets and binding sites. Using a pre-trained machine-learned force field, the authors compute these adsorption-energy distributions for nearly 160 metallic and bimetallic alloys relevant to CO2-to-methanol conversion, generating over 877,000 adsorption energies. They then compare the distributions with the Wasserstein distance, cluster the materials hierarchically, and find that many known active catalysts—Cu, CuZn, NiZn, Ga2Cu, InPt3, Ni—fall into one cluster. On that basis they propose ZnRh and ZnPt3, which sit in the same cluster but have not been tested, as likely high-activity candidates, with melting points above Cu/CuZn that suggest better stability under reaction conditions.

What carries the argument

The load-bearing object is the adsorption energy distribution (AED): for each material, a normalized histogram with 0.1 eV bins covering the adsorption energies of four intermediates (*H, *OH, *OCHO, *OCH3) across all generated low-Miller-index surface terminations and adsorption sites, with padding so the four adsorbate distributions do not overlap. The AED replaces the usual practice of quoting a single surface minimum or a d-band center. Comparison is done with the first Wasserstein distance, which measures how much probability mass must be moved to turn one distribution into another, followed by agglomerative hierarchical clustering with variance-minimizing linkage. A companion statistical summary (SAAED) condenses each AED into minimum, spread, and higher moments so individual binding energies can be checked against the Sabatier principle, for example against the oxygen-binding volcano for Cu-based methanol synthesis.

What would settle it

Measure the methanol yield, selectivity, and stability of ZnRh and ZnPt3 under standard CO2 hydrogenation conditions (roughly 800 K, H2/CO2 feed) and compare with CuZn or industrial Cu/ZnO/Al2O3; if their performance is not in the same league despite AED similarity to known active catalysts, the descriptor's central assumption fails.

Watch

Extended reading notes

Core claim

The paper's central claim is that an adsorption energy distribution (AED)—a normalized histogram of adsorption energies for *H, *OH, *OCHO, and *OCH3 collected over multiple Miller-index surfaces and symmetry-distinct sites—can serve as a material fingerprint for catalytic activity in CO2-to-methanol conversion. After validating the machine-learned force field against density functional theory on Pt, Zn, and NiZn and excluding materials whose estimated error exceeds 0.25 eV, the authors build AEDs for 158 materials. Hierarchical clustering of the Wasserstein distance matrix yields 19 clusters, one of which (cluster 10) groups known active catalysts with ZnRh and ZnPt3. The paper therefore proposes ZnRh and ZnPt3 as promising candidates that should have high activity and, because their melting temperatures exceed those of Cu and CuZn, possibly better durability.

Load-bearing premise

The assumption that carries the load is that two materials with similar adsorption-energy distributions will have similar catalytic activity; the paper states this explicitly and does not validate it with experiment or microkinetic modeling.

Editorial extensions

If this is right

  • ZnRh and ZnPt3 become concrete experimental targets: if they show methanol activity comparable to CuZn or NiZn, the AED-similarity route is a validated screening tool.
  • The same pipeline can be pointed at other reactions by swapping the four adsorbates for that reaction's key intermediates, so the descriptor is not limited to CO2-to-methanol.
  • Materials with broad AEDs, such as Y3Zn11, V6Ga5, Y3In5, and V4Zn5, are predicted to be poorer catalysts because only a small fraction of their surface binds intermediates near the optimum.
  • The melting-point argument implies that ZnRh and ZnPt3 may retain stability where Cu-based catalysts degrade, which is directly testable in long-run reactor tests.
  • Excluding 29 materials with large estimated errors—many of them magnetic—means the current search space is biased against magnetic alloys; including spin polarization could extend the method.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • A natural next step is to test whether cluster membership is robust to the choice of adsorbates; adding CO or other intermediates could shift ZnRh and ZnPt3 relative to the known catalysts and would reveal which energy matches are doing the work.
  • Because the descriptor averages over facets without weighting them by their surface areas, combining AEDs with Wulff-construction facet weights could sharpen the comparison for nanoparticles whose exposed facets change under reaction conditions.
  • If the similarity assumption holds, the same clustering logic could be applied across reaction families: a catalyst found active for CO2-to-methanol might seed searches for methanation or higher-alcohol synthesis by using the appropriate intermediate set.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

3 major / 4 minor

Summary. The paper introduces a new catalytic descriptor, the adsorption energy distribution (AED), which aggregates machine-learned force field (MLFF) adsorption energies for four reaction intermediates across many surface facets and binding sites of ~160 metallic alloys. The authors validate the OCP equiformer_V2 MLFF on Pt, Zn, and NiZn, then use a sparse single-point DFT validation protocol (EMAE) to estimate errors and exclude 29 materials with EMAE above 0.25 eV. They compute Wasserstein distances between the AEDs and perform hierarchical clustering with Ward linkage, identifying ZnRh and ZnPt3 as untested candidate catalysts because they cluster with known active materials such as CuZn, NiZn, and Ga2Cu. The paper is transparent about the central assumption that AED similarity to a known good catalyst is a meaningful activity indicator, and it makes the resulting dataset available on Zenodo.

Significance. If the workflow is robust, the AED descriptor is a genuinely new way to represent facet and site heterogeneity in catalyst screening, and it is reaction-tunable through adsorbate choice. The dataset of over 877,000 adsorption energies and the open data publication are concrete strengths, as is the explicit EMAE validation protocol. The candidate nominations, however, rest on cluster assignments that are never checked against the stated MLFF accuracy, and on an untested similarity-to-activity assumption; the former is an internal consistency issue that is checkable and fixable, while the latter limits the strength of the claims that can be drawn.

major comments (3)
  1. [§3.5 / §2 (Validation and Data Cleaning)] The clustering threshold of 2.5×10−3 eV is not reconciled with the MLFF validation MAEs of 0.06–0.16 eV reported in Table 1 or with the 0.25 eV EMAE cutoff used for data cleaning. For the first Wasserstein distance in 1D, a uniform shift of one distribution by δ changes its distance to an unshifted distribution by approximately δ. A systematic MLFF error of 0.1 eV in the adsorption energies of one material therefore shifts its W1 distance to another material by roughly 0.1 eV, which is 40 times larger than the merge threshold. This means that the placement of ZnRh and ZnPt3 in cluster 10, and hence the central candidate claim, may be an artifact of treating MLFF energies as exact. The authors should report EMAEs for ZnRh and ZnPt3, propagate MLFF uncertainties into the distance matrix (for example by bootstrapping adsorption energies with the measured MAE), and demonstrate that the cluster assignments are stable under perturbations of the order of the validation MAE.
  2. [§2 (Unsupervised Learning)] The paper states explicitly that the candidate proposals rest on the assumption that AED similarity to a known good catalyst indicates promising catalytic activity, but this assumption is not validated by any independent test. To support the specific claim that ZnRh and ZnPt3 'should have a high activity,' the authors should either check whether experimental activity rankings among the known catalysts in cluster 10 are reproduced by AED distances, or perform microkinetic modeling for the proposed candidates, or clearly label the candidates as untested hypotheses. As written, the recommendation is stronger than the evidence supports.
  3. [§2 (Validation and Data Cleaning)] The exclusion of 29 materials with EMAE above 0.25 eV is a post hoc cleaning step whose relation to the clustering is unclear. The 0.25 eV threshold is two orders of magnitude larger than the clustering threshold of 2.5×10−3 eV, yet the EMAEs of the retained materials, at least for ZnRh and ZnPt3, are not reported in the text or in the supplement; the reader cannot verify that the AEDs of the proposed candidates are within the accuracy needed to support their cluster assignment. The paper should justify the 0.25 eV cutoff in terms of its effect on the downstream Wasserstein distances and report the relevant EMAE values for the proposed candidates.
minor comments (4)
  1. [Fig. 2 caption] The caption uses 'EMEA' where 'EMAE' is meant.
  2. [References] References [64] and [65] are identical (both cite Ong et al., pymatgen); one should be replaced with the correct Custodian reference.
  3. [§2 (Adsorption Energy Distributions)] The phrase 'the price energy zero is of no relevance' appears to be a typo for 'the precise energy zero is of no relevance'.
  4. [Table 1 caption] The caption contains 'absorbate-material combination', which should be 'adsorbate-material combination'.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the candidate prediction is an unsupervised similarity argument whose descriptor and distances are computed independently of the target activity labels.

full rationale

The central claim—that ZnRh and ZnPt3 are promising because their adsorption energy distributions cluster with experimentally known active catalysts—is an inductive analogy, not a derivation that encodes the answer. The AED descriptor is built from OCP MLFF adsorption energies (validated against DFT in Table 1 and Figure 2) before any activity labels are examined. The Wasserstein distance matrix and Ward clustering (Section 3.5, Eq. 1) operate only on these AEDs; no parameter is fitted to methanol-yield data, and the known 'good' catalysts enter as external literature labels used to interpret clusters ex post. The paper explicitly discloses the load-bearing assumption: 'Our proposals for interesting candidates are based on the assumption that AED similarity with a known good catalyst is a meaningful indicator for promising catalytic activity' (Section 2, Unsupervised Learning), and it enumerates limitations (facet-area insensitivity, support/oxide effects, selectivity). The only self-citation is the Zenodo data record [43], which is data availability, not an argument. A legitimate robustness concern remains—validation MAEs of 0.06–0.16 eV are not propagated into the 2.5e-3 eV Wasserstein merge threshold, so the cluster assignments could change under realistic MLFF error—but that is a correctness/error-propagation issue, not circularity, because the prediction is not equivalent to its inputs by construction. No step in the derivation chain reduces to its own inputs, so the circularity score is 0.

Assumptions & free parameters 3 free parameters · 4 assumptions · 0 invented entities

No new physical entities are introduced. The central result depends on the hand-chosen clustering threshold, the EMAE exclusion cutoff, and the unvalidated similarity-to-activity assumption.

free parameters (3)
  • Clustering threshold = 0.0025 (Wasserstein distance)
    Hand-chosen cut-off for hierarchical clustering; changing it changes which materials group together and thus which candidates are proposed.
  • EMAE exclusion cutoff = 0.25 eV
    Materials with estimated mean absolute error above this threshold are excluded from the dataset; 29 materials removed, mostly magnetic.
  • Histogram bin width = 0.1 eV
    Bin width chosen for the AED histograms; affects the resolution of the distributions and the Wasserstein distances.
assumptions (4)
  • domain assumption The OCP equiformer_V2 MLFF provides sufficient accuracy for qualitative AED screening.
    Reported accuracy 0.23 eV and measured MAE 0.16 eV on three test materials; assumed to hold for all 158 materials, with EMAE sampling as a partial check.
  • ad hoc to paper Similarity in AEDs to known active catalysts indicates catalytic activity.
    Explicitly stated in Section 2; this is the central premise linking clustering results to candidate predictions, and it is not independently validated.
  • domain assumption RPBE DFT is an adequate ground truth for validation and bulk optimization.
    Used for bulk optimization and single-point validation; DFT functional choice is an established approximation in catalysis.
  • domain assumption The four adsorbates *H, *OH, *OCHO, and *OCH3 capture the key reaction intermediates.
    Selected from experimental literature on Cu(211) model catalyst; other intermediates like CO may matter for some alloys, as the authors note.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Machine Learning Accelerated Descriptor Design for Catalyst Discovery in CO$_2$ to Methanol Conversion." pith.science (2026). https://pith.science/paper/CELBWFJQ

@misc{pith2026241213838,
  author       = {Pith},
  title        = {Pith review of: Machine Learning Accelerated Descriptor Design for Catalyst Discovery in CO$_2$ to Methanol Conversion},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/CELBWFJQ}},
  note         = {Machine review of arXiv:2412.13838}
}
abstract

Transforming CO$_2$ into methanol represents a crucial step towards closing the carbon cycle, with thermoreduction technology nearing industrial application. However, obtaining high methanol yields and ensuring the stability of heterocatalysts remain significant challenges. Herein, we present a sophisticated computational framework to accelerate the discovery of thermal heterogeneous catalysts, using machine-learned force fields. We propose a new catalytic descriptor, termed adsorption energy distribution, that aggregates the binding energies for different catalyst facets, binding sites, and adsorbates. The descriptor is versatile and can be adjusted to a specific reaction through careful choice of the key-step reactants and reaction intermediates. By applying unsupervised machine learning and statistical analysis to a dataset comprising nearly 160 metallic alloys, we offer a powerful tool for catalyst discovery. We propose new promising candidates such as ZnRh and ZnPt$_3$, which to our knowledge, have not yet been tested, and discuss their possible advantage in terms of stability.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

69 extracted references · 33 canonical work pages

  1. [1]

    Nature Communications 10(5698) (2019) https://doi.org/10.1038/s41467-019-13638-9

    Ye, R.-P., Ding, J., Gong, W., Argyle, M.D., Zhong, Q., Wang, Y., Russell, C.K., Xu, Z., Russell, A.G., Li, Q., Fan, M., Yao, Y.-G.: CO 2 hydrogenation to high- value products via heterogeneous catalysis. Nature Communications 10(5698) (2019) https://doi.org/10.1038/s41467-019-13638-9

  2. [2]

    Journal of Catalysis 372, 33–38 (2019) https://doi.org/10.1016/j.jcat.2019.01.042

    Rohr, B.A., Singh, A.R., Nørskov, J.K.: A theoretical explanation of the effect of oxygen poisoning on industrial haber-bosch catalysts. Journal of Catalysis 372, 33–38 (2019) https://doi.org/10.1016/j.jcat.2019.01.042

  3. [3]

    Applied Catalysis A: General 138(2), 311–318 (1996) https://doi

    Saito, M., Fujitani, T., Takeuchi, M., Watanabe, T.: Development of copperzinc oxide-based multicomponent catalysts for methanol synthesis from carbon dioxide and hydrogen. Applied Catalysis A: General 138(2), 311–318 (1996) https://doi. org/10.1016/0926-860X(95)00305-3

  4. [4]

    Journal of CO 2 Utilization 39, 101166 (2020) https://doi.org/10.1016/j.jcou.2020.101166

    Ny´ ari, J., Magdeldin, M., Larmi, M., J¨ arvinen, M., Santasalo-Aarnio, A.: Techno- economic barriers of an industrial-scale methanol CCU-plant. Journal of CO 2 Utilization 39, 101166 (2020) https://doi.org/10.1016/j.jcou.2020.101166

  5. [5]

    Renewable and Sustainable Energy Reviews 31, 221–257 (2014) https://doi.org/10.1016/j.rser.2013.11.045

    Ganesh, I.: Conversion of carbon dioxide into methanol – a potential liquid fuel: Fundamental challenges and opportunities (a review). Renewable and Sustainable Energy Reviews 31, 221–257 (2014) https://doi.org/10.1016/j.rser.2013.11.045

  6. [6]

    Nature Catalysis 1, 696–703 (2018) https://doi.org/10.1038/s41929-018-0142-1

    Tran, K., Ulissi, Z.W.: Active learning across intermetallics to guide discovery of electrocatalysts for CO2 reduction and H2 evolution. Nature Catalysis 1, 696–703 (2018) https://doi.org/10.1038/s41929-018-0142-1

  7. [7]

    Nanomaterials 12(15), 2527 (2022) https://doi.org/10.3390/nano12152527 17

    Wang, L., Etim, U.J., Zhang, C., Amirav, L., Zhong, Z.: CO 2 activation and hydrogenation on Cu-ZnO/Al 2O3 nanorod catalysts: An in situ FTIR study. Nanomaterials 12(15), 2527 (2022) https://doi.org/10.3390/nano12152527 17

  8. [8]

    Sci China Chem 66, 3645–3652 (2023) https://doi.org/10.1007/s11426-023-1789-3

    Li, D., Wang, Z., Jin, S., Zhu, M.: Deactivation and regeneration of the commer- cial Cu/ZnO/Al 2O3 catalyst in low-temperature methanol steam reforming. Sci China Chem 66, 3645–3652 (2023) https://doi.org/10.1007/s11426-023-1789-3

Show all 69 references
  1. [9]

    Nature Catalysis 2, 659–670 (2019) https://doi.org/10.1038/s41929-019-0298-3

    Bruix, A., Margraf, J.T., Andersen, M., Reuter, K.: First-principles-based mul- tiscale modelling of heterogeneous catalysis. Nature Catalysis 2, 659–670 (2019) https://doi.org/10.1038/s41929-019-0298-3

  2. [10]

    ACS Catal

    Posada-Borb´ on, A., Gr¨ onbeck, H.: A first-principles-based microkinetic study of CO 2 reduction to CH 3OH over In 2O3(110). ACS Catal. 11(15), 9996–10006 (2021) https://doi.org/10.1021/acscatal.1c01707

  3. [11]

    2013.07.006

    Che, M.: Nobel prize in chemistry 1912 to Sabatier: Organic chemistry or cataly- sis? Catalysis Today 218-219, 162–171 (2013) https://doi.org/10.1016/j.cattod. 2013.07.006 . Catalysis: From the active sites to the processes

  4. [12]

    Journal of Catalysis 328, 36–42 (2015) https://doi.org/10.1016/j.jcat.2014.12.033

    Medford, A.J., Vojvodic, A., Hummelshøj, J.S., Voss, J., Abild-Pedersen, F., Studt, F., Bligaard, T., Nilsson, A., Nørskov, J.K.: From the Sabatier princi- ple to a predictive theory of transition-metal heterogeneous catalysis. Journal of Catalysis 328, 36–42 (2015) https://do...

  5. [13]

    Nature Communications 15(1), 359 (2024) https://doi.org/10.1038/s41467-023-44261-4

    Chen, Z.W., Li, J., Ou, P., Huang, J.E., Wen, Z., Chen, L., Yao, X., Cai, G., Yang, C.C., Singh, C.V., Jiang, Q.: Unusual Sabatier principle on high entropy alloy catalysts for hydrogen evolution reactions. Nature Communications 15(1), 359 (2024) https://doi.org/10.1038/s41467...

  6. [14]

    Journal of Physics: Condensed Matter 20(6), 064239 (2008) https://doi.org/10.1088/ 0953-8984/20/6/064239

    Jones, G., Bligaard, T., Abild-Pedersen, F., Nørskov, J.K.: Using scaling rela- tions to understand trends in the catalytic activity of transition metals. Journal of Physics: Condensed Matter 20(6), 064239 (2008) https://doi.org/10.1088/ 0953-8984/20/6/064239

  7. [15]

    Nature Reviews Chemistry 6(2), 89–111 (2022) https://doi.org/10.1038/ s41570-021-00340-y

    Vogt, C., Weckhuysen, B.M.: The concept of active site in heterogeneous catal- ysis. Nature Reviews Chemistry 6(2), 89–111 (2022) https://doi.org/10.1038/ s41570-021-00340-y

  8. [16]

    npj Com- putational Materials 2(1), 16028 (2016) https://doi.org/10.1038/npjcompumats

    Ward, L., Agrawal, A., Choudhary, A., Wolverton, C.: A general-purpose machine learning framework for predicting properties of inorganic materials. npj Com- putational Materials 2(1), 16028 (2016) https://doi.org/10.1038/npjcompumats. 2016.28

  9. [17]

    ACS Catalysis 7(10), 6600–6608 (2017) https://doi.org/10.1021/acscatal.7b01648

    Ulissi, Z.W., Tang, M.T., Xiao, J., Liu, X., Torelli, D.A., Karamad, M., Cummins, K., Hahn, C., Lewis, N.S., Jaramillo, T.F., Chan, K., Nørskov, J.K.: Machine- learning methods enable exhaustive searches for active bimetallic facets and reveal active site motifs for CO 2 reduc...

  10. [18]

    ACS Catalysis 9, 2752–2759 (2019) https://doi.org/10.1021/acscatal.8b04478

    Andersen, M., Levchenko, S.V., Scheffler, M., Reuter, K.: Beyond scaling relations 18 for the description of catalytic materials. ACS Catalysis 9, 2752–2759 (2019) https://doi.org/10.1021/acscatal.8b04478

  11. [19]

    Advanced Sci- ence 6(21), 1900808 (2019) https://doi.org/10.1002/advs.201900808 https://advanced.onlinelibrary.wiley.com/doi/pdf/10.1002/advs.201900808

    Himanen, L., Geurts, A., Foster, A.S., Rinke, P.: Data-driven materials science: Status, challenges, and perspectives. Advanced Sci- ence 6(21), 1900808 (2019) https://doi.org/10.1002/advs.201900808 https://advanced.onlinelibrary.wiley.com/doi/pdf/10.1002/advs.201900808

  12. [20]

    Small Methods 5(11), 2100987 (2021) https: //doi.org/10.1002/smtd.202100987

    Zhang, N., Yang, B., Liu, K., Li, H., Chen, G., Qiu, X., Li, W., Hu, J., Fu, J., Jiang, Y., Liu, M., Ye, J.: Machine learning in screening high performance electrocatalysts for CO 2 reduction. Small Methods 5(11), 2100987 (2021) https: //doi.org/10.1002/smtd.202100987

  13. [21]

    : AdsorbML: a leap in efficiency for adsorption energy calculations using generalizable machine learning potentials

    Lan, J., Palizhati, A., Shuaibi, M., et al. : AdsorbML: a leap in efficiency for adsorption energy calculations using generalizable machine learning potentials. npj Comput Mater 9, 172 (2023) https://doi.org/10.1038/s41524-023-01121-5

  14. [22]

    Advanced Sci- ence 10(22), 2301020 (2023) https://doi.org/10.1002/advs.202301020 https://onlinelibrary.wiley.com/doi/pdf/10.1002/advs.202301020

    Mou, L.-H., Han, T., Smith, P.E.S., Sharman, E., Jiang, J.: Machine learning descriptors for data-driven catalysis study. Advanced Sci- ence 10(22), 2301020 (2023) https://doi.org/10.1002/advs.202301020 https://onlinelibrary.wiley.com/doi/pdf/10.1002/advs.202301020

  15. [23]

    npj Computational Materials 4(1), 25 (2018) https://doi.org/ 10.1038/s41524-018-0081-z

    Zhang, Y., Ling, C.: A strategy to apply machine learning to small datasets in materials science. npj Computational Materials 4(1), 25 (2018) https://doi.org/ 10.1038/s41524-018-0081-z

  16. [24]

    npj Computational Materials 6(1), 177 (2020) https://doi.org/10.1038/s41524-020-00447-8

    Mamun, O., Winther, K.T., Boes, J.R., Bligaard, T.: A bayesian framework for adsorption energy prediction on bimetallic alloy catalysts. npj Computational Materials 6(1), 177 (2020) https://doi.org/10.1038/s41524-020-00447-8

  17. [25]

    npj Computational Materials 9(1), 115 (2023) https: //doi.org/10.1038/s41524-023-01070-z

    Fiedler, L., Modine, N.A., Schmerler, S., Vogel, D.J., Popoola, G.A., Thompson, A.P., Rajamanickam, S., Cangi, A.: Predicting electronic structures at any length scale with machine learning. npj Computational Materials 9(1), 115 (2023) https: //doi.org/10.1038/s41524-023-01070-z

  18. [26]

    : Towards atom-level understand- ing of metal oxide catalysts for the oxygen evolution reaction with machine learning

    Lunger, J.R., Karaguesian, J., Chun, H., et al. : Towards atom-level understand- ing of metal oxide catalysts for the oxygen evolution reaction with machine learning. npj Computational Materials 10, 80 (2024) https://doi.org/10.1038/ s41524-024-01273-y

  19. [27]

    Engineering 39, 25–44 (2024) https://doi.org/10.1016/j.eng.2023.07.021

    Liu, X., Peng, H.-J.: Toward next-generation heterogeneous catalysts: Empow- ering surface reactivity prediction with machine learning. Engineering 39, 25–44 (2024) https://doi.org/10.1016/j.eng.2023.07.021

  20. [28]

    Accounts of Chemical Research 53(10), 2119–2129 (2020) https://doi.org/10

    Kang, P.-L., Shang, C., Liu, Z.-P.: Large-scale atomic simulation via machine learning potentials constructed by global potential energy surface exploration. Accounts of Chemical Research 53(10), 2119–2129 (2020) https://doi.org/10. 19 1021/acs.accounts.0c00472

  21. [29]

    npj Computational Materials 9(1), 2 (2023) https://doi.org/ 10.1038/s41524-022-00959-5

    Chen, D., Shang, C., Liu, Z.-P.: Machine-learning atomic simulation for hetero- geneous catalysis. npj Computational Materials 9(1), 2 (2023) https://doi.org/ 10.1038/s41524-022-00959-5

  22. [30]

    Science 376(6593), 603–608 (2022) https://doi.org/10.1126/science.abj7747 https://www.science.org/doi/pdf/10.1126/science.abj7747

    Amann, P., Kl¨ otzer, B., Degerman, D., K¨ opfle, N., G¨ otsch, T., L¨ omker, P., Rame- shan, C., Ploner, K., Bikaljevic, D., Wang, H.-Y., Soldemo, M., Shipilin, M., Goodwin, C.M., Gladh, J., Stenlid, J.H., B¨ orner, M., Schlueter, C., Nilsson, A.: The state of zinc in methano...

  23. [31]

    Joule 3(3), 834–845 (2019) https://doi.org/10.1016/j.joule.2018.12.015

    Batchelor, T.A., Pedersen, J.K., Winther, S.H., Castelli, I.E., Jacobsen, K.W., Rossmeisl, J.: High-entropy alloys as a discovery platform for electrocatalysis. Joule 3(3), 834–845 (2019) https://doi.org/10.1016/j.joule.2018.12.015

  24. [32]

    ACS catalysis10(3), 2169–2176 (2020) https://doi.org/10.1021/acscatal.9b04343

    Pedersen, J.K., Batchelor, T.A., Bagger, A., Rossmeisl, J.: High-entropy alloys as catalysts for the CO2 and CO reduction reactions. ACS catalysis10(3), 2169–2176 (2020) https://doi.org/10.1021/acscatal.9b04343

  25. [33]

    ACS Catalysis 11(10), 6059–6072 (2021) https://doi.org/ 10.1021/acscatal.0c04525

    Chanussot, L., Das, A., Goyal, S., Lavril, T., Shuaibi, M., Riviere, M., Tran, K., Heras-Domingo, J., Ho, C., Hu, W., Palizhati, A., Sriram, A., Wood, B., Yoon, J., Parikh, D., Zitnick, C.L., Ulissi, Z.: Open Catalyst 2020 (OC20) dataset and community challenges. ACS Catalysis...

  26. [34]

    ACS Catal

    Tran, R., Lan, J., Shuaibi, M., Wood, B.M., Goyal, S., Das, A., Heras-Domingo, J., Kolluru, A., Rizvi, A., Shoghi, N., Sriram, A., Therrien, F., Abed, J., Voznyy, O., Sargent, E.H., Ulissi, Z., Zitnick, C.L.: The Open Catalyst 2022 (OC22) dataset and challenges for oxide elect...

  27. [35]

    Entropy 19(2), 47 (2017) https:// doi.org/10.3390/e19020047

    Ramdas, A., Garc ´ ıa Trillos, N., Cuturi, M.: On Wasserstein two-sample testing and related families of nonparametric tests. Entropy 19(2), 47 (2017) https:// doi.org/10.3390/e19020047

  28. [36]

    Journal of Cleaner Production 339, 130653 (2022) https://doi.org/10.1016/j.jclepro.2022.130653

    Bahri, S., Pathak, S., Singhahluwalia, A., Malav, P., Upadhyayula, S.: Meta- analysis approach for understanding the characteristics of CO2 reduction catalysts for renewable fuel production. Journal of Cleaner Production 339, 130653 (2022) https://doi.org/10.1016/j.jclepro.2022.130653

  29. [37]

    APL Materials 1(1), 011002 (2013) 20 https://doi.org/10.1063/1.4812323 https://pubs.aip.org/aip/apm/article- pdf/doi/10.1063/1.4812323/13163869/011002 1 online.pdf

    Jain, A., Ong, S.P., Hautier, G., Chen, W., Richards, W.D., Dacek, S., Cholia, S., Gunter, D., Skinner, D., Ceder, G., Persson, K.A.: Commentary: The Materials Project: A materials genome approach to accelerating materials innovation. APL Materials 1(1), 011002 (2013) 20 https...

  30. [38]

    Hammer, B., Hansen, L.B., Nørskov, J.K.: Improved adsorption energetics within density-functional theory using revised Perdew-Burke-Ernzerhof functionals. Phys. Rev. B 59, 7413–7421 (1999) https://doi.org/10.1103/PhysRevB.59.7413

  31. [39]

    Alam, M.I., Cheula, R., Moroni, G., Nardi, L., Maestri, M.: Mechanistic and multiscale aspects of thermo-catalytic co 2 conversion to c1 products. Catal. Sci. Technol. 11, 6601–6629 (2021) https://doi.org/10.1039/D1CY00922B

  32. [40]

    Journal of Advances in Nanomaterials 2(1), 1–10 (2017) https://doi.org/10.22606/jan.2017.21001

    Wu, Z., Cole, J., Fang, H.L., Qin, M., He, Z.: Revisiting catalyst structure and mechanism in methanol synthesis. Journal of Advances in Nanomaterials 2(1), 1–10 (2017) https://doi.org/10.22606/jan.2017.21001

  33. [41]

    GitHub (2024)

    F AIR-Chem: fairchem: A F AIR-Chem Project Repository. GitHub (2024). https: //github.com/F AIR-Chem/fairchem

  34. [42]

    arXiv preprint arXiv:2306.12059 (2023) https://doi.org/10.48550/arXiv.2306.12059

    Liao, Y.-L., Wood, B., Das, A., Smidt, T.: Equiformerv2: Improved equiv- ariant transformer for scaling to higher-degree representations. arXiv preprint arXiv:2306.12059 (2023) https://doi.org/10.48550/arXiv.2306.12059 . Published as a conference paper at ICLR 2024

  35. [43]

    https://doi.org/10.5281/zenodo.13370374

    Pisal, P., Krejci, O.: Final Geometries and Energies, Statistical Analysis and Estimated Errors of Single Metals and Bimetallics for CO 2 to Methanol Conversion. https://doi.org/10.5281/zenodo.13370374 . https://doi.org/10.5281/ zenodo.13370374

  36. [44]

    Journal of the American Statistical Association 58(301), 236–244 (1963) https://doi.org/10

    Jr., J.H.W.: Hierarchical grouping to optimize an objective function. Journal of the American Statistical Association 58(301), 236–244 (1963) https://doi.org/10. 1080/01621459.1963.10500845

  37. [45]

    ChemCatChem 15(1), 202201150 (2023) https://doi.org/10.1002/cctc.202201150 https://chemistry- europe.onlinelibrary.wiley.com/doi/pdf/10.1002/cctc.202201150

    Dongapure, P., Tekawadia, J., Thundiyil, S., Caha, I., Deepak, F.L., Devi, R.N.: Mechanistic insights into near ambient pressure activity of intermetal- lic NiZn/TiO 2 catalyst for CO 2 conversion to methanol. ChemCatChem 15(1), 202201150 (2023) https://doi.org/10.1002/cctc.20...

  38. [46]

    Chemical Reviews 124(8), 4543–4678 (2024) https://doi.org/10.1021/acs.chemrev.3c00148

    Beck, A., Newton, M.A., Water, L.G.A., Bokhoven, J.A.: The enigma of methanol synthesis by cu/zno/al2o3-based catalysts. Chemical Reviews 124(8), 4543–4678 (2024) https://doi.org/10.1021/acs.chemrev.3c00148

  39. [47]

    Science 336(6083), 893–897 (2012) https://doi.org/10.1126/science.1219831 21 https://www.science.org/doi/pdf/10.1126/science.1219831

    Behrens, M., Studt, F., Kasatkin, I., K¨ uhl, S., H¨ avecker, M., Abild- Pedersen, F., Zander, S., Girgsdies, F., Kurr, P., Kniep, B.-L., Tovar, M., Fischer, R.W., Nørskov, J.K., Schl¨ ogl, R.: The active site of methanol synthesis over Cu/ZnO/Al 2O3 industrial catalysts. Scie...

  40. [48]

    Nature Communications 11(1), 3898 (2020) https://doi.org/10.1038/s41467-020-17631-5

    Laudenschleger, D., Ruland, H., Muhler, M.: Identifying the nature of the active sites in methanol synthesis over cu/zno/al2o3 catalysts. Nature Communications 11(1), 3898 (2020) https://doi.org/10.1038/s41467-020-17631-5

  41. [49]

    Nature Chemistry 6(4), 320 (2014) https://doi

    Studt, F., Sharafutdinov, I., Abild-Pedersen, F., Elkjær, C.F., Hummelshøj, J.S., Dahl, S., Chorkendorff, I., Nørskov, J.K.: Discovery of a Ni-Ga catalyst for carbon dioxide reduction to methanol. Nature Chemistry 6(4), 320 (2014) https://doi. org/10.1038/NCHEM.1873

  42. [50]

    Applied Catalysis B: Environmental 34(4), 255–266 (2001) https://doi.org/10.1016/S0926-3373(01)00203-X

    Toyir, J., Ramirez de la Piscina, P., Fierro, J.L.G., Homs, N.: Catalytic per- formance for co 2 conversion to methanol of gallium-promoted copper-based catalysts: influence of metallic precursors. Applied Catalysis B: Environmental 34(4), 255–266 (2001) https://doi.org/10.101...

  43. [51]

    Catalysis Science & Technology7(15), 3375–3387 (2017) https://doi.org/10.1039/ C7CY01021D

    Medina, J.C., Figueroa, M., Manrique, R., Rodr ´ ıguez Pereira, J., Srinivasan, P.D., Bravo-Su´ arez, J.J., Baldovino Medrano, V.G., Jim´ enez, R., Karelovic, A.: Cat- alytic consequences of ga promotion on cu for co 2 hydrogenation to methanol. Catalysis Science & Technology7...

  44. [52]

    Chemical Engineering Science 200, 167–175 (2019) https://doi.org/10.1016/j.ces.2019.02.004

    Men, Y.-L., Liu, Y., Wang, Q., Luo, Z.-H., Shao, S., Li, Y.-B., Pan, Y.-X.: Highly dispersed pt-based catalysts for selective co 2 hydrogenation to methanol at atmospheric pressure. Chemical Engineering Science 200, 167–175 (2019) https://doi.org/10.1016/j.ces.2019.02.004

  45. [53]

    : Thermally stable ni foam-supported inverse CeAlOx/Ni ensemble as an active structured catalyst for CO 2 hydrogena- tion to methane

    Tang, X., Song, C., Li, H., et al. : Thermally stable ni foam-supported inverse CeAlOx/Ni ensemble as an active structured catalyst for CO 2 hydrogena- tion to methane. Nat Commun 15(3115), 3115 (2024) https://doi.org/10.1038/ s41467-024-47403-4

  46. [54]

    Energy & Fuels 36(1), 156–169 (2022) https://doi.org/10.1021/acs

    Hu, F., Ye, R., Lu, Z.-H., Zhang, R., Feng, G.: Structure–activity relationship of Ni-based catalysts toward CO 2 methanation: Recent advances and future per- spectives. Energy & Fuels 36(1), 156–169 (2022) https://doi.org/10.1021/acs. energyfuels.1c03645 https://doi.org/10.10...

  47. [55]

    Lempelto, A., Gell, L., Kiljunen, T., Honkala, K.: Exploring co 2 hydrogenation to methanol at a cuzn–zro2 interface via dft calculations. Catal. Sci. Technol. 13, 4387–4399 (2023) https://doi.org/10.1039/D3CY00549F

  48. [56]

    ACS Catalysis 9(11), 10253–10259 (2019) https://doi.org/10.1021/ acscatal.9b03449

    Wang, J., Tang, C., Li, G., Han, Z., Li, Z., Liu, H., Cheng, F., Li, C.: High- performance mazrox (ma = cd, ga) solid-solution catalysts for co 2 hydrogenation to methanol. ACS Catalysis 9(11), 10253–10259 (2019) https://doi.org/10.1021/ acscatal.9b03449

  49. [57]

    ACS Catalysis 14(17), 13126–13135 (2024) https://doi.org/10.1021/acscatal.4c03206

    Cheula, R., Tran, T.A.M.Q., Andersen, M.: Unraveling the effect of dopants 22 in zirconia-based catalysts for co 2 hydrogenation to methanol. ACS Catalysis 14(17), 13126–13135 (2024) https://doi.org/10.1021/acscatal.4c03206

  50. [58]

    ACS Catalysis 13(3), 1875–1892 (2023) https: //doi.org/10.1021/acscatal.2c04872

    Cannizzaro, F., Hensen, E.J.M., Filot, I.A.W.: The promoting role of ni on in2o3 for co2 hydrogenation to methanol. ACS Catalysis 13(3), 1875–1892 (2023) https: //doi.org/10.1021/acscatal.2c04872

  51. [59]

    Transactions of Tianjin University 28(4), 245–264 (2022) https://doi.org/10.1007/s12209-022-00326-x

    Gao, D., Li, W., Wang, H., Wang, G., Cai, R.: Heterogeneous catalysis for CO 2 conversion into chemicals and fuels. Transactions of Tianjin University 28(4), 245–264 (2022) https://doi.org/10.1007/s12209-022-00326-x

  52. [60]

    Applied Catalysis B: Environment and Energy 356, 124210 (2024) https://doi.org/10.1016/j.apcatb.2024.124210

    Liu, L., Gao, Y., Zhang, H., Kosinov, N., Hensen, E.J.M.: Ni and zro2 promotion of in2o3 for co 2 hydrogenation to methanol. Applied Catalysis B: Environment and Energy 356, 124210 (2024) https://doi.org/10.1016/j.apcatb.2024.124210

  53. [61]

    Kresse, G., Hafner, J.: Ab initio molecular-dynamics simulation of the liquid- metal–amorphous-semiconductor transition in germanium. Phys. Rev. B 49, 14251–14269 (1994) https://doi.org/10.1103/PhysRevB.49.14251

  54. [62]

    Kresse, G., Furthm¨ uller, J.: Efficient iterative schemes for ab initio total-energy calculations using a plane-wave basis set. Phys. Rev. B 54, 11169–11186 (1996) https://doi.org/10.1103/PhysRevB.54.11169

  55. [63]

    Computational Materials Science 139, 140–152 (2017) https://doi.org/10.1016/ j.commatsci.2017.07.030

    Mathew, K., Montoya, J.H., Faghaninia, A., Dwarakanath, S., Aykol, M., Tang, H., Chu, I.-h., Smidt, T., Bocklund, B., Horton, M., Dagdelen, J., Wood, B., Liu, Z.-K., Neaton, J., Ong, S.P., Persson, K., Jain, A.: Atomate: A high-level inter- face to generate, execute, and analy...

  56. [65]

    Computational Materials Science 68, 314–319 (2013) https://doi.org/10.1016/j.commatsci.2012

    Ong, S.P., Richards, W.D., Jain, A., Hautier, G., Kocher, M., Cholia, S., Gunter, D., Chevrier, V.L., Persson, K.A., Ceder, G.: Python materials genomics (pymat- gen): A robust, open-source python library for materials analysis. Computational Materials Science 68, 314–319 (201...

  57. [66]

    Concurrency and Computation: Practice and Experience 27(17), 5037–5059 (2015) https://doi.org/10.1002/cpe.3505 23 https://onlinelibrary.wiley.com/doi/pdf/10.1002/cpe.3505

    Jain, A., Ong, S.P., Chen, W., Medasani, B., Qu, X., Kocher, M., Brafman, M., Petretto, G., Rignanese, G.-M., Hautier, G., Gunter, D., Persson, K.A.: Fireworks: a dynamic workflow system designed for high- throughput applications. Concurrency and Computation: Practice and Expe...

  58. [67]

    https://arxiv.org/abs/2204.02782

    Gasteiger, J., Shuaibi, M., Sriram, A., G¨ unnemann, S., Ulissi, Z., Zitnick, C.L., Das, A.: GemNet-OC: Developing Graph Neural Networks for Large and Diverse Molecular Simulation Datasets (2022). https://arxiv.org/abs/2204.02782

  59. [68]

    Nature Methods 17, 261–272 (2020) https://doi.org/10.1038/ s41592-019-0686-2

    Virtanen, P., Gommers, R., Oliphant, T.E., Haberland, M., Reddy, T., Courna- peau, D., Burovski, E., Peterson, P., Weckesser, W., Bright, J., van der Walt, S.J., Brett, M., Wilson, J., Millman, K.J., Mayorov, N., Nelson, A.R.J., Jones, E., Kern, R., Larson, E., Carey, C.J., Po...

  60. [69]

    https://arxiv.org/abs/1109.2378

    M¨ ullner, D.: Modern hierarchical, agglomerative clustering algorithms (2011). https://arxiv.org/abs/1109.2378

  61. [70]

    Bar-Joseph, Z., Gifford, D.K., Jaakkola, T.S.: Fast optimal leaf ordering for hier- archical clustering. Bioinformatics 17(1), 22–29 (2001) https://doi.org/10.1093/ bioinformatics/17.suppl 1.S22 https://academic.oup.com/bioinformatics/article- pdf/17/suppl 1/S22/50522365/bioin...

Pith tools

Reviewed August 11, 2026 · model on record in the stance chip above.