Fig. 1 An overview of the steps involved in identifying the adsorption energy for an adsorbate-surface combination.?

計算催化具有篩選大量材料的潛力，能夠與耗時且昂貴的實驗研究形成互補。使用第一性原理方法尋找異相催化劑的一項關鍵任務是吸附能的計算。吸附能是通過計算吸附質–表面所有可能構型中的最小值來確定的，它是計算自由能圖并確定催化劑表面最佳反應途徑的起點。

傳統上，確定全局最優的吸附質–表面構型依賴于啟發式方法和研究者的直覺。隨著高通量篩選需求的日益增加，僅僅使用啟發式和直覺變得極具挑戰。

來自美國加州人工智能基礎研究所的Janice Lan等，證明了使用機器學習勢能夠更加準確、高效地識別低能吸附質–表面構型。他們使用啟發式和隨機的策略對大量可能的吸附質構型進行采樣，并使用機器學習勢對結構進行弛豫。對于最佳的k個能量，作者進一步使用單點密度泛函理論（DFT）或完全DFT弛豫以改善計算結果。通過這種方法，可以在精度和效率之間取得適當的權衡，其中一種選擇能夠在87.36%的時間內找到能量最低的構型，同時在計算上實現約2000倍的加速。

Fig. 4 Illustration of the lowest energy configurations as found by?DFT-Heur+Rand, SchNet, GemNet-OC, and SCN-MD-Large on the?OC20-Dense validation set.

在作者之前的工作中，使用機器學習模型加速尋找低能吸附質–表面構型的過程曾一度依賴于針對特定吸附質/催化劑組合的定制模型，這限制了更加廣泛的應用。作者現在的工作所使用的可泛化機器學習勢，有望大大擴展以往方法的多功能性，并同時繼續降低人力和計算成本。該文近期發布于npj Computational Materials 9: 172 (2023).

Fig. 5 ML+SP success rate at k = 5 across the different subsplits?of the OC20-Dense test set and several baseline models.?

Editorial Summary

The design of novel heterogeneous catalysts plays an essential role in the synthesis of everyday fuels and chemicals. To accommodate the growing demand for energy while combating climate change, efficient, low-cost catalysts are critical to the utilization of renewable energy. Given the enormity of the material design space, efficient screening methods are highly sought after. Computational catalysis offers the potential to screen vast numbers of materials to complement more time- and cost-intensive experimental studies. A critical task for first-principles approaches to heterogeneous catalyst discovery is the calculation of adsorption energies. The adsorption energy is the global minimum energy across all potential adsorbate placements and configurations, and is the starting point for the calculation of the free energy diagrams to determine the most favorable reaction pathways on a catalyst surface. Traditionally, the identification of the globally optimal adsorbate-surface configuration relies on heuristic methods and researcher intuition. As the desire to perform high-throughput screening increases, it becomes challenging to use heuristics and intuition alone.?

Janice Lan et al. from the Fundamental AI Research, CA, USA, demonstrated that machine learning (ML) potentials can be leveraged to identify low-energy adsorbate-surface configurations more accurately and efficiently. The authors sampled a large number of potential adsorbate configurations using both heuristic and random strategies and perform relaxations using ML potentials. The best k-relaxed energies can then be refined using single-point density functional theory (DFT) calculations or with full DFT relaxations. Using this approach, the appropriate trade-offs can be made between accuracy and efficiency, with one balanced option finding the lowest energy configuration 87.36% of the time, while achieving a ~2000× speedup in computation. Prior attempts to use machine learning models to accelerate the search for low-energy adsorbate-surface configurations have typically relied on bespoke models for each adsorbate/catalyst combination, which limits broader applicability. In this work, the generalizable machine learning potentials are promising to greatly expand the versatility of these methods while continuing to reduce the human and computational cost. This article was recently published in npj Computational Materials 9: 172 (2023).

原文Abstract及其翻譯

Janice Lan,Aini Palizhati,?Muhammed Shuaibi,?Brandon M. Wood,?Brook Wander,?Abhishek Das,?Matt Uyttendaele,?C. Lawrence Zitnick?&?Zachary W. Ulissi?

Abstract

Computational catalysis is playing an increasingly significant role in the design of catalysts across a wide range of applications. A common task for many computational methods is the need to accurately compute the adsorption energy for an adsorbate and a catalyst surface of interest. Traditionally, the identification of low-energy adsorbate-surface configurations relies on heuristic methods and researcher intuition. As the desire to perform high-throughput screening increases, it becomes challenging to use heuristics and intuition alone. In this paper, we demonstrate machine learning potentials can be leveraged to identify low-energy adsorbate-surface configurations more accurately and efficiently. Our algorithm provides a spectrum of trade-offs between accuracy and efficiency, with one balanced option finding the lowest energy configuration 87.36% of the time, while achieving a ~2000× speedup in computation. To standardize benchmarking, we introduce the Open Catalyst Dense dataset containing nearly 1000 diverse surfaces and ~100,000 unique configurations.

摘要

計算催化在廣泛應用的催化劑設計中發揮著越來越重要的作用。許多計算方法的一項共同任務是需要精確計算吸附質在催化劑表面的吸附能。傳統上，低能吸附質–表面構型的確定依賴于啟發式方法和研究者的直覺。隨著高通量篩選需求的日益增加，僅僅使用啟發式和直覺變得極具挑戰。在本文中，我們證明了使用機器學習勢能夠更加準確、高效地識別低能吸附質–表面構型。我們的算法提供了精度和效率之間的權衡范圍，其中一種選擇能夠在87.36%的時間內找到能量最低的構型，同時在計算上實現約2000倍的加速。為了標準化基準測試，我們引入了開源催化劑密集型（Open Catalyst Dense）數據集，其中包含近1000個不同的表面和約100000萬種不同的構型。