Structural database of small PtAu and hydrogenated PtAu nanoparticles

This is a database of PtAu and hydrogenated PtAu:H nanoparticles generated with a GAP interatomic potential for PtAu:H [1]. The files are provided in ASE's extended XYZ format. The database contains the following entries:

lowest_PtAu.xyz: 11 nanoparticles from 0% to 100% Au content at 10% intervals. These are the lowest in energy for each composition among a significantly larger database of candidates (with 770 entries). They were generated with a "cooking" protocol [2].

hydrogenated_ordered.xyz: 880 PtAu:H nanoparticles generating by adding hydrogen to the 11 "bare" nanoparticles in "lowest_PtAu.xyz". The hydrogen was added using a heuristic Markov-chain Monte Carlo approach with a defined chemical potential for hydrogen. For each chemical potential (8 values within the range -3 eV to -2.1 eV), 10 configurations were run per starting bare nanoparticle, with 2000 steps of the heuristic algorithm (which includes full geometric optimization at each step). The attached structures are the last snapshot of each Monte Carlo trajectory, and represent the expected hydrogen coverage at the given chemical potential. The individual entries within the XYZ file have tags with self-explanatory names for the chemical potential ("mu") and the random initialization ("i_rand").

The reference paper with more details will be added here once it is available as a preprint or published in a journal.

Financial support from the Research Council of Finland and computational resources from CSC (the Finnish IT Center for Science) and Aalto University's Science-IT project are gratefully acknowledged.

References

J. Kloppenburg and M.A. Caro. GAP interatomic potential for PtAu:H nanoparticle simulation. DOI: 10.5281/zenodo.11184039.

J. Kloppenburg, L. B. Pártay, H. Jónsson, and M. A. Caro. "A general-purpose machine learning Pt interatomic potential for an accurate description of bulk, surfaces and nanoparticles". J. Chem. Phys. 158, 134704 (2023).

本数据集为使用GAP原子间势（Gaussian Approximation Potential）针对PtAu:H体系生成的PtAu及氢化PtAu:H纳米粒子数据库[1]。所有文件以原子模拟环境（Atomic Simulation Environment，ASE）的扩展XYZ格式存储。本数据库包含以下条目：

lowest_PtAu.xyz：包含11个纳米粒子，金（Au）含量覆盖0%至100%，间隔为10%。在包含770条候选条目的大规模数据库中，这些粒子为对应组分下能量最低的结构。它们通过「cooking」制备流程生成[2]。

hydrogenated_ordered.xyz：通过向lowest_PtAu.xyz中的11个「裸」纳米粒子引入氢原子，共生成880个PtAu:H纳米粒子。氢原子的引入采用启发式马尔可夫链蒙特卡洛方法，并设定了明确的氢化学势。对于8个处于-3 eV至-2.1 eV区间的化学势值，每个初始裸纳米粒子对应运行10组构型，每组包含2000步启发式算法迭代（每步均包含完整的几何优化）。本文件收录的结构为每条蒙特卡洛轨迹的最终快照，代表对应化学势下的预期氢覆盖率。XYZ文件中的各条目带有含义明确的标签，分别对应氢化学势（「mu」）与随机初始化编号（「i_rand」）。

有关更多细节的研究论文将在其以预印本形式发布或正式刊发于期刊后补充至此处。

衷心感谢芬兰研究理事会（Research Council of Finland）提供的经费支持，以及芬兰IT科学中心（CSC）与阿尔托大学科学信息技术项目（Aalto University's Science-IT project）提供的计算资源。

参考文献
[1] J. Kloppenburg 与 M.A. Caro. 用于PtAu:H纳米粒子模拟的GAP原子间势. DOI: 10.5281/zenodo.11184039.
[2] J. Kloppenburg, L. B. Pártay, H. Jónsson, 与 M. A. Caro. 用于精准描述块体、表面与纳米粒子的通用机器学习Pt原子间势. 《化学物理学报》158, 134704 (2023).