Structure of Pre-Nucleation Structures (PNCs) of Pure Calcium Carbonate and with guest ions (Mg, Al, Si)

This dataset contains .json files of the analysis of PNCs structures (Calcium Carbonates with guest ionslike Mg, Al or Si) obtained with an evolutionnary algorithm. There is one file per composition studied. The .json files are compatible with the software Chemiscope [1] in order to vizualise and explore interactively the PNCs and their structural descriptors. The vizualisation is set in such way that the darker and bigger the point, the lower the recsaled enthalpy. The structures are ordered from the lowest rescaled enthalpy to the highest, in other words structure 0 has a rescaled enthalpy of 0 eV/atom and the final structure will have a rescaled enthalpy close but smaller than 0.14 eV/atom, the cutoff we applied to study the structures.Some figures are also part of the repository, to illustrate some paragraph. A python script is include too, reverse_chemiscope.py, in order to extract the structures from the .json files. There is also a technical explanation of the workflow we built.本数据集包含经进化算法（evolutionary algorithm）得到的PNCs结构（即掺杂Mg、Al或Si等客体离子的碳酸钙）的分析结果.json文件，每一种研究组分对应一个独立的.json文件。上述.json文件可与Chemiscope软件[1]兼容，用于交互式可视化与探索该类PNCs结构及其结构描述符。 可视化界面的设置规则为：点的颜色越深、尺寸越大，则其重整化焓（rescaled enthalpy）越低。所有结构按重整化焓由低到高排序，具体而言，第0号结构的重整化焓为0 eV/原子，末位结构的重整化焓接近且略小于0.14 eV/原子，该阈值为本次研究的结构筛选截断值。本数据集还附带部分配图，用于辅助对应段落的阐释。 此外本数据集还附带一份Python脚本reverse_chemiscope.py，用于从.json文件中提取目标结构。本数据集还包含针对我们所搭建的分析流程的技术说明文档。