Data for "Elementary Decomposition Mechanisms of Lithium Hexafluorophosphate in Battery Electrolytes and Interphases"

Contained here is a JavaScript Object Notation (JSON)-formatted file called pfx_named_data.json. This file contains the structures (as serialized Pymatgen Molecule objects) and thermochemical properties of the reaction endpoints and TS reported in the article "Elementary Decomposition Mechanisms of Lithium Hexafluorophosphate in Battery Electrolytes and Interphases" (see DOI:10.26434/chemrxiv-2022-4bd1p-v2). <br> The key for each key-value pair in pfx_named_data.json is the name of the species as reported in the main text or the Supporting Information. For instance, the data for TS11 would be found under the key "TS11". For reactions where species, namely LiF, HF, and CO2, are removed, two entries for the relevant endpoint are provided. The species with LiF, HF, and/or CO2 present are named "M<em>n"</em>, where <em>n</em> is the appropriate index; the species with the species removed are named "M<em>n</em>-<em>x"</em>, where <em>x</em> is the species that is removed. Where multiple species are removed, the name takes the form "M<em>n</em>-<em>x</em>-<em>y</em>", where <em>x</em> and <em>y</em> are the species removed. <br> All structures were optimized in Jaguar using the range-separated hybrid generalized gradient approximation (GGA) density functional ωB97X-D, def2-SVPD basis set, and Conductor-like Screening Model (COSMO) implementation of the polarizable continuum model (PCM) with water as the solvent. In Jaguar, all basis functions representing f and higher orbitals were removed to further reduce cost, making the basis more precisely def2-SVPD(-f). All TS were confirmed to have one imaginary frequency and to connect to the expected endpoints. The electronic energies of all TS and reaction endpoints (reactants and products) were corrected with single-point energy evaluations in Jaguar using range-separated hybrid meta-GGA functional ωB97M-V with the def2-TZVPD basis set in COSMO. <br> To load this data in Python, use monty (https://github.com/materialsvirtuallab/monty): <br> from monty.serialization import loadfn data = loadfn("pfx_named_data.json")

本数据集为一份名为pfx_named_data.json的JavaScript对象表示法（JSON）格式文件。该文件收录了论文《电池电解质与界面中六氟磷酸锂的基本分解机理》（DOI:10.26434/chemrxiv-2022-4bd1p-v2）中报道的反应终点与过渡态（TS）的结构（以序列化的Pymatgen分子对象形式存储）及热化学性质。<br>pfx_named_data.json中各键值对的键名与正文或补充材料中报道的物种名称完全一致。例如，TS11的相关数据可通过键名"TS11"获取。对于涉及移除LiF、HF与CO₂这类物种的反应，我们为相关反应终点提供了两组条目：保留LiF、HF和/或CO₂的物种命名为"M<em>n</em>"，其中<em>n</em>为对应索引；移除上述物种后的物种命名为"M<em>n</em>-<em>x</em>"，其中<em>x</em>为被移除的物种；若需移除多个物种，则命名格式为"M<em>n</em>-<em>x</em>-<em>y</em>"，<em>x</em>与<em>y</em>分别代表被移除的物种。<br>所有结构均在Jaguar软件中完成几何优化，采用范围分离杂化广义梯度近似（generalized gradient approximation, GGA）密度泛函ωB97X-D、def2-SVPD基组，并使用类导体屏蔽模型（Conductor-like Screening Model, COSMO）实现的可极化连续介质模型（polarizable continuum model, PCM），以水作为溶剂。为进一步降低计算成本，Jaguar软件中移除了所有代表f轨道及更高轨道的基函数，使基组更精准地匹配def2-SVPD(-f)。所有过渡态均经振动频率分析确认仅含一个虚频，且可连接至预期的反应终点。所有过渡态与反应终点（反应物与产物）的电子能量，均通过Jaguar软件进行单点能计算校正，计算采用范围分离杂化meta-GGA泛函ωB97M-V与def2-TZVPD基组，并在COSMO模型下完成。<br>若需在Python环境中加载该数据集，可使用monty库（https://github.com/materialsvirtuallab/monty），示例代码如下：<br>from monty.serialization import loadfn<br>data = loadfn("pfx_named_data.json")