论文标题
零点晶格扩展和带隙重归其化:Grüneisen方法与自由能最小化
Zero-point lattice expansion and band gap renormalization: Grüneisen approach versus free energy minimization
论文作者
论文摘要
与静态晶格图片相比,零点晶格膨胀(ZPLE)是材料中存在声子引起的晶格参数的一个小变化。它对带隙能的零点重归其化(ZPR)显着贡献,但是其后果并未像由电子 - 波相互作用所呈现的那样彻底研究。在通常的第一原理方法中,一个人通过最大程度地减少$ t = 0 $ k Helmholtz自由能来评估ZPLE。在这项工作中,我们表明,通常忽略零点效应的Grüneisen参数的形式主义,可以有效地用于计算各向同性和各向异性材料的ZPLE,计算成本要低得多。我们会系统地测试这种形式主义对22立方体和葡萄菌材料的测试,并获得了极好的一致性,可以为ZPLE和由此带来的带隙ZPR的自由能最小化结果。我们使用结果来验证估计ZPLE诱导的ZPR的经验表达,并公布其对从实验数据估算ZPLE涉及的温度范围的敏感性。我们的发现最终表明,ZPLE对带隙ZPR的贡献可以达到20%,至80%以上的电子 - phonon相互作用贡献较重或更多的离子材料,包括含有光原子的材料。因此,如果尝试将理论ZPR结果与实验数据进行比较,则必须考虑在平等基础上的两种贡献。
The zero-point lattice expansion (ZPLE) is a small variation of the lattice parameters induced by the presence of phonons in a material compared to the static lattice picture. It contributes significantly to the zero-point renormalization (ZPR) of the band gap energy, but its consequences have not been investigated as thoroughly as those stemming from electron-phonon interactions. In the usual first-principles approach, one evaluates the ZPLE by minimizing the $T=0$ K Helmholtz free energy. In this work, we show that the formalism based on the Grüneisen parameters, which commonly neglects zero-point effects, can be efficiently used to compute ZPLE for both isotropic and anisotropic materials at much lower computational cost. We systematically test this formalism on 22 cubic and wurtzite materials and obtain excellent agreement with free energy minimization results for both the ZPLE and the resulting band gap ZPR. We use our results to validate an empirical expression estimating the ZPLE-induced ZPR and unveil its sensitivity to the temperature range involved in estimating the ZPLE from experimental data. Our findings finally reveal that the ZPLE contribution to the band gap ZPR can reach 20% to more than 80% of the electron-phonon interaction contribution for heavier or more ionic materials, including materials containing light atoms. Considering both contributions on an equal footing is thus essential should one attempt to compare theoretical ZPR results with experimental data.