Advancing solid-state band gap predictions
Advancing solid-state band gap predictions
复制标题
推进固态带隙预测
DOI：
10.1073/pnas.2113648118
复制
发表时间：
2021
期刊：
影响因子：
--
通讯作者：
Scuseria, Gustavo E.
中科院分区：
文献类型：
--
作者：
Scuseria, Gustavo E.
也许材料最重要的特性是它的基本带隙。这是系统中增加和减少一个电子之间的能量差。它将金属与绝缘体区分开来，并向我们提供有关材料对外部影响的电子响应的信息。这在电池、半导体、合金、电子设备和光伏材料等众多技术应用中至关重要。多年来，利用量子模拟进行带隙预测已经取得了稳步进展，但缺乏一种适用于具有不同带隙的广谱材料的准确方法。Wing等人(1)在第一性原理预测基本带隙方面迈出了决定性的一步，其测量精度可与实验误差条相媲美，这一壮举建立在过去几十年多个研究小组的贡献之上。多年来，计算材料建模的主力一直是密度泛函理论（DFT），特别是对于量子效应占优势的带隙等特性。DFT是一个相对简单的模型，它的根先于量子力学的表述(2)。它本质上把电子描述为独立的粒子，通过电子密度的交换相关能泛函得到的有效势来相互作用。有点矛盾的是，这个函数的存在很容易被证明，但它的实际实现却比最初设想的要具有挑战性得多。几十年来，研究人员系统地提高了DFT的准确性，使其成为化学、物理和材料科学的宝贵辅助手段。然而，在广泛的固态材料中，基本带隙仍然难以高精度预测。最适合计算带隙的DFT形式被称为广义Kohn-Sham理论(3)，在这个模型中，泛函取决于电子可以占据的所有轨道，而不仅仅是产生电子密度的组合。简而言之，Wing等人(1)提出的方法建立在两个现有支柱之上：1)混合泛函数(4)，混合了一部分半局部DFT交换（5)和一部分非局部精确Hartree-Fock （HF）交换，以及2）在短距离和长距离之间分离电子间库仑排斥(6)。Wing等人(1)添加的关键成分是一种巧妙的ansatz(7)，用于在采用局部轨道框架(9)时确定固体(8)中最佳调谐的筛选参数。
Perhaps the most important property of a material is its fundamental band gap. This is the energy difference between adding and subtracting one electron from a system. It distinguishes metals from insulators and gives us information about the electronic response of the material to external influences. This is crucial in myriad technological applications like batteries, semiconductors, alloys, electronic devices, and photovoltaic materials, to name a few. Band gap predictions employing quantum simulations have steadily progressed over the years, but an accurate method valid across the broad spectrum of materials with diverse band gaps was lacking. Wing et al.(1) take a decisive step forward in the firstprinciples prediction of fundamental band gaps with accuracy rivaling experimental error bars in their measurement, a feat that builds upon the contributions of multiple research groups during the last few decades. For many years, the workhorse of computational materials modeling has been density functional theory (DFT), particularly for properties like band gaps where quantum effects are preponderant. DFT is a relatively simple model whose roots precede the formulation of quantum mechanics (2). It essentially describes electrons as independent particles interacting via an effective potential obtainable from an exchange-correlation energy functional of the electron density. Somewhat paradoxically, the existence of this functional can be easily proven, but its practical realization has turned out to be much more challenging than perhaps originally envisioned. For several decades, researchers have systematically improved the accuracy of DFT, making it a valuable aid for chemistry, physics, and materials science. However, across the broad range of solid-state materials, the fundamental band gap has remained stubbornly difficult to predict with high accuracy. The form of DFT most appropriate to calculate band gaps is known as generalized Kohn–Sham theory (3), a model where the functional depends on all the orbitals that electrons can occupy, not just the combination that yields the electron density. Succinctly explained, the method proposed by Wing et al.(1) builds upon two existing pillars: 1) hybrid functionals (4) that mix a portion of semilocal DFT exchange (5) with a portion of nonlocal exact Hartree–Fock (HF) exchange, and 2) separation of the interelectronic Coulomb repulsion between short range and long range (6). The key ingredient added by Wing et al.(1) is an ingenious ansatz (7) for determining the optimally tuned screening parameter in a solid (8) when adopting a localized orbital framework (9).