Impact of structural modifications on dielectric properties of epitaxial rare-earth oxides on silicon

实现先进的未来器件的一种非常有前途的方法是使用单晶，紧密晶格匹配的氧化物，它将沉积在所选的衬底上。二元晶态稀土氧化物已被证明是一组非常有前途的外延生长材料。在某些条件下，薄的稀土氧化物层外延生长在Si上表现出介电常数是远远大于已知的体值。这种增强效应的原因还没有完全弄清楚。在正在进行的项目中，我们研究了晶格失配引起的应变对介电性能的影响。我们可以表明，立方晶格的四角畸变不足以解释K的增强。因此，我们提出了更严重的应变诱导结构相变形成单斜相。在该项目的执行过程中，我们将更详细地研究这个阶段转换。第一个结果表明，在Si上的Gd 2 O3（在拉伸应力下）和Nd 2 O3（在压缩应力下）表现出几乎相同的介电常数的增强。通过在Ge和Si衬底上生长Nd 2 O3，我们将得到一个具有相同应力量（在两种情况下晶格失配相等）但符号相反（在Ge上拉伸但在Si上压缩）的模型系统。 对于这两种情况下，具有不同厚度的层的K增强的调查应该提供这种相变的应力依赖性的详细图片。初步的研究表明，Gd 2 O3层生长在Si上，在低温下导致四角畸变的立方晶格。温度可能不足以进行结构相变。我们希望通过在不同温度下生长和评估薄Gd 2 O3来了解这种相变的动力学。将研究四角畸变层的互补生长后退火。我们的研究表明，N掺杂在外延的Gd 2 O3薄膜可以有助于改善其介电性能。的漏电流密度的大幅减少和消失的电容-电压特性中观察到的Gd 2 O3：N层的滞后表明，在Gd 2 O3中的氮掺入有效地消除了氧空位引起的氧化物层中的缺陷的不利影响。然而，需要考虑的是，N的掺入也可以通过改变其电子结构来显著影响高K氧化物的电学性质。为了通过N掺杂来优化外延稀土氧化物的电性能，因此必须研究它们的电子结构作为N浓度的函数的演变。作为我们的工作的N掺杂外延Gd 2 O3层的延伸，我们需要调查的影响，N掺入的Gd 2 O3：N层的电子结构，这是必不可少的，以优化其电气性能。

A very promising way to realize advanced future devices is using single-crystalline, closely lattice matched oxides, which will be deposited on the substrate of choice. Binary crystalline rare earth oxides have proved to be a very promising group of dielectrics for epitaxial growth. Under certain conditions, thin rare earth oxide layers epitaxially grown on Si exhibit dielectric constants that are much larger than the known bulk values. The reason for that enhancement effect is not fully understood yet. In the on-going project, we investigate the effect of lattice-mismatch induced strain on dielectric properties. We could show that tetragonal distortion of the cubic lattice is not sufficient to explain the enhancement in K. Therefore, we propose more severe strain induced structural phase deformations into a monoclinic phase. In the prosecution of that project, we will investigate this phase transition in more detail. First results show that Gd2O3 (under tensile stress) and Nd2O3 (under compressive stress) on Si exhibit nearly the same enhancement in the dielectric constant. By growing Nd2O3 on Ge and Si substrates, we will have a model system with identical amount of stress (the lattice mismatch is equal in both cases) but opposite sign (tensile on Ge but compressive on Si). Investigation of the K enhancement for layers with varying thicknesses for both cases should deliver a detailed picture of the stress dependence of this phase transition. Preliminary investigations reveal that Gd2O3 layer growth on Si at low temperatures leads to tetragonal distorted cubic lattices. The temperature was probably not sufficient for a structural phase transition. We expect to understand the dynamics of such a phase transition by growing and evaluating thin Gd2O3 at different temperatures. Complimentary, post-growth anneals of tetragonal distorted layers will be investigated. Our investigations revealed that N doping in epitaxial Gd2O3 thin films can be instrumental in improving their dielectric properties. Substantial reduction of the leakage current density and disappearance of hysteresis in capacitance-voltage characteristics observed in the Gd2O3:N layers indicate that nitrogen incorporation in Gd2O3 effectively eliminates the adverse effects of the oxygen vacancy induced defects in the oxide layer. However, one needs to consider that N incorporation can also dramatically influence the electrical properties of high-K oxides by changing their electronic structures. In order to optimize the electrical properties of the epi-rare earth oxides by means of N doping, it is therefore imperative to investigate the evolution of their electronic structure as a function of N concentration. As an extension to our work on N doped epitaxial Gd2O3 layers we need to investigate the effect of N incorporation on the electronic structure of Gd2O3:N layers which is essential to optimize their electrical properties.