Controlling and quantifying two-level systems, disorder and ideality in tetrahedrally bonded amorphous thin films

在凝聚态物理领域，缺乏结构秩序的材料——被认为是“无定形材料”或“玻璃”——与晶体相比，相对未被探索。晶体由空间上重复的原子组成，这使得数学上的简单形式可以用来计算和预测这些系统的性质。然而，非晶系统没有这种结构的可重复性，因此人们对其了解较少。然而，这种理解的缺乏并不排除无序系统的适用性或科学影响；塑料、硅酸盐玻璃和非晶硅光伏就是与日常生活、工业和技术相关的例子。非晶超导体是一个显著的例子，说明了基本的科学性质是如何超越结构缺陷的。无序材料的性质在很大程度上取决于材料是如何产生的，但如何描述不同方法产生的不同非晶结构并不清楚，即使是单一元素材料，也不清楚完全无序材料中缺陷的性质。很明显，无序存在于不同的长度和能量尺度上，从局部的原子大小的无序到更大的尺度。有趣的是，存在一种“理想玻璃”的概念，它在保持完全无序的同时，缺乏这种无序的缺陷，从而接近晶体的独特性，包括其性质的可重复性和可预测性。该项目将确定不同制备方法和不同原子产生的无序和缺陷类型之间的关系，以及无序材料“理想”的可调性。这样的确定将提高对具有技术和基础科学意义的无序材料的理解和控制。该项目还将教育和培训学生，并帮助增加科学的多样性参与；PI和研究生都是女性，并且积极参与努力，使STEM少数族裔和社会经济少数群体能够接触到物理学。技术摘要：非晶态或玻璃态的热力学性质是一个长期和当前都很有趣的问题。结构秩序的缺乏使这些系统在数学上难以处理，并使解决无序如何影响热力学性质成为一个挑战。能量景观的局部和全局广义最小值（kT）与构型熵有关。理想的玻璃具有较低的构型熵，接近晶体对应的构型熵，这意味着存在一种独特的无序状态，没有缺陷。更小尺度（kT）的局部极小值产生玻璃的异常低温特性，这些特性可以通过隧道或双能级系统（TLS）很好地描述，这被广泛认为是普遍的，尽管对于导致这些小尺度极小值的原因存在分歧。更有争议的是低温和高温热力学性质之间的联系（如果有的话）。近年来，发现了低温普遍行为的显著例外，表明存在不同类别的失序。一个相关的问题涉及非晶系统中缺陷的性质和可能的相互依赖性；例如，在非晶硅中，已知存在TLS和悬空键，并且依赖于原子密度，但不直接相关。这种（和其他）四面体结合材料与传统研究的玻璃有根本不同；它们不能从液态淬灭，它们的四面体键导致了一个过度约束的连续随机网络，并且难以为传统的量热法大量生产，这在以前阻碍了大多数热力学测量。有趣的是，这些恰恰是最容易通过气相沉积工艺制成薄膜的材料。将采用各种生长技术制备四面体键合材料薄膜，以研究TLS与密度/结构和无序理想性之间的联系；该项目将测试理想眼镜没有TLS或缺陷的假设。使用独特的基于膜的纳米热计，将在0.1-1000K的宽温度范围内测量热容量和热力学性质。该温度范围涵盖了低温度（1K）下的TLS和非晶硅的高温度玻璃化转变温度。这些量热计极快的加热（10^5 K/秒）和冷却（10^4 K/秒）速率和宽的温度范围使这些以前不可能的实验成为可能，包括确定零度构型熵。

Non-technical AbstractIn the field of Condensed Matter Physics, materials that lack structural order -- deemed "amorphous materials" or "glasses" -- are, compared to crystals, relatively unexplored. Crystals consist of spatially repeated atoms, which permit mathematically simple formalisms that can be used to calculate and predict properties of these systems. Amorphous systems, however, have no such structural repeatability, and thus are less understood. This lack of understanding, however, does not preclude the applicability or scientific impact of disordered systems; plastics, silicate glasses, and amorphous silicon photovoltaics are examples that are pertinent to daily life, industry, and technologies. Amorphous superconductors are a remarkable example of how a fundamental scientific property transcends structural imperfection. The properties of a disordered material depend strongly on how the material was produced, but it is not clear how to describe the different amorphous structures produced by different methods, even for a single element material, nor what the nature of a defect is in a fully disordered material. It is clear that disorder exists on different length and energy scales, ranging from local, atomic-sized disorder to larger scales. Intriguingly, there exists the notion of an "ideal glass", which while remaining thoroughly disordered, lacks imperfections in that disorder and thus approaches the uniqueness of a crystal, including reproducibility and predictability of its properties. The project will determine the relationship between types of disorder and defects produced by different preparation methods and for different atoms, and the tunability of the "ideality" of disordered materials. Such a determination will yield improved understanding and control of disordered materials of technological and fundamental scientific significance. The project will also educate and train students and help to increase diversity participation in science; the PI and graduate student are women, and actively engage in efforts to make physics accessible to underrepresented STEM ethnic and socioeconomic minorities. Technical AbstractA problem of both longstanding and current interest is the thermodynamic nature of the amorphous or glassy state. The lack of structural order makes these systems less mathematically tractable and makes it a challenge to resolve how disorder affects the thermodynamic properties. Local and global minima on a broad scale (kT) in the energy landscape are relevant to the configurational entropy. An ideal glass has low configurational entropy, approaching that of the crystalline counterpart, thus implying the existence of a unique disordered state that lacks defects. Local minima on a much smaller scale (kT) produce anomalous low temperature properties of glasses that are well described by tunneling or two level systems (TLS), which are widely considered universal although disagreement exists as to what causes these small scale minima. Even more controversial is the connection (if any) between the low and high temperature thermodynamic properties. In recent years, significant exceptions to low temperature universal behavior have been found, suggesting that different classes of disorder exist. A related question concerns the nature and possible interdependence of defects in an amorphous system; e.g. in amorphous silicon, both TLS and dangling bonds are known to exist, and are dependent on atomic density, but are not directly correlated. This (and other) tetrahedrally-bonded materials are fundamentally different than the traditionally studied glasses; they cannot be quenched from the liquid state, their tetrahedral bonding leads to an overconstrained continuous random network, and the difficulty in producing large quantities for conventional calorimetry has previously prevented most thermodynamic measurements. Intriguingly, these are precisely the materials most easily made as thin films by vapor deposition processes. Various growth techniques will be used to produce thin films of tetrahedrally bonded materials to study the link between TLS and density/structure and ideality within disorder; the project will test the hypothesis that ideal glasses do not have TLS or defects. Using unique membrane-based nanocalorimeters, heat capacity and thermodynamic properties will be measured over a wide temperature range, 0.1-1000K. This temperature range covers the TLS at low T (1K) and the proposed high T glass transition temperature for amorphous silicon. The enormously fast heating (10^5 K/sec) and cooling (10^4 K/sec) rates of these calorimeters and wide temperature range permit these previously impossible experiments, including determination of zero temperature configurational entropy.