Atomic layer deposition (ALD) of complex oxide thin films for integration into semiconductor devices: Precursor engineering and process optimization

随着半导体器件尺寸的不断减小，在新型集成结构中，硅氧化物基层与高钾复杂氧化物层之间的交换成为可能。复合氧化物将在存储器和逻辑器件的不同领域取代二氧化硅层。以动态随机存取存储器（dram）为例，关键是在横向尺寸减小的情况下保持25 fF的恒定电池电容。这些技术挑战可以通过原子层沉积（ALD）来实现，这些极薄（几十纳米）的高k薄膜具有均匀的厚度和均匀的化学计量，形成具有高纵横比的三维（3D）结构。复杂氧化物薄膜的其他应用需要低沉积温度和ALD的均匀性，包括替代栅极氧化物，碳纳米管（CNTs）涂层，压电应用的铁电薄膜，以及预计将成为新的通用存储器的电阻开关元件（rram）中的绝缘层。在这个项目中，我们的目标是评估用于沉积复杂氧化物薄膜的定制ALD工艺，这应该可以通过专门设计ALD前体的专门工程来实现。深入了解优化ALD前驱体特征的关键因素，多组分氧化ALD工艺中前驱体的相互关系，以及具有相同ALD窗口的不同元素前驱体的设计是使ALD成为复杂氧化物标准沉积技术的主要挑战。该项目将重点研究两类作为第三代电介质的复合氧化物：1)高钾碱土钛酸盐、锆酸盐和半酸盐（如（Ba,Sr）Ti03、Ba(Ti,Zr)03和Ba(Ti,Hf)03)，以及2）通过加入稀土元素来稳定4族金属氧化物的特殊结构（即高钾）相（具体例子为Hf02:La203、Y203:Zr02）。ALD的主要挑战之一是选择和可用性合适的前驱体，这些前驱体在气相中具有适当的挥发性，稳定性，但在底物和生长表面上具有反应性。这些标准与标准金属有机化学气相沉积（MOCVD）有很大的不同。因此，本项目的目标之一是开发具有合适配体设计或修饰的高钾氧化物ALD前体。新型前驱体将主要利用金属酰胺与胍、脒、丙二酸、环戊二烯及其衍生物等螯合配体结合开发。前驱体将在ALD和液体注射(LI)- ALD工艺中进行测试和表征，并根据ALD窗口和生长速率优化前驱体组合，例如，将用于在最先进的ALD反应器中生长多组分氧化膜。一个专门的目标是通过选择前驱体和沉积条件来控制薄膜的微观结构，即非晶和多晶或立方和单斜。随着当前的技术和新兴的纳米技术朝着采用ALD等纳米级薄膜生长技术的方向发展，本研究项目的成功需要一个多学科的方法。双方强大的科学背景在前驱体开发和薄膜器件结构制造方面是互补的。这将为实现与现代技术的新材料和工艺开发直接相关的拟议项目铺平道路。

The ever decreasing size of semiconductor devices enforces the exchange of silicon oxide based layers by higher-k: complex oxide layers in novel integrated structures. Complex oxides are going to replace Si02 layers in different areas of memory as well as logic devices. Considering dynamic random access memory devices (DRAMs) as an example the key point is to sustain a constant cell capacitance of 25 fF at even decreased lateral dimensions. These technological challenges can be achieved by the atomic layer deposition (ALD) of extremely thin (few tens of a nanometer) higher-k films with uniform thickness and homogeneous stoichiometry into three dimensional (3D) structures with high aspect ratio. Other applications of complex oxide thin films which require the low deposition temperature and the homogeneity of ALD cover the fields of alternative gate oxides, coatings on carbon-nanotubes (CNTs), ferroelectric thin films for piezoelectric applications, and insulating layers in resistive switching elements (RRAMs) which are projected to become the new universal memory. In this project we aim to evaluate tailored ALD processes for the deposition of complex oxide thin films which should be possible by a dedicated engineering of specially designed ALD precursors. The deeper understanding of the key-factors which characterize optimized ALD precursors, the interrelafion of the precursors in a multi-component oxide ALD process, and the design of different element precursors with the same ALD window are major challenges for qualifying ALD as a standard deposition technique for complex oxides. The project will concentrate on two classes of complex oxides which are explored as third generation dielectrics: 1) higher-k alkaline earth titanates, zirconates, and hafnates (e.g. (Ba,Sr)Ti03, Ba(Ti,Zr)03, and Ba(Ti,Hf)03), and 2) stabilization of special structural (i.e. higher-k) phases of group 4 metal oxides by incorporation of rare earth elements (specific examples Hf02:La203, Y203:Zr02). One of the main challenges of ALD is the choice and the availability of suitable precursors with appropriate properties in terms of being volatile, stable in the gas-phase, but reactive on the substrate and growing surface. These are quite different criteria to those of standard metal organic chemical vapour deposition (MOCVD). Therefore one objective of this project is to develop precursors for ALD of higher-k oxides with suitable ligand design or modification. Novel precursors will be developed mainly using metal amides in combination with chelating ligands like guanidinates, amidinates, malonates, cyclopentadienyls and their respective derivatives. The precursors will be tested and characterised in ALD and liquid injection (LI)- ALD processes, and optimized precursor combinations with respect to ALD window and growth rate for example, will be used to grow multi-component oxide films in state-of-the-art ALD reactors. One dedicated aim is to control the films microstructure i.e. amorphous and polycrystalline or cubic and monoclinic by choice of precursors and deposition conditions. As the current technologies and the emerging nanotechnologies move towards adopting the nano-scale thin film growth techniques like ALD, a multidisciplinary approach is required for the success of this research project. The strong scientific backgrounds of both the partners are complementary to each other in terms of precursor development and fabrication of thin film device structures. This should pave the way for realisation of the proposed project which is directly relevant for development of new materials and processes for modern day technology.