GOALI: Strained Layer Heterostructures for GaN-on-Si Epitaxy

目标：用于 GaN-on-Si 外延的应变层异质结构

• 批准号: 1410765
• 负责人: Joan Redwing
• 金额: $ 39万
• 依托单位: Pennsylvania State Univ University Park
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2014
• 资助国家: 美国
• 起止时间: 2014-07-01 至 2018-06-30
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=1410765&HistoricalAwards=false
• 关键词: GOALI; Strained; Layer; Heterostructures; GaN

Non-technical Description: Gallium nitride is an important semiconductor due to its wide ranging applications in solid-state lighting and power electronics. Pennsylvania State University and Efficient Power Conversion (EPC) Corporation are partnering in this GOALI project to develop materials growth technology to integrate gallium nitride thin films with silicon wafers, which are the standard substrates used for integrated circuit fabrication. The ability to directly combine these two materials would allow flexibility in circuit design for a variety of applications. However, the hexagonal crystal structure of gallium nitride is incompatible with the cubic crystal structure of (100) silicon. This project seeks to overcome these limitations by intentionally introducing strain into the silicon to alter the surface structure and thereby reduce the crystal mismatch. Real-time wafer curvature measurements are used to directly measure strain in the silicon substrate and study dynamic changes in film stress during gallium nitride deposition. Post-growth structural characterization is used to elucidate mechanisms of stress generation and relaxation. The project supports the Ph.D. thesis work of two graduate students at Penn State University who are exposed to issues relevant to process scale-up and manufacturing in a global semiconductor foundry through the collaboration with EPC. Undergraduates and high-school students from underrepresented groups and economically challenged regions of Pennsylvania are participating in the research through summer programs at Penn State University. Technical Description: This GOALI project is on the epitaxial growth of AlGaN/GaN heterostructures on on-axis (100) Si substrates for hybrid power electronics circuits. Strained Si/SiGe virtual substrates are being explored as a route to control the dominant reconstruction of the (100) Si surface and thereby reduce the in-plane rotational misalignment of AlN nuclei. In-situ wafer curvature measurements are used to monitor stress relaxation in the SiGe layers and measure tensile strain in the Si layer at the growth temperature and study its impact on GaN/AlN heteroepitaxy. The effects of n-type dopant chemistry on stress evolution in GaN are also being examined. The project seeks a fundamental understanding of the role of Si surface strain on GaN/AlN epitaxy, which can then be exploited to develop AlGaN/GaN two-dimensional electron gas heterostructures on (100) Si for power devices. The project team has complementary expertise in group III-nitride heteroepitaxy on Si substrates (Penn State University) and process scale-up and commercialization of GaN-on-Si for power field-effect transistors (EPC). In addition, EPC is well positioned to capitalize on hybrid circuit designs that will be enabled by the GaN on (100) Si growth technology developed in this project.

非技术描述：氮化镓是一种重要的半导体，在固态照明和电力电子中有着广泛的应用。宾夕法尼亚州立大学和高效功率转换（EPC）公司在这个GOALI项目中合作开发材料生长技术，将氮化镓薄膜与硅晶片集成在一起，硅晶片是集成电路制造的标准衬底。直接结合这两种材料的能力将为各种应用的电路设计提供灵活性。然而，氮化镓的六方晶体结构与（100）硅的立方晶体结构不相容。该项目旨在通过有意地在硅中引入应变来改变表面结构，从而减少晶体不匹配，从而克服这些限制。利用实时晶圆曲率测量直接测量硅衬底中的应变，研究氮化镓沉积过程中薄膜应力的动态变化。生长后结构表征用于阐明应力产生和松弛的机制。该项目支持宾夕法尼亚州立大学的两名研究生的博士论文工作，他们通过与EPC合作，接触到与全球半导体代工厂的工艺放大和制造相关的问题。来自宾夕法尼亚州代表性不足的群体和经济困难地区的本科生和高中生通过宾夕法尼亚州立大学的暑期项目参与了这项研究。技术描述：这个GOALI项目是用于混合电力电子电路的轴向（100）Si衬底上的AlGaN/GaN异质结构的外延生长。人们正在探索应变Si/SiGe虚拟衬底作为控制（100）Si表面主导重构的途径，从而减少AlN核的面内旋转错位。利用原位晶圆曲率测量来监测SiGe层中的应力松弛，测量生长温度下Si层中的拉伸应变，并研究其对GaN/AlN异质外延的影响。研究了n型掺杂化学对氮化镓中应力演化的影响。该项目寻求对Si表面应变在GaN/AlN外延中的作用的基本理解，然后可以利用它来开发用于功率器件的（100）Si上的AlGaN/GaN二维电子气体异质结构。项目团队在Si衬底上的iii组氮化物异质外延（宾夕法尼亚州立大学）和功率场效应晶体管（EPC）的GaN-on-Si工艺放大和商业化方面具有互补的专业知识。此外，EPC在利用该项目开发的GaN on (100) Si生长技术实现混合电路设计方面处于有利地位。