Novel InSb quantum dots monolithically grown on silicon for low cost mid-infrared light emitting diodes

中红外光谱区（2-5 μ m）包含了许多污染物和有毒气体和液体的基本吸收带，引起了全世界的极大兴趣。其中包括二氧化碳、一氧化碳和氯化氢等气体，这些气体需要在石油钻井平台、煤矿、垃圾填埋场和汽车尾气等许多行业进行精确的现场多组分监测。强吸收波段也存在于药物中间体、药品、麻醉品和生物化学品中，其吸收强度通常比近红外强2个数量级，从而在环境监测、生物医学、工业过程控制和健康与安全等领域实现高选择性和高灵敏度的检测。还有3.6和3.8 um之间的大气传输窗口，可以在民用和军事情况下实现自由空间光通信和热成像，以及为国土安全开发红外对抗措施。然而，由于缺乏高效和负担得起的光源和探测器，这些应用尚未得到充分利用。这项工作提出了一种基于锑化铟（InSb）量子点在低成本硅（Si）衬底上的新型发光二极管（LED）架构的生长和制造。这将彻底改变我们利用这些设备的方式，并导致光学系统的成本和尺寸的显着缩放，从而使其能够广泛采用。它还将使光子元件能够直接嵌入到电子电路中，这将开辟中红外光子集成电路的新领域。这将在集成“芯片实验室”传感器和紧凑型生物芯片等领域产生全新的技术，为英国带来巨大的商业利益和机会。在过去的几年中，使用带间级联激光器和II型超晶格的中红外器件的发展取得了重大进展。然而，这些结构非常复杂，制造成本昂贵，并且生长在锑化镓（GaSb）衬底上，这些衬底质量差，成本高（约为Si成本的50倍）并且只能用于小尺寸。在硅上的生长将是最理想的，以实现成本效益的制造，并确保未来的商业成功。III-V在Si上直接外延生长的主要障碍是III-V/Si界面之间的大晶格失配，导致大密度的螺纹位错（td），严重影响器件性能。本项目将通过在硅上生长的低缺陷密度GaSb缓冲层上实现基于InSb量子点的新器件设计来克服这一问题。与体结构或量子阱结构相比，量子点的主要优点是机械鲁棒性和对TD的极低灵敏度，以及抑制非辐射俄歇复合以提高量子效率。在量子阱器件中，每一个通过它传播的螺纹位错都将作为一个非辐射中心，大大降低了器件的性能。然而，在量子点中，每个TD只会“杀死”一个或几个孤立的点，这不会显著影响器件性能，因为缓冲层中的TD密度可以降低到中低水平。低缺陷密度的GaSb缓冲层应通过新颖的“界面错配阵列（IMF）”和位错滤波层来实现，这些层设计用于弯曲和湮灭III-V/Si界面产生的TD。Si基中红外led将与学术（南安普顿大学和蒙彼利埃大学）和工业（化合物半导体技术和气体传感解决方案）项目合作伙伴密切合作开发，以评估实际应用中的器件性能，这将有助于实现未来的商业化。

There is great worldwide interest in the mid-infrared spectral region (2-5 um) as it contains the fundamental absorption bands of a number of pollutant and toxic gases and liquids. These include gases such as carbon dioxide, carbon monoxide and hydrogen chloride which require accurate, in-situ multi-component monitoring in a number of industries such as oil-rigs, coal mines, land-fill sites and car exhausts. Strong absorption bands also exist for drug intermediates, pharmaceuticals, narcotics and biochemicals where the absorption strength is typically 2 orders of magnitude stronger than in the near-infrared allowing highly selective and sensitive detection in the fields of: environmental monitoring, bio-medicine, industrial process control and health and safety. There is also an atmospheric transmission window between 3.6 and 3.8 um which enables free space optical communication and thermal imaging in both civil and military situations as well as the development of infrared countermeasures for homeland security. However, these applications have yet to be fully exploited due to the lack of efficient and affordable light sources and detectors. This work proposes the growth and fabrication of a new light emitting diode (LED) architecture based on indium antimonide (InSb) quantum dots onto low cost silicon (Si) substrates. This will revolutionize how we utilize these devices and lead to a dramatic scaling in the cost and size of the optical systems to enable their widespread uptake. It will also enable the photonic components to be directly embedded into electronic circuits which would open up a new field of mid-infrared photonic integrated circuits. This would generate entirely new technology in areas such as integrated 'lab-on-a-chip' sensors and compact biochips bringing great commercial benefits and opportunities to the UK.In the last few years, there has been significant progress in the development of mid-infrared devices using interband cascade lasers and type II superlattices. However these structures are extremely complex and expensive to fabricate and are grown on gallium antimonide (GaSb) substrates which are of poor quality, high cost (~50 times the cost of Si) and are only available in small sizes. Growth onto silicon would be most desireable to enable cost effective manufacture and to ensure future commercial success. The major obstacle in direct epitaxial growth of III-Vs onto Si is the large lattice mismatch between the III-V/Si interface, resulting in a large density of threading dislocations (TDs) which strongly deteriorate the device performance. This project shall overcome this by implementation of a new device design based on InSb quantum dots on low defect density GaSb buffer layers grown on Si. The key advantages are the mechanical robustness and very low sensitivity of the quantum dots to TD compared to bulk or quantum well structures, and the suppression of non-radiative Auger recombination to increase the quantum efficiency. In a quantum well device, every threading dislocation which propagates through it will act as a non-radiative centre drastically reducing the device performance. However in a QD, each TD will only 'kill' one or a few isolated dots which will not significantly affect device performance providing the TD density in the buffer layer can be reduced to moderate-to-low levels. Low defect density GaSb buffer layers shall be realized through novel 'interfacial misfit arrays (IMF)' and dislocation filtering layers designed to bend and annihilate TD generated at the III-V/Si interface. The Si based mid-infrared LEDs will be developed in close collaboration with academic (University of Southampton and University of Montpellier) and industrial (Compound Semiconductor Technologies and Gas Sensing Solutions) project partners to evaluate device performance for use in practical applications which will help to achieve future commercialisation.