外延
气相
材料科学
光电子学
金属有机气相外延
相(物质)
增长率
化学
纳米技术
物理
图层(电子)
热力学
几何学
有机化学
数学
作者
R. Lang,Christoph Klein,Jens Ohlmann,Frank Dimroth,David Lackner
出处
期刊:Journal of vacuum science & technology
[American Institute of Physics]
日期:2024-02-21
卷期号:42 (2)
被引量:3
摘要
The aim of this review paper is to summarize a decade of research focused on enhancing metalorganic vapor-phase epitaxy (MOVPE) growth rates of GaAs, driven by the imperative for most cost-effective and energy-efficient III–V compounds’ production. While MOVPE is renowned for producing high-quality devices, it has been constrained by production cost. For example, MOVPE was traditionally thought to have moderate growth rates that limit the throughput of the cost-intensive reactors. Recent research endeavors, however, have demonstrated ultrafast growth rates, exceeding 280 μm/h, with a remarkable group III precursor utilization efficiency of over 50%. It is worth noting that even with increased growth rates, the surface quality remains unaffected in terms of roughness and morphology. Nonetheless, optoelectronic properties, such as minority carrier lifetime, deteriorate for both p- and n-doped materials under constant growth conditions. This is attributed to an increase in the defect density of arsenic antisites, particularly EL2 and HM1 defects, as revealed by deep-level transient spectroscopy investigations. Some of these losses can be mitigated by optimizing growth conditions, such as elevating the temperature and reducing the V/III ratio. The latter not only restores some of the material quality but also increases the growth rate and reduces precursor consumption. Still, fully recovering the original reference lifetimes remains a challenge. Solar cell results indicate that structures with predominantly n-type absorbers are less affected by reduced minority carrier lifetimes. A remarkable 24.5% efficiency was achieved in a GaAs single-junction solar cell grown at 120 μm/h, representing less than 1 min of growth time for the absorber layers.
科研通智能强力驱动
Strongly Powered by AbleSci AI