Title: Molecular Beam Epitaxial Growth Optimization for Next Generation Optoelectronic Devices Based on III-V Semiconductors
Date: July 31, 2020
Time: 10:00 AM
Place: REMOTE PARTICIPATION
Supervisor(s): Wasilewski, Zbig
Molecular beam epitaxy (MBE) has been serving as the best tool for group III-V compound growths due to its capability of depositing epilayers with high single crystalline quality, high purity and Angstrom-scale thickness precision. In spite of decades of work, further optimizations of MBE growth conditions are required in order to meet the increasing demands of next generation and novel devices. For instance, low temperature (LT) deposition far away from thermal equilibrium growth conditions brings several benefits such as suppressed undesirable interface inter-diffusion, dopant diffusion and segregation. More importantly, it has been proven to be a very useful method in forming non-stoichiometric III-V compounds. Although this creates defects like group V antisites and interstitials, they can be utilized in making terahertz (THz) photoconductive antennas (PCAs). Since the development of THz technology in the late 20th century, THz PCAs have been drawing tremendous research interest due to their low-cost and portability. The amount of excess As embedded which alters the LT grown III-V material properties is sensitive to the V/III flux ratio, and more importantly, the growth temperature at the range of ~200–300°C. It is crucial to determine a critical substrate temperature that simultaneously maximizing the excess As incorporation without losing single crystallinity, as both features are essential for high performance broadband THz PCAs applications. However, substrate temperature monitoring becomes difficult when the growth is switched to LT regime. Conventional substrate temperature measurement techniques such as pyrometry have limited precision at LT due to the exponential decrease of the substrate thermal radiation intensity with the temperature, hence an alternative mean for precise temperature monitoring is required. On the other hand, InAlGaAs has been an emerging candidate for 1319 nm laser pumped, high efficiency photonic power converters (PPCs). To acquire smooth surface morphology and minimize crystalline defects, such quaternary material is preferred to be lattice-matched on InP(001) substrates. A clear correlation between group III compositions, lattice constant and bandgap of InAlGaAs, as well as a reproducible approach to grow lattice-matched InAlGaAs structures need to be established to optimize the PPC performance.
A comprehensive study is conducted to optimize MBE growth conditions of several III-V materials, including LT GaAs, mid-temperature (MT) and LT InGaAs-InAlAs superlattice (SL), and InAlGaAs, with parameters such as substrate temperature, As overpressure and doping concentration being repetitively tuned. In particular, an integrated spectral pyrometry (ISP) technique is proposed, which is particularly well suited to monitor the growth temperature of InAs and InSb. Potentially, ISP can also be applied to other LT growths with semiconductors of smaller bandgaps when pyrometer does not work. In-growth and post-growth surface morphologies were investigated with the aid of reflection high energy electron diffraction (RHEED) and Nomarski microscopy. LT GaAs and InGaAs-InAlAs SL based THz PCAs were characterized with 780 and 1550 nm pumped time domain spectroscopy (TDS) measurements and correlation between THz signal amplitude, pulse width and bandwidth with growth and anneal temperatures are discussed. The material properties of InAlGaAs structures were investigated with emphasis of discussing challenges of lattice mismatch and in-growth flux drift studied by high resolution x-ray diffraction (HRXRD) fitting. The performance of InAlGaAs tunnel junctions was optimized by MBE, with state-of-the-art, 1319 nm pumped single junction (SJ) and two junction (TJ) PPC devices illustrated. The effects of growth temperature on current-voltage characteristics and unintentionally strained InAlGaAs epilayers on quantum efficiency are discussed.