Micro light emitting diodes (micro-LEDs) based on III-V compound semiconductor have been regarded as an ultimate solution for next generation display due to high luminous efficiency, low power consumption, fast response time, and thermal stability. Their superior characteristics are essential for a variety display application, including high-resolution television with high contrast ratio, high brightness smart watches and augmented-reality (AR) devices. However, there are some thechnical issues to commercialize micro display. Micro-LEDs for display application need to consider their optical and electical properties such as luminance uniformity, efficiency degradation by size reduction, wavelength shift depending on current density, light propagation from emitter and transfer method for fabrication. The poor luminance uniformity and wavelength shift deteriorate the quality of display, and the efficiency degradation problem as miniaturising micro-LED limits to develop the high-resolution display, respectively. Furthermore, transfer method must be developed to reduce production cost and emission beam shape from micro-LEDs must be considered for commercial micro displays. In conclusion, the improvement and development of technology related to micro-LEDs could be a breakthrough for commercializing the high quality micro-LED display.

We will summarize recent progress made within our group on self-assembled quantum dot device development for quantum repeater and quantum computer applications. A particular emphasis will be on semiconductor quantum dots embedded in circular Bragg grating cavities. For scalability, spatially deterministic placement of quantum dots in bullseye cavities is pursued and tuning by electric and strain fields are implemented. To apply electric fields, a new device design for electrically contactable circular Bragg grating cavities in labyrinth geometry is employed.

Dr. Yung-Chung Kao is the Chairman/CEO & Founder of IntelliEPI, a leading merchant supplier of III-V compound semiconductor epitaxy materials based on Molecular Beam Epitaxy (MBE) technology to serve electronic and optoelectronic industries. The company was founded in 1999 and is currently trading in the Taipei Exchange.

Dr. Kao has over 40 years of extensive technical experience in compound semiconductor materials development and manufacturing. Previously, he was at Texas Instruments, Inc. from 1987 to 1998 and served as a Senior Member of Technical Staff where he headed the III-V MBE group at TI Central Research Laboratory. Dr. Kao has authored/co-authored well over 100 technical publications, 2 book chapters, and has been granted 13 US patents, in the areas ranging from MBE technology development, semiconductor materials, to advanced MMIC devices.. He received his Ph.D. in Electrical Engineering from UCLA in 1987, MSEE from Texas A&M University, and his BS in Physics from National Tsing Hua University, Hsinchu, Taiwan in 1978.

This presentation will describe the evolution of epi-foundries from early day traditional merchant single source epitaxy services to nowadays hybrid epitaxy model with not only epi-tools integration but also the addition of regrowth and post-growth metallization steps to the epi-device structures. Since III-V semiconductors in particular have the highest level in epi-layers maturity and complexity, we will focus our discussion using III-V epi-devices.

Molecular beam epitaxy (MBE) and metal organic chemical vapor deposition (MOCVD) are two most important modern epitaxy techniques that used extensively to prepare epi-wafers for III-V semiconductor electronic/optoelectronic devices. Most of the foundries were formed about 25 to 35 years ago, during the early part of the mobile phone era. The merchant epi-foundries bought either MBE or MOCVD equipment, now up to 8x6in multiwafer production configuration, from a handful of equipment manufacturers and to supply custom designed, tightly specified, epitaxial films on either GaAs or InP substrates for a variety of customers. In this type of service, epitaxial design come from the customer and the epi-foundry then converts the customer’s design into overlaying epitaxial films on one or multiple substrates using either MBE or MOCVD apparatus. As-grown epi-wafers characterization is done to ensure successful completion of the job. And wafers are delivered to customers or IC foundries for processing and fabrication. Up to date, almost all the advanced and mature III-V compound semiconductors devices are prepared on II-V epitaxial wafers.

Since every epi-technique has its limitations and no single growth machine can be configured with all the species. There are also apparent risks of cross contamination if certain species are used in the same chamber. In other words, each epitaxial tool’s configuration, various epi-layers attributes (control of thickness, composition, and doping range), and interface quality between layers, determine the epitaxial devices performance upper limits. As all the dataintensive applications in communication networks demand ever higher bandwidth and capacity, the performance requirements on all optoelectronics and RF devices are almost pushed beyond what current epi-tools can provide. In order to reach higher power, efficiency, and speed, epitaxial industry needs to expand it capabilities and relax the restrictions to support the 5G, and beyond. We will discuss the following epitaxial combination possibilities, or called hybrid epitaxy, that are either established at our facilities or under proposition to be included:

• MOCVD & MBE growth combination: Dilute-nitride QW VCSEL, quantum dot QW Laser
• Epitaxy processing & re-growth: VCSEL with tunnel junction, MBE n+GaN on MOCVD GaN HEMT
• Inter-systems linkage: UHV inter-connection of As/P MBE system with As/Sb/N MBE system; MBE and MOCVD
• MBE systems with in situ sensors installed for monitoring and control
• MBE UHV linkage to processing chamber: post InP HBT epi-growth in situ metallization, pre epitaxial growth hydrogen cleaning
• Cluster tool connection to many single wafer MBE/MOCVD chambers: for 200mm & 300mm GaN/Si

In this talk, we will use various device structures prepared under hybrid epitaxy to illustrate the feasibility to push device performance to higher limits without substantially increase in cost. The fundamental tools used are still MBE and MOCVD. The hardware linkage tools and various techniques used to protect and/or to clean the interface, respectively, will be discussed in the presentation.

Masaya Notomi received B.E., M.E. and Ph. D. degrees in applied physics from University of Tokyo in 1986, 1988, and 1997. He joined NTT Optoelectronics Laboratories, NTT Corporation, Japan in 1988, and moved to NTT Basic Research Laboratories in 1999. He is currently Senior Distinguished Scientist of NTT Basic Research Laboratories, and heading NTT Nanophotonics Center. Since 2017, he has been cross-appointed as a professor in Department of Physics, Tokyo Institute of Technology, Japan. He has been working on semiconductor quantum nanostructures, and physics and applications of nanophotonics, including photonic crystals and plasmonics. His works involve novel phenomena arising from nanophotonic structures, enhancement of light-matter interactions, and applications for integrated optoelectronic computations. He received IEEE/LEOS Distinguished Lecturer Award (2006), JSPS (Japan Society for the Promotion of Science) Prize (2009), Japan Academy Medal (2009), Commendation for Science and Technology by the Japanese Minister of Education, Culture, Sports, Science and Technology (2010), and Distinguished Achievement and Contributions Award of IEICE (The Institute of Electronics, Information and Communication Engineers) (2021). IEEE Fellow since 2013.

Recently, there are growing expectations for optoelectronic computing because of increasing demands for the low-latency computation capacity beyond CMOS processors, especially for deep learning applications. Optics enables energy-efficient and ultralow-latency computations for a certain area, such as for vector-matrix multiplications (VMMs). We regard that optoelectronic accelerators based on optical VMMs will be crucial for future ICT, which would be consisting of (1) linear transformation optical circuits, (2) nonlinear elements, and (3) efficient OE/EO conversion devices. In this talk, we will show integrated nanophotonics will play a crucial role for these three factors. The most important advantage of optics for computation lies in (1). We will show the impact of controllable light interference in photonic integrated circuits. Despite of (1), we need some nonlinearity for efficient computation systems, which used to be weakness of optics. As for (2), we will show that a combination of functional devices and materials, including III/V semiconductors or two-dimensional materials enables integrable efficient nonlinear elements in optical circuits. Finally, we emphasize that any optical circuits should be efficiently merged with electronic circuits. As for (3), we believe that nanophotonic devices would lead to a paradigm shift for optoelectronic integration, and we will show some of examples.

Superconducting flux qubits are based on a superposition of clock-wise and anti-clockwise currents formed by millions of Cooper pairs. In order to excite the system in a superposition state, the half-quantum flux of magnetic field is passed through the superconducting circuit containing one or several Josephson junctions. The system is forced to generate a circular current to either reduce the magnetic flux to zero or to build it up to a full-quantum flux. We argue that a valuable alternative to superconducting flux qubits may be offered by qubits based on superfluid currents of quasiparticles of liquid light: exciton-polaritons, propagating in plane of semiconductor microcavities [1]. Circular currents of exciton-polaritons mimic the superconducting flux qubits being composed by a large number of bosonic quasiparticles that compose a single quantum state of a many-body condensate. The essential difference comes from the fact that polaritons are electrically neutral, and the magnetic field would not have a significant effect on a polariton current. We note however, that the phase of a polariton condensate must change by an integer number of 2π, when going around the ring. If one introduces a π-phase delay line in the ring, the system is obliged to propagate a clockwise or anticlockwise circular current to reduce the total phase gained over one round-trip to zero or to build it up to 2π. We show that such a π-delay line can be provided by a dark-soliton embedded into a ring condensate and pinned to a potential well created by the C-shape non-resonant pump- spot. The physics of resulting split-ring polariton condensates is essentially similar to the physics of flux qubits. In particular, they exhibit pronounced Bloch oscillations passing periodically through clockwise and anticlockwise current states. We argue that qubits based on split-ring polariton condensates may be characterized by a very high figure of merit due to the topological protection of superfluid circular currents. Moreover, as the Bose-Einstein condensation and superfluidity of exciton-polaritons were observed at the room temperature [2], quantum networks based on polariton qubits would not require cryogenic operation temperatures. This makes them a valuable alternative to superconducting qubits [3].