III–V optoelectronics, together with optical fiber technology, have been the foundation of the modern Internet. InP-based devices are the core components of optical telecommunication systems, while GaAs-based surface-emitting lasers are mass-produced as low-cost, compact light sources for data centers. InGaAs photodetectors offer the highest performance among near-infrared detectors, with detection wavelengths extendable up to ~2.4 μm by increasing indium composition, enabling operation without cryogenic cooling. InSb is a key material for mid-wavelength infrared (MWIR) detection, particularly for space and astronomical applications. Furthermore, advances in quantum well and quantum dot infrared photodetectors based on superlattice structures have expanded the detectable wavelength range into the long-wavelength infrared (LWIR) region.
Despite these advantages, the high fabrication cost of III–V devices remains a major limitation. III–V substrates are relatively small, fragile, and expensive, accounting for nearly 50% of the total device cost. Replacing III–V wafers with larger, lower-cost, and mechanically robust silicon wafers is therefore highly attractive. In addition, silicon is an inefficient light emitter due to its indirect bandgap, whereas III–V materials excel in light generation. Integrating III–V materials on silicon thus enables complementary functionalities that cannot be achieved by either material alone.
Direct epitaxial growth of III–V materials on silicon, however, presents significant challenges, including lattice mismatch, thermal expansion differences, and defect formation. Our group has been addressing these challenges to demonstrate high-performance III–V optoelectronic devices on silicon substrates, targeting applications in silicon photonics, solar cells, and infrared detectors.