Indium phosphideInP
The native platform of telecom light. Direct bandgap, lattice-matched quaternaries spanning 1.0–1.65 μm, and monolithic integration of lasers, amplifiers, modulators, and detectors on one substrate.
InP is the only mainstream platform where the laser, the modulator, the amplifier, and the detector can all be grown and fabricated monolithically. Nearly every DFB laser shipping in a telecom transceiver is an InP device, and full InP PICs with hundreds of components underpin coherent optics.
Key properties
| Property | Value | Conditions |
|---|---|---|
| Refractive index | 3.16 | 1550 nm, 300 K |
| Bandgap | 1.344 eV (direct) | 300 K |
| Bandgap wavelength | 923 nm | 300 K |
| Emission coverage (quaternaries) | ~1.0–1.65 μm | InGaAsP / InAlGaAs on InP |
| Thermo-optic coefficient | K⁻¹ | 1550 nm |
| Thermal conductivity | 68 W·m⁻¹·K⁻¹ | 300 K |
| Typical passive waveguide loss | 1.5–3 dB/cm | generic foundry processes |
What the numbers mean in practice
The direct bandgap is the whole point: efficient radiative recombination makes InP the gain medium, not just the substrate. Alloying with the lattice-matched quaternaries InGaAsP or InAlGaAs tunes the gap across the entire O- through L-band range while keeping strain manageable, which is how one substrate technology covers 1310 nm datacom and 1550 nm coherent alike. InAlGaAs's larger conduction-band offset improves electron confinement, which shows up directly as better high-temperature performance (higher ) in uncooled lasers.
The costs relative to silicon: wafers are smaller (2–4 in. common, 6 in. emerging against silicon's 200–300 mm), more expensive, and more fragile; passive losses run roughly an order of magnitude above SOI and two above nitride; and the lower thermal conductivity (68 vs. 149 W·m⁻¹·K⁻¹) makes heat extraction from high-power actives harder. This trade is exactly why heterogeneous integration exists: III-V gain bonded onto silicon or nitride passives tries to keep the best of each.
At the bench, remember that the substrate is transparent at your measurement wavelength: light launched into an InP chip can happily travel through the substrate and reappear at the detector, producing spurious signals and etalon fringes that have wasted many alignment sessions.
References
- G. D. Pettit, W. J. Turner, "Refractive index of InP," J. Appl. Phys. 36, 2081 (1965).
- S. Adachi, Physical Properties of III-V Semiconductor Compounds, Wiley (1992).
- M. Smit, K. Williams, J. van der Tol, "Past, present, and future of InP-based photonic integration," APL Photonics 4, 050901 (2019).
- I. Vurgaftman, J. R. Meyer, L. R. Ram-Mohan, "Band parameters for III-V compound semiconductors and their alloys," J. Appl. Phys. 89, 5815 (2001).