Quantum Photonics

Quantum photonics applies the same integrated-chip platform used for classical photonic computing to manipulate individual photons for quantum information processing. The case for photons as quantum information carriers is strong: they travel at the speed of light, interact weakly with the environment (low decoherence), and can be routed through standard silicon photonic waveguides and beam splitters. The case against is operational temperature — superconducting nanowire single-photon detectors (SNSPDs), the highest-efficiency photon detectors available (>90%), require cryogenic operation, creating a practical deployment gap that is not yet solved.

The materials landscape spans four platforms: silicon (CMOS-compatible, no efficient on-chip emission), silicon nitride (low loss, broad bandwidth), lithium niobate (fast modulation, nonlinear efficiency), and III-V semiconductors like InP and GaAs (efficient light emission, higher loss). Each excels in a different part of the quantum photonic stack. Integration of these materials — heterogeneous photonic integration — is the engineering frontier: putting efficient single-photon sources, low-loss routing, and high-efficiency detection on a single chip.

Single-photon sources remain the primary bottleneck. Quantum dots in photonic crystal waveguides achieve near-unity coupling probability (β ≈ 98%), but are probabilistic emitters — they produce photon pairs, not guaranteed single photons on demand. Multiplexing strategies to achieve effective determinism add system complexity. Meanwhile, silicon's indirect bandgap prevents efficient on-chip light emission entirely, requiring heterogeneous bonding of III-V gain materials.

The applications divide into three categories: quantum computing (linear optical quantum computing using photons as qubits, beam splitters and phase shifters as gates), quantum communication (quantum key distribution and quantum repeaters for unbreakable encryption, with silicon photonics as the integration platform), and quantum sensing (precision measurement beyond the standard quantum limit using entangled light). Of these, quantum communication via QKD is the most commercially mature and the most direct application of the integrated silicon photonics manufacturing base.

Key Claims

• SNSPDs achieve >90% detection efficiency — Best-in-class single-photon detectors, but require cryogenic operation. Evidence: strong (Quantum Photonics on a Chip)
• Quantum dots achieve β ≈ 98% coupling probability — Near-unity photon emission into photonic crystal waveguides. Evidence: strong (Quantum Photonics on a Chip)
• Silicon photonics is the preferred integration platform for QKD — CMOS manufacturing compatibility enables scalable quantum communication devices. Evidence: moderate (Quantum Photonics on a Chip)
• Ideal single-photon sources remain unrealized — All current implementations are probabilistic; determinism requires multiplexing. Evidence: strong (Quantum Photonics on a Chip)
• Cryogenic operation limits practical deployment — SNSPDs require sub-kelvin temperatures; room-temperature alternatives (SPADs at ~65% efficiency, Ge APDs at 5.27% at 80K) are substantially inferior. Evidence: strong (Quantum Photonics on a Chip)

Benchmarks & Data

• SNSPD efficiency: >90% at 1550 nm, cryogenic (Quantum Photonics on a Chip)
• Quantum dot coupling: β ≈ 98% in photonic crystal waveguides (Quantum Photonics on a Chip)
• SPAD quantum efficiency: ~65%, room temperature (Quantum Photonics on a Chip)
• Ge avalanche photodiode: 5.27% efficiency at 80K (Quantum Photonics on a Chip)
• Transition-edge sensors: resolve up to 6 photons at 1550 nm (Quantum Photonics on a Chip)

Relationship to Classical Photonic Computing

Quantum photonics shares fabrication platforms (silicon photonics, silicon nitride, lithium niobate) with classical photonic computing but targets fundamentally different regimes — single-photon manipulation rather than coherent-field computation. Some advances are synergistic: improvements in low-loss waveguides, photon detectors, and heterogeneous integration benefit both fields. The commercial infrastructure being built for classical CPO (TSMC COUPE, UCIe optical) may eventually accelerate quantum photonic integration by proving the manufacturing pathways at scale.

Open Questions

• Can room-temperature single-photon detectors reach >90% efficiency to eliminate cryogenic requirements?
• What is the path to deterministic (non-probabilistic) single-photon sources on chip?
• Can heterogeneous integration of III-V sources and silicon photonic waveguides achieve wafer-scale yield?
• At what qubit count does photonic quantum computing approach quantum advantage for practical problems?
• How does photonic QKD compare to post-quantum cryptography for practical secure communication?

Related Concepts

• Photonic Neural Networks — Classical photonic computing sharing the same silicon photonics platform
• Co-Packaged Optics — CPO infrastructure that may benefit from quantum photonic detector advances

Changelog

• 2026-04-14 — Initial compilation from 1 source (April 14 ingestion batch)