Photonic Qubit

Figure

Description

The photonic qubit is the umbrella discrete-variable optical qubit family in which quantum information is encoded into a small Hilbert space of one or a few optical modes, most commonly a single photon in a polarization, path/dual-rail, or time-bin basis. A representative single-photon state is

$$|\psi ⟩=\alpha |1,0⟩+\beta |0,1⟩,$$

where the two logical basis states may correspond to horizontal/vertical polarization, two spatial rails, or two well-separated time bins. This entry is about that single-photon discrete-variable family rather than continuous-variable photonic computing or oscillator-code encodings such as GKP.

Photonic qubits are attractive because photons suffer extremely little decoherence while propagating through low-loss waveguides or optical fiber, making them natural carriers of quantum information across a chip or between modules. The tradeoff is that photons interact only weakly, so universal computation usually relies on measurement-induced entangling operations, ancillary photons, feed-forward, or photonic resource-state constructions rather than a strong native two-photon interaction.

In practice, photonic quantum computing splits into several closely related layers: the encoding itself (dual-rail, polarization, time-bin), gate-based linear optics (KLM-style), cluster-state measurement-based photonics, and fusion-based fault-tolerant architectures. The umbrella entry should therefore be read as the common physical substrate beneath those more specific child entries.

Hamiltonian

There is no single universal microscopic Hamiltonian for all photonic-qubit platforms, but passive control of a discrete-variable photonic qubit is well represented by a bilinear mode Hamiltonian,

$$H_{\mathrm{passive}}=\sum _{ij}{h}_{ij}{a}_i^{†}{a}_j,$$

which generates beam-splitter and phase-shifter transformations among optical modes while conserving total photon number. For a two-mode encoding,

$$H_{\mathrm{BS}}=\kappa a^{†}b+{\kappa }^{\ast }{b}^{†}a,H_{\varphi }=\varphi a^{†}a,$$

which, for real $\kappa$, reduces to the familiar beam-splitter form and provides a representative $SU(2)$ control set for single-qubit rotations in the logical subspace.

Universal computation is not obtained from this passive Hamiltonian alone. Instead, the crucial nontrivial resource is measurement-induced effective nonlinearity: two-photon interference, ancilla preparation, photon counting, and classical feed-forward generate heralded entangling operations or larger cluster/resource states. For that reason, a photonic-qubit umbrella entry should treat the bilinear optics Hamiltonian as representative of single-qubit transport and control, not as the whole computational story.

Motivation

• Minimal transport decoherence: photons are the natural flying qubits for on-chip routing and long-distance networking.
• Encoding flexibility: the same logical qubit idea can live in polarization, path, or time-bin modes depending on whether the priority is integrated optics, interferometric stability, or fiber transmission.
• Room-temperature optical layer: most sources, interferometers, and waveguides operate without millikelvin infrastructure, though many high-performance detectors do require cryogenics.
• Erasure-friendly failure mode: photon loss often appears as a flagged absence event rather than an unheralded over-rotation, which is attractive for loss-aware fault-tolerance.
• Main hardware challenge: the platform pays for those advantages with loss sensitivity, source/detector inefficiency, mode mismatch, and the lack of a deterministic native two-photon gate.

Experimental Status

Scalable linear-optical route established — Knill, Laflamme, and Milburn (2001):

• Showed that universal quantum computation is possible with single-photon sources, passive linear optics, photon counting, and feed-forward.
• Established the canonical photonic trade: weak native interactions can be replaced by measurement-induced nonlinear operations at resource-cost overhead.

First landmark entangling-gate experiment — O’Brien et al. (2003):

• Demonstrated a post-selected all-optical controlled-NOT gate.
• Confirmed that nontrivial two-qubit photonic logic is experimentally reachable without a direct optical nonlinearity.

Encoding-specific diversification (2000s-2020s):

• Dual-rail/path encoding became the canonical integrated-LOQC language.
• Time-bin encoding became the workhorse for deployed fiber links and long-distance photonic networking.
• Polarization encoding remained central for tabletop and many source-characterization experiments.

Fault-tolerant architectural shift — Bartolucci et al. (2023):

• Modern photonic roadmaps increasingly emphasize cluster-state and fusion-based resource-state consumption rather than literal KLM gate teleportation.
• The important modern question is less “can photons compute?” and more “can loss, feed-forward latency, and source multiplexing be engineered below threshold?”

Recent 2024-2026 direction check:

• A targeted re-check did not uncover a single peer-reviewed 2025-2026 experimental result that supersedes the canonical umbrella milestones above for the whole discrete-variable photonic-qubit family.
• The most relevant recent motion is architectural and systems-level: fusion-based fault tolerance, large photonic resource-state generation, and networked spin-photon modules such as the 2024 Distributed Quantum Computing in Silicon preprint.

Key Metrics

Metric	Value	Notes	Fidelity reference
Representative logical encodings	polarization, path/dual-rail, time-bin	Same single-photon DV qubit family, different optical modes	Kok et al. 2007
Native 1Q control	Deterministic linear optics	Beam splitters and phase shifters implement single-qubit mode rotations	Knill et al. 2001
Native 2Q interaction	None in passive optics	Universal 2Q logic is usually measurement-induced and probabilistic	Knill et al. 2001
First landmark optical 2Q gate demo	2003	Post-selected all-optical CNOT	O’Brien et al. 2003
Dominant scaling bottleneck	Photon loss + indistinguishability + detector efficiency	Weak interaction is manageable only if those optical imperfections stay below threshold	Kok et al. 2007
Modern FT architecture direction	Fusion / cluster-state resource consumption	Emphasizes small entangled resource states plus destructive fusion measurements	Bartolucci et al. 2023
Recent systems milestone	Remote entanglement plus teleported gate between silicon spin-photon nodes	Important 2024 direction, but still a preprint and not a canonical family-wide benchmark	Photonic Inc. 2024

References

Foundational proposals and reviews

• E. Knill, R. Laflamme, and G. J. Milburn, “A scheme for efficient quantum computation with linear optics,” Nature 409, 46 (2001) — arXiv:quant-ph/0006088
• P. Kok, W. J. Munro, K. Nemoto, T. C. Ralph, J. P. Dowling, and G. J. Milburn, “Linear optical quantum computing with photonic qubits,” Rev. Mod. Phys. 79, 135 (2007) — arXiv:quant-ph/0512071

Landmark experiment

• J. L. O’Brien, G. J. Pryde, A. G. White, T. C. Ralph, and D. Branning, “Demonstration of an all-optical quantum controlled-NOT gate,” Nature 426, 264 (2003)

Modern fault-tolerant architecture and recent direction

• S. Bartolucci et al., “Fusion-based quantum computation,” Nat. Commun. 14, 912 (2023) — arXiv:2101.09310
• Photonic Inc., “Distributed Quantum Computing in Silicon,” arXiv:2406.01704 (2024)

Linked Papers

• knill-2001-klm
• kok-2005-review-article-linear-optical
• obrien-2003-all-optical-cnot
• bartolucci-2023-fbqc
• photonic-2024-distributed-qc

Evergreen context

• erasure-error-vs-pauli-error — photonic platforms are unusually important for erasure-aware thinking because missing photons are often flagged absence events rather than hidden unitary errors.
• noise-bias-and-asymmetric-error-channels — the platform only scales cleanly when its dominant imperfections stay in the structured loss, mismatch, and detection-failure regime rather than washing out into generic depolarizing noise.
• threshold-theorem — photonic architectures are one of the clearest demonstrations that threshold engineering can trade deterministic interactions for resource overhead, feed-forward, and loss-tolerant coding.

Related Entries

• dual-rail-photonic-qubit — canonical path-encoded child entry
• time-bin-photonic-qubit — fiber-native temporal encoding
• linear-optical-photonic-qubit — gate-based LOQC framing of the same family
• photonic-cluster-state-mbqc-qubit — measurement-based photonic computation
• fusion-based-photonic-qubit — fault-tolerant resource-state architecture