T-Center Qubit (Silicon Spin-Photon)

Figure

Description

The silicon T centre is a telecom-band spin-photon interface built from a carbon-pair plus hydrogen defect complex in silicon, commonly modeled as a C-C-H complex. Its optical zero-phonon line sits near 1326 nm in the telecom O-band, so it can interface directly with standard fibre components without the frequency conversion usually needed for visible-wavelength solid-state emitters.

A single T centre combines an optically addressable electron spin with nearby nuclear-spin memories. Depending on isotopic configuration, the register can include the electron spin, the hydrogen nuclear spin, and up to two nearby ¹³C nuclear spins. That combination makes the platform interesting not just as a photon source, but as a small local memory-and-processing node inside a silicon photonics stack.

Experimentally, the platform has moved in distinct steps. Bergeron et al. (2020) established the integrated telecom photon-spin interface in silicon. Higginbottom et al. (2022) reported optical observation of single spins in silicon T centres. DeAbreu et al. (2023) demonstrated waveguide-integrated T centres. A 2024 Photonic Inc. preprint then reported remote entanglement between T centres in separate cryostats and a preliminary, post-selected teleported CNOT sequence over a selected basis set. In 2025, electrically triggered spin-photon devices in silicon showed that T-centre nanophotonics can also be driven electrically, not only optically.

Hamiltonian

A representative ground-state spin Hamiltonian is

$H_{\mathrm{gs}}={\mu }_B\mathbf{B}\cdot {\mathbf{g}}_e\cdot \mathbf{S}-{\sum }_i{\gamma }_{n,i}\mathbf{B}\cdot {\mathbf{I}}_i+{\sum }_i\mathbf{S}\cdot {\mathbf{A}}_i\cdot {\mathbf{I}}_i,$

where $\mathbf{S}$ is the electron spin, ${\mathbf{I}}_i$ are the coupled nuclear spins (¹H and, when present, ¹³C), ${\mathbf{g}}_e$ is the electron $g$ tensor, and ${\mathbf{A}}_i$ are hyperfine tensors. For an encyclopedia entry this captures the essential physics: a Zeeman-split electron spin coupled to one or more long-lived nuclear memories.

The optical interface is mediated by a bound-exciton manifold with transition energy near the telecom O-band zero-phonon line. A minimal driven model is

$H_{\mathrm{opt}}\approx {\omega }_0|X⟩⟨X|+H_{\mathrm{spin}}^{(X)}+\Omega (t)(|g⟩⟨X|+|X⟩⟨g|),$

with ${\omega }_0/2\pi \approx 226$ THz (1326 nm). In real devices, orientation-dependent selection rules, excited-state structure, and spectral diffusion matter for entanglement generation and readout.

Motivation

• Telecom-native: The 1326 nm transition is already in the telecom O-band, easing fibre networking and chip-to-chip photonic interconnects.
• Silicon-native: The platform lives inside silicon and can leverage mature silicon photonics and semiconductor processing.
• Small local register: One defect can host an optically active electron spin plus nuclear-spin memory qubits.
• Modular architecture fit: Heralded remote entanglement offers a natural path to non-local interconnects between cryogenic modules.
• Networking first, then computation: T centres are especially compelling where communication, memory, and distributed gates matter simultaneously.

Experimental Status

Integrated telecom spin-photon interface — Bergeron et al. (2020):

• Established the silicon-integrated telecom photon-spin interface
• Optical transition near 1326 nm in the telecom O-band
• Electron-spin coherence on the millisecond scale in isotopically enriched ²⁸Si

Single-spin optical observation — Higginbottom et al. (2022):

• Optical observation of individual T-centre spins in silicon
• Cavity-coupled nanophotonic devices with single-spin readout capability

Waveguide integration — DeAbreu et al. (2023):

• Waveguide-integrated silicon T centres demonstrated
• Strengthened the case for scalable on-chip routing and collection

Distributed quantum computing proof-of-concept — Photonic Inc. (2024 preprint):

• Remote Bell-pair generation between T centres in separate cryostats connected by telecom fibre
• Measured Bell-pair fidelity reached 0.60(0.08) at a 40 ns time-bin width
• Reported a preliminary, post-selected teleported CNOT truth table over a selected basis set, rather than a full deterministic remote gate benchmark

Electrical triggering — Dobinson et al. (2025):

• Electrically triggered silicon spin-photon devices demonstrated
• Included electrically driven O-band emission and electrically generated single-photon / spin-photon functionality in T-centre nanophotonics

Key Metrics

Metric	Value	Notes	Fidelity reference
Emission wavelength	1326 nm	Telecom O-band zero-phonon line	Bergeron et al. 2020
Electron spin T₂	~2 ms	Reported in isotopically enriched ²⁸Si	Bergeron et al. 2020
Nuclear-spin memory	>1 s	Long-lived local memory register	Bergeron et al. 2020
Remote Bell-pair fidelity	0.60(0.08)	40 ns time-bin width, preliminary 2024 preprint	Photonic Inc. 2024
Teleported gate sequence	Demonstrated	Post-selected preliminary tCNOT truth table over a selected basis set	Photonic Inc. 2024
Electrical triggering	Demonstrated	Electrically driven T-centre spin-photon devices in silicon	Dobinson et al. 2025

Scaling Considerations

• Silicon photonics compatibility: Waveguide and cavity integration are real experimental milestones, not just a platform promise.
• Telecom networking advantage: Native O-band emission removes a major systems burden that many other solid-state spin-photon platforms carry.
• Distributed-computing relevance: The platform has already crossed from isolated emitter physics into remote-entanglement experiments, albeit still at proof-of-concept scale.
• Architectural caveat: Claims about qLDPC-friendly non-local connectivity are best viewed as architectural motivation, not yet as an experimentally validated fault-tolerant stack.
• Device roadmap: Electrical triggering is an important maturity step because it points toward more integrated cryogenic photonic hardware.

References

Foundational interface and control milestones

• L. Bergeron et al., “Silicon-Integrated Telecommunications Photon-Spin Interface,” PRX Quantum 1, 020301 (2020)
• D. B. Higginbottom et al., “Optical observation of single spins in silicon,” Nature 607, 266-270 (2022)
• A. DeAbreu et al., “Waveguide-integrated silicon T centres,” Optics Express 31, 15045-15057 (2023)

Distributed and device-integration milestones

• Photonic Inc., “Distributed Quantum Computing in Silicon” (2024 preprint) — remote Bell-pair generation and a preliminary post-selected teleported-gate sequence, arXiv:2406.01704
• M. Dobinson et al., “Electrically triggered spin-photon devices in silicon,” Nature Photonics (2025)

Linked Papers

• bergeron-2020-silicon-integrated-telecommunications
• higginbottom-2022-optical-observation-of-single-spins-in-silicon
• deabreu-2023-waveguide-integrated-silicon-t-centres
• photonic-2024-distributed-qc
• dobinson-2025-electrically-triggered-spin-photon-devices-in-silicon

Evergreen context

• divincenzo-criteria — T centres are notable because the same silicon defect platform addresses stationary-spin storage and optical communication, rather than bolting a separate transducer onto an otherwise disconnected qubit.
• coherence-time-hierarchy — the platform naturally fits a layered-memory picture: an optically active electron spin for interfacing, plus slower but much longer-lived local nuclear-spin storage.

Related Entries

• nv-center-qubit — diamond color center benchmark for spin-photon networking, but visible-wavelength rather than telecom-native
• siv-color-center-qubit — strong optical diamond color-center family with different coherence and wavelength tradeoffs
• silicon-spin-qubit — silicon spin hardware without the native telecom optical interface
• quantum-transduction — contrast case where frequency conversion is needed because the emitter is not already in the telecom band
• qldpc-codes — architectural context for why non-local interconnects are attractive
• dual-rail-photonic-qubit — photonic flying-qubit layer that can interface with remote entanglement protocols