The quantum revolution demands new alliances between old adversaries. Silicon—the workhorse of modern electronics—seems an unlikely partner for quantum photonics. Its indirect bandgap makes it a reluctant light emitter. Its lack of optical activity appears to disqualify it from the quantum stage. Yet today, silicon photonics stands poised to enable scalable quantum networks that could transform computing, communication, and our understanding of reality itself.

This transformation didn’t happen overnight. It required researchers to reimagine silicon‘s limitations as opportunities, to engineer defects with atomic precision, and to coax quantum behavior from a material that nature never intended for optical applications. The result is a technological tour de force: single atoms embedded in silicon crystals that can store quantum information, emit single photons on demand, and entangle with light across vast distances.

What follows is a deep exploration of this unlikely marriage between silicon and quantum light. We’ll examine how physicists have overcome fundamental barriers, what challenges remain, and why silicon quantum photonics might hold the key to practical quantum technologies. From the physics of spin-photon coupling to the economics of quantum startups, from philosophical puzzles about time and entanglement to practical questions about packaging and thermal management—this guide provides a comprehensive view of a field that sits at the intersection of fundamental science and transformative technology.

Whether you’re a researcher seeking technical insights, an investor evaluating quantum opportunities, or simply curious about how light and matter conspire to process information in ways classical physics forbids, this exploration offers both depth and perspective on one of the 21st century’s most promising technological frontiers.

The Complete Guide To Silicon Quantum Photonics: Light-Based Quantum Computing And Communication Technologies – From Wrestling With Silicon’s Indirect Bandgap To Contemplating Photons That Experience No Time

The marriage of silicon electronics with quantum photonics represents one of the most ambitious technological convergences of our time. Light, with its natural immunity to many forms of decoherence and ability to travel at ultimate speed, offers tantalizing advantages for quantum information. Yet silicon, with its indirect bandgap and lack of optical activity, seems an unlikely host for quantum photonics. The story of how researchers have coaxed silicon into the quantum optical age—through defect engineering, cavity design, and hybrid integration—demonstrates human ingenuity in overcoming fundamental material limitations.

Now, let’s answer your questions on silicon quantum photonics!

1. What are the physical mechanisms for achieving strong spin-photon coupling in silicon, given silicon’s indirect bandgap?

Silicon’s indirect bandgap presents a fundamental challenge for quantum photonics that has driven decades of creative solutions. In direct bandgap semiconductors like gallium arsenide, electrons can transition between conduction and valence bands by emitting or absorbing a photon while conserving crystal momentum. Silicon’s conduction band minimum and valence band maximum occur at different points in momentum space, requiring a phonon to conserve momentum during optical transitions. This three-particle process is inherently weak, with radiative lifetimes extending to milliseconds—far too slow for quantum applications.

The solution comes from defect engineering that creates localized states within silicon’s bandgap. These defect states break the crystal’s translational symmetry, relaxing momentum conservation requirements. A bound electron at a defect site has uncertainty in position, which by Heisenberg’s principle creates uncertainty in momentum sufficient to enable direct optical transitions. The challenge lies in creating defects with the right properties: strong optical transitions, minimal non-radiative decay, and spin-dependent optical selection rules for quantum applications.

Color centers—optically active point defects—have emerged as the leading platform for spin-photon coupling in silicon. These atomic-scale defects trap electrons in localized orbitals with discrete energy levels, similar to atoms but embedded in the silicon matrix. The T center, consisting of two carbon atoms and a hydrogen atom, exemplifies successful defect engineering. Its zero-phonon line at 1326 nm falls within the telecommunications band, enabling compatibility with existing fiber infrastructure. The T center’s spin states couple to optical transitions through spin-orbit interaction, enabling optical initialization and readout of quantum information.

The physics of spin-photon coupling in color centers relies on spin-selective optical transitions. In the simplest case, optical excitation promotes an electron from a spin-ground state to a spin-excited state, with selection rules determined by angular momentum conservation. The excited state can decay via photon emission, with the photon’s polarization entangled with the final spin state. This creates a quantum interface between stationary spin qubits and flying photonic qubits—essential for quantum networks.

Cavity quantum electrodynamics enhances the naturally weak coupling between spins and photons. By embedding color centers in photonic cavities with mode volumes approaching (λ/n)³, the local electromagnetic field intensity increases dramatically. The Purcell effect enhances spontaneous emission rates by factors exceeding 1000, reducing radiative lifetimes from microseconds to nanoseconds. Strong coupling regime, where spin-photon interaction exceeds decay rates, enables coherent quantum operations between spins and photons.

Recent developments in silicon photonic crystal cavities achieve quality factors exceeding 10⁶ with mode volumes below 0.1 (λ/n)³. These cavities concentrate electromagnetic energy to near the theoretical limit, enhancing light-matter interaction. Integration of color centers with such cavities has demonstrated Purcell factors exceeding 100, spin-photon entanglement with fidelities above 90%, and coherent spin control via cavity-mediated interactions. The combination of high-quality color centers and advanced cavity designs overcomes silicon’s indirect bandgap limitations.

Alternative approaches exploit electric dipole transitions in quantum dots or donors near silicon surfaces. While these lack the atomic-like optical transitions of color centers, they offer electrical tunability and compatibility with existing quantum dot architectures. Hybrid schemes combining quantum dots for computation with color centers for optical interface may provide optimal system architectures. The diversity of approaches reflects the richness of silicon as a quantum platform despite its apparent optical limitations.

2. How do T centers and other color centers in silicon compare to competing platforms for quantum photonic applications?

T centers in silicon have rapidly emerged as compelling candidates for quantum photonics, but they compete with established platforms like nitrogen-vacancy (NV) centers in diamond, quantum dots in III-V semiconductors, and rare-earth ions in crystals. Each platform offers distinct advantages and faces unique challenges. Understanding these trade-offs is crucial for predicting which technologies will dominate different quantum photonic applications.

The T center’s primary advantage lies in its telecommunications-wavelength emission at 1326 nm, enabling direct integration with existing fiber infrastructure without frequency conversion. This contrasts sharply with NV centers in diamond, which emit at 637 nm and require complex frequency conversion for fiber transmission. The T center’s narrow optical linewidth (< 1 GHz) approaches the transform limit, enabling high-fidelity photonic quantum gates. Recent measurements show optical coherence times exceeding 1 microsecond, sufficient for quantum networking protocols.

Fabrication scalability gives silicon color centers a significant advantage. T centers form through carbon implantation and annealing—processes compatible with standard semiconductor manufacturing. The ability to deterministically place T centers using focused ion beams or lithographic masks enables precise integration with photonic circuits. This contrasts with the stochastic formation of NV centers in diamond or the complex epitaxial growth required for III-V quantum dots. Silicon’s mature fabrication ecosystem allows rapid iteration and scaling to millions of devices.

However, T centers face challenges in optical efficiency and spin properties. The quantum efficiency for photon emission remains below 10%, limited by non-radiative decay channels and imperfect cavity coupling. NV centers achieve higher quantum efficiencies (up to 70% in optimized structures) and benefit from longer spin coherence times at room temperature. III-V quantum dots can achieve near-unity quantum efficiency and GHz emission rates but require cryogenic operation and suffer from spectral diffusion.

Other silicon color centers offer complementary properties. The G center (substitutional carbon-silicon vacancy) emits at telecommunications wavelengths with narrower linewidth than T centers but shows weaker spin-photon coupling. The W center (tri-interstitial silicon) provides bright emission but lacks suitable spin states for quantum information. Transition metal defects like chromium or molybdenum create centers with millisecond spin coherence but weaker optical transitions. This diversity allows optimization for specific applications.

Rare-earth ions in silicon represent another promising direction. Erbium ions emit at 1540 nm—the optimal telecommunications wavelength—with exceptional optical coherence. Recent breakthroughs achieving single erbium ion detection and control in silicon open paths to quantum repeaters and memories. The 4f electronic shells of rare-earth ions are shielded from the environment, providing hours-long spin coherence times. However, the weak oscillator strength requires ultra-high finesse cavities for efficient spin-photon coupling.

System-level considerations often dominate platform selection. Silicon color centers integrate naturally with silicon photonics and electronics, enabling monolithic quantum photonic processors. Diamond NV centers excel for sensing applications but require hybrid integration for scalable systems. III-V quantum dots offer superior performance but at higher cost and complexity. The optimal platform depends on whether the application prioritizes performance metrics (efficiency, coherence) or system metrics (scalability, integration).

The rapid progress in silicon color centers suggests they may dominate applications requiring large-scale integration: quantum repeater networks, distributed quantum computing, and integrated quantum photonic processors. Competing platforms will likely persist in specialized niches: diamond for sensing, III-V for single-photon sources, rare-earths for quantum memories. The future may see hybrid systems combining the best features of each platform. Silicon’s role as the integration platform—hosting both native color centers and heterogeneously integrated components—positions it centrally in the quantum photonic ecosystem.

3. What are the fundamental limits on photon emission rates and collection efficiencies in silicon-based single-photon sources?

The performance of single-photon sources ultimately determines the viability of quantum photonic technologies. In silicon-based systems, fundamental physical limits interact with practical engineering constraints to bound achievable emission rates and collection efficiencies. Understanding these limits—from radiative lifetime bounds to waveguide coupling physics—reveals both the potential and limitations of silicon quantum photonics.

The fundamental emission rate limit comes from the oscillator strength of the optical transition, governed by quantum mechanics and electromagnetism. For electric dipole transitions, the spontaneous emission rate follows Fermi’s golden rule: Γ = (ω³|μ|²)/(3πε₀ℏc³), where μ is the transition dipole moment. Color centers in silicon typically have dipole moments of 1-10 Debye, yielding natural lifetimes of 10-100 nanoseconds. This sets an upper bound on emission rates around 100 MHz without cavity enhancement.

Cavity quantum electrodynamics can enhance emission rates through the Purcell effect: F_P = (3Q/4π²)(λ/n)³/V, where Q is quality factor and V is mode volume. State-of-the-art silicon photonic crystal cavities achieve Q > 10⁶ and V < 0.1(λ/n)³, yielding Purcell factors exceeding 1000. This could theoretically enable GHz emission rates. However, practical limits arise from cavity fabrication tolerances, spectral alignment between emitter and cavity, and maintaining high Q while enabling photon extraction.

Collection efficiency—the fraction of emitted photons captured into useful modes—faces fundamental and practical constraints. In bulk silicon, total internal reflection at the silicon-air interface (critical angle 17°) limits collection to about 2% without photonic structures. Even perfect anti-reflection coatings and solid immersion lenses can only increase this to about 30% due to the high refractive index contrast. This fundamental constraint drives the need for integrated photonic approaches.

Photonic crystal cavities can achieve near-unity collection efficiency by engineering the local density of optical states. By creating a complete photonic bandgap except for the cavity mode, spontaneous emission is funneled exclusively into the designed mode. Theoretical β-factors (fraction of emission into the cavity mode) approaching 99% are possible. However, fabrication imperfections, absorption losses, and the need for outcoupling reduce practical collection efficiencies to 50-90%.

Waveguide-coupled architectures offer an alternative approach with different trade-offs. By placing emitters in single-mode waveguides, collection efficiencies exceeding 95% into the waveguide mode are achievable. The challenge becomes efficient coupling from waveguides to optical fibers or free space. Edge coupling suffers from mode mismatch and reflection losses, typically achieving 70-80% efficiency. Grating couplers enable wafer-scale testing but have limited bandwidth and efficiency.

Non-radiative decay presents a fundamental limit on quantum efficiency. Color centers in silicon compete with non-radiative channels including multi-phonon relaxation, Auger processes, and energy transfer to nearby defects. The quantum yield depends on temperature, local strain, and electromagnetic environment. Current T centers show quantum efficiencies around 5-10%, limited primarily by non-radiative decay. Understanding and eliminating these channels remains an active research area.

The interplay between emission rate and collection efficiency creates system-level trade-offs. High-Q cavities enhance emission rates but make photon extraction challenging. Low-Q cavities ease collection but provide minimal rate enhancement. Optimal designs balance these factors for specific applications. Quantum key distribution prioritizes high rates over perfect efficiency, while quantum computing demands high-fidelity photon-spin entanglement even at lower rates.

Ultimate limits may require new approaches. Plasmonic enhancement could achieve stronger field confinement than dielectric cavities but faces high absorption losses. Metamaterial designs might engineer emission patterns for optimal collection. Integration with superconducting single-photon detectors on the same chip eliminates fiber coupling losses. The path to practical silicon single-photon sources requires co-optimization across materials, devices, and systems—pushing multiple boundaries simultaneously rather than optimizing individual metrics.

4. What are the technical challenges in integrating silicon photonic components (waveguides, resonators, modulators) with quantum emitters?

The integration of quantum emitters with silicon photonic circuits represents a complex engineering challenge that spans materials science, nanofabrication, and quantum optics. While silicon photonics has achieved remarkable success in classical communications, adding quantum functionality requires unprecedented precision in placement, spectrum alignment, and environmental control. These challenges multiply when scaling from single devices to integrated quantum photonic processors.

Spatial alignment between quantum emitters and photonic structures demands nanometer precision. A color center must be positioned within the optical mode maximum to achieve strong coupling—typically requiring placement accuracy better than 50 nm in all three dimensions. Deterministic positioning techniques include focused ion beam implantation, atomic force microscope manipulation, and lithographically defined implantation masks. However, ion straggling, diffusion during annealing, and measurement uncertainty limit practical placement accuracy to 10-50 nm.

Spectral alignment presents equal challenges. Quantum emitters have transition energies determined by local strain, electric fields, and defect configuration, leading to inhomogeneous broadening across devices. Meanwhile, photonic cavities have resonances determined by geometry with fabrication variations of ±1 nm causing GHz-scale frequency shifts. Achieving spectral overlap between randomly distributed emitter frequencies and cavity resonances requires either extensive pre-screening or active tuning mechanisms.

Thermal management becomes critical when integrating active components. Thermo-optic modulators, common in silicon photonics, create temperature gradients that shift both emitter frequencies and cavity resonances. A 1 K temperature change shifts silicon’s refractive index by 2×10⁻⁴, detuning cavities by hundreds of linewidths. Quantum emitters show temperature-dependent emission wavelengths and linewidths. Maintaining spectral alignment during device operation requires sophisticated thermal engineering and potentially active stabilization.

Electrical isolation while maintaining optical integration challenges device design. Many quantum emitters require electrical control for initialization, manipulation, or tuning. However, metal electrodes near photonic structures introduce optical losses and modify electromagnetic modes. Transparent conductive oxides offer partial solutions but have higher resistance and limited cryogenic performance. Three-dimensional integration with electrical layers separated from optical layers shows promise but increases fabrication complexity.

Material compatibility constraints limit process choices. Quantum emitter formation often requires high-temperature annealing (>1000°C) that can damage photonic structures or cause unwanted diffusion. Conversely, photonic device fabrication may involve plasma etching or chemical treatments that modify emitter properties. Developing process flows that preserve both emitter quality and photonic performance requires careful sequencing and protective strategies.

The heterogeneous integration of different material systems adds another dimension of complexity. While silicon excels for passive photonics, optimal quantum emitters might exist in other materials—diamond for NV centers, or III-V semiconductors for quantum dots. Bonding, transfer printing, or pick-and-place assembly can combine materials, but maintaining optical alignment and quality across material interfaces remains challenging. Thermal expansion mismatches create strain and potential delamination during temperature cycling.

Scaling to many-emitter systems multiplies these challenges. Quantum photonic circuits for computing or communication require arrays of identical emitters with uniform coupling to photonic structures. Statistical variations in emitter properties, placement accuracy, and fabrication tolerances create a yield problem similar to semiconductor manufacturing but with tighter requirements. Redundancy, error correction, and post-fabrication trimming strategies borrowed from classical electronics may prove essential.

Recent successes demonstrate feasible integration paths. Hybrid approaches combining top-down lithography with bottom-up self-assembly show promise for scalable emitter placement. In-situ tuning using electrical, strain, or optical fields can compensate for spectral misalignment. Machine learning algorithms optimize device operation across multiple parameters simultaneously. The challenge shifts from perfect fabrication to intelligent system design that accommodates imperfection while maintaining quantum functionality. This paradigm shift—from fighting variability to managing it—may prove key to practical quantum photonic integration.

5. How can we achieve efficient fiber-to-chip coupling while maintaining the coherence properties of quantum states?

The interface between optical fibers and silicon photonic chips represents a critical bottleneck in quantum photonic systems. Unlike classical communications where power loss merely reduces signal amplitude, quantum applications require preserving delicate superposition states and entanglement during the coupling process. Any perturbation that distinguishes between quantum states—whether from reflection, scattering, or mode conversion—can destroy quantum coherence. Achieving efficient, coherence-preserving coupling demands innovations in design, fabrication, and packaging.

The fundamental challenge stems from mode size mismatch. Single-mode optical fibers support gaussian modes with 8-10 μm diameter, while silicon waveguides confine light to submicron dimensions due to high index contrast. Direct butt-coupling between fiber and waveguide achieves only 10-20% efficiency due to mode overlap mismatch. This 7-10 dB insertion loss is unacceptable for quantum applications where every photon counts and photon loss can break entanglement.

Edge coupling with inverse tapers offers one solution. By gradually tapering the silicon waveguide to nanoscale tips (< 200 nm), the optical mode expands to better match the fiber mode. State-of-the-art inverse tapers achieve 1-2 dB coupling loss (60-80% efficiency) with broad bandwidth. However, the fragile nanoscale tips are susceptible to damage and contamination. More critically for quantum applications, surface roughness on the taper can cause polarization mixing and decoherence of polarization-encoded quantum states.

Grating couplers enable wafer-scale testing but face limitations for quantum applications. By diffracting light vertically, gratings allow coupling from above the chip rather than the edge. Optimized gratings achieve 3-6 dB coupling efficiency with easier alignment tolerances. However, gratings are inherently polarization-sensitive and wavelength-specific. For quantum states encoded in polarization or frequency, gratings can act as which-path detectors that destroy superposition. Advanced 2D grating designs partially address these issues but increase complexity.

The preservation of quantum coherence during coupling requires attention to multiple decoherence mechanisms. Polarization mode dispersion in fibers and waveguides can separate horizontally and vertically polarized components, destroying polarization entanglement. Careful design of waveguide geometry for polarization-independent propagation is essential. Chromatic dispersion affects time-bin encoded states, requiring dispersion compensation. Even mechanical vibrations coupling through the fiber can introduce phase noise that degrades interference visibility.

Three-dimensional integration approaches show promise for improved coupling. By fabricating polymer or silicon nitride waveguides with intermediate index and mode size on top of silicon photonic circuits, multi-stage mode conversion achieves better impedance matching. These approaches have demonstrated sub-1 dB coupling efficiency while maintaining polarization. The additional fabrication complexity is justified by improved quantum state fidelity.

Advanced packaging solutions address mechanical and environmental stability. Silicon photonic chips wire-bonded or flip-chip bonded to carrier substrates provide strain relief and thermal management. Fiber arrays permanently attached using UV-cured epoxy or laser welding ensure stable alignment. However, differential thermal expansion between silicon (2.6 ppm/K) and fiber (0.5 ppm/K) creates stress during temperature cycling. Low-stress die attach materials and kinematic mounting schemes help maintain alignment.

Active alignment and stabilization may prove necessary for the most demanding quantum applications. Piezoelectric fiber positioners can maintain sub-nanometer alignment stability. On-chip Mach-Zehnder interferometers monitor coupling efficiency and phase, enabling feedback control. For polarization-sensitive applications, on-chip polarization controllers using thermal or carrier-injection phase shifters can compensate for fiber birefringence. These active approaches trade complexity for performance.

The ultimate solution may be eliminating fiber coupling entirely through full integration. If quantum sources, processors, and detectors all reside on the same chip, fiber interfaces are needed only for long-distance quantum communication. This vision requires continued advancement in integrated quantum photonics but promises to sidestep coupling challenges. Until then, the fiber-chip interface remains a critical engineering challenge where classical solutions must be adapted to preserve quantum coherence—a reminder that quantum technologies demand rethinking established approaches at every level.

6. What packaging and thermal management solutions are needed for integrated silicon quantum photonic chips?

Packaging quantum photonic chips presents unique challenges that go beyond conventional semiconductor packaging. These devices must operate at cryogenic temperatures while maintaining optical, electrical, and mechanical connections to the outside world. Thermal cycling between room temperature and millikelvin environments creates extreme mechanical stresses. Meanwhile, quantum coherence demands isolation from electromagnetic interference, mechanical vibration, and thermal fluctuations. Success requires rethinking packaging from first principles with quantum requirements in mind.

Thermal management begins with understanding heat loads and cooling requirements. Quantum emitters and detectors often require millikelvin temperatures for optimal performance, while integrated electronics may operate at 4 K or above. This temperature gradient across a single chip creates thermoelectric effects and differential thermal contraction. Heat loads from optical absorption (typically microwatts per photon at GHz rates), electrical dissipation in control circuits, and thermal radiation from warmer stages must all be managed within the microwatt cooling power of dilution refrigerators.

Materials selection for quantum packaging demands careful consideration of thermal and mechanical properties. Traditional semiconductor packages use materials optimized for room temperature operation. At cryogenic temperatures, differential thermal contraction can generate gigapascal-level stresses. Silicon contracts by 0.23% cooling to 4 K, while copper contracts by 0.32%. These mismatches can crack die, delaminate bonds, or shift optical alignment. Materials like Invar, silicon carbide, or matched glass-ceramics minimize thermal stress but may compromise thermal conductivity or electromagnetic shielding.

The optical packaging must maintain fiber alignment across temperature cycles while minimizing heat conduction. V-groove arrays etched in silicon provide self-aligned fiber placement matched to the chip’s thermal expansion. However, optical adhesives typically become brittle at cryogenic temperatures and may crack or delaminate. Alternative approaches use mechanical clamping, metallic solders, or glass frit bonding. Each method trades optical performance against reliability and thermal conductivity.

Electrical interconnects face competing requirements of high bandwidth and low thermal conductivity. Quantum control requires numerous high-frequency connections—typically 10-100 per qubit for full control. Standard coaxial cables conduct too much heat from room temperature. Superconducting niobium-titanium coax provides low loss below the critical temperature but requires careful thermal anchoring. Flexible superconducting circuits on polyimide enable high-density interconnects with manageable heat loads. The transition from normal metals to superconductors must be carefully managed to avoid junction heating.

Electromagnetic shielding becomes critical when single photons and individual electron spins carry quantum information. Magnetic shielding using mu-metal or superconducting shields protects against external fields that could cause decoherence. Radio frequency interference from control electronics requires careful filtering and shielding. Even Johnson noise from resistive elements can inject photons that trigger false detector counts. Multi-layer shielding strategies combining magnetic, electric, and superconducting shields provide the ultra-quiet environment quantum devices require.

Vibration isolation prevents mechanical noise from destroying quantum coherence. Building vibrations, pulse tube cooler oscillations, and acoustic noise couple into the package, creating time-varying strain fields. These strains modulate optical cavity frequencies and emitter energies, causing decoherence. Soft suspension systems, mass-spring dampers, and active vibration cancellation all play roles. The package itself must avoid mechanical resonances in the frequency range of quantum operations (typically MHz to GHz).

Modular packaging architectures enable testing and replacement of components. Unlike classical integrated circuits where packaging is permanent, quantum devices may require iterative optimization. Demountable optical connectors, zero-insertion-force electrical sockets, and kinematic mounting systems allow component swapping. This modularity comes at the cost of increased complexity and potential reliability issues. The trade-off between integration and flexibility remains an active area of development.

Recent packaging innovations show promising directions. Three-dimensional integration using through-silicon vias enables separation of quantum and classical layers with minimal interconnect length. Photonic wire bonding creates low-loss optical connections between chips. Advanced thermal interface materials like synthetic diamond heat spreaders provide excellent thermal conductivity with matched expansion. As quantum photonic systems scale from laboratory demonstrations to practical devices, packaging co-design with chips and systems becomes essential. The future belongs to holistic approaches that consider quantum requirements from the earliest design stages rather than adapting classical solutions.

7. What is the competitive landscape for silicon quantum photonics companies, and how do they differentiate their approaches?

The silicon quantum photonics landscape features a diverse ecosystem of startups, established semiconductor companies, and research institutions racing to commercialize quantum optical technologies. Unlike quantum computing where a few large players dominate, quantum photonics has spawned numerous specialized companies targeting different applications and technology approaches. Understanding their differentiation strategies reveals both the opportunities and challenges in this emerging market.

PsiQuantum stands out as the most ambitious and well-funded player, having raised over $665 million to build a million-qubit photonic quantum computer. Their approach uses silicon photonics for all quantum operations—single-photon sources, interferometers for gates, and integrated detectors. The key differentiation is architectural: they pursue fusion-based quantum computing that trades lower efficiency for better error tolerance and manufacturability. By partnering with GlobalFoundries, they leverage existing semiconductor fabs rather than building custom facilities. Their strategy targets fault-tolerant quantum computing directly, skipping the NISQ era entirely.

Xanadu takes a contrasting approach with continuous-variable quantum computing using squeezed light states. Their silicon photonic chips implement gaussian boson sampling and other algorithms suited to photonic hardware. By focusing on near-term applications in drug discovery and finance, they’ve achieved early revenue while building toward universal quantum computing. Their recent demonstration of quantum computational advantage with 216 squeezed-state modes validates the approach. The differentiation lies in choosing quantum algorithms naturally suited to photonics rather than forcing photons to emulate qubits.

Nu Quantum, spun out from Cambridge University, focuses on quantum networking components rather than computing. Their silicon photonic chips integrate single-photon sources and detectors optimized for quantum communication. By targeting the quantum internet market—connecting quantum computers and enabling quantum key distribution—they avoid direct competition with computing-focused companies. Their modular component approach allows integration with various quantum computing platforms, positioning them as an enabling technology provider.

Sparrow Quantum specializes in deterministic single-photon sources using quantum dots in photonic crystal structures. While not strictly silicon photonics, their InGaAs (Indium–Gallium–Arsenide) quantum dots could be integrated with silicon photonic circuits. They’ve achieved record-breaking photon indistinguishability (>99%) and efficiency (>98%), crucial for photonic quantum computing. Their differentiation is performance—providing the highest quality photon sources even at premium prices for applications where fidelity matters more than cost.

Traditional semiconductor companies bring different strengths to quantum photonics. Intel leverages its silicon fabrication expertise to integrate quantum dots with photonic circuits on the same chip. Their approach emphasizes manufacturability and scaling, using standard CMOS processes where possible. IBM explores silicon photonics for quantum interconnects, linking superconducting quantum processors optically. These incumbents benefit from deep process knowledge and capital resources but may lack the focus and agility of startups.

Chinese companies like Quantum CTek and Origin Quantum pursue integrated strategies combining quantum communication, computing, and sensing. Backed by government initiatives, they benefit from patient capital and protected markets. Their differentiation often emphasizes complete quantum systems rather than components, targeting domestic infrastructure projects. The geopolitical dimension adds complexity to international competition and collaboration.

Academic spinouts continue to emerge with novel approaches. Oxford‘s Orca Computing uses temporal encoding of photons for quantum computing, requiring fewer physical resources. MIT‘s lightmatter targets hybrid classical-quantum processors for machine learning. Each brings unique IP from university research, often focusing on specific technical breakthroughs that larger companies might acquire or license.

The competitive dynamics favor different strategies at different scales. Component suppliers can succeed with specialized products serving multiple customers. Platform companies need massive capital to reach meaningful qubit counts. Application-specific companies can find niches in optimization, simulation, or communication. The market remains fluid enough that multiple winners are likely, each serving different segments.

Partnerships and ecosystems play crucial roles in differentiation. Companies aligning with major cloud providers gain distribution advantages. Those partnering with telecom companies access quantum communication markets. Semiconductor foundry relationships enable scaling. The ability to navigate these partnerships while maintaining technological differentiation often determines success. As the market matures, we expect consolidation around platforms while specialized component suppliers thrive in niches—similar to the classical semiconductor industry structure.

8. Which market applications (quantum communication, computing, sensing) are likely to commercialize first for silicon photonics?

The commercialization timeline for silicon quantum photonics varies dramatically across applications, driven by technical maturity, market readiness, and competitive alternatives. While quantum computing captures headlines and investment, nearer-term applications in communication and sensing may provide the revenue streams that sustain the industry through longer development cycles. Understanding these market dynamics helps predict which applications will drive early adoption and generate returns for investors.

Quantum communication represents the nearest-term commercial opportunity for silicon photonics. Quantum key distribution (QKD) systems already operate commercially, though mostly using discrete components rather than integrated photonics. Silicon photonic integration could reduce QKD system costs by 10-100x while improving reliability. The market drivers are compelling: rising cybersecurity threats, quantum computing’s threat to classical encryption, and government mandates for quantum-safe communications. Banks, governments, and critical infrastructure operators represent early adopters willing to pay premiums for provable security.

Silicon photonics offers specific advantages for quantum communication. Integrated interferometers enable compact QKD transmitters and receivers. On-chip single-photon detectors eliminate fiber coupling losses. Wavelength division multiplexing allows classical and quantum channels on the same fiber. Recent demonstrations achieve megabit-per-second quantum key rates over metropolitan distances. The main barriers are cost relative to classical encryption and the need for dedicated fiber infrastructure. We expect commercial deployment in high-value point-to-point links within 2-3 years, expanding to network applications within 5-7 years.

Quantum sensing applications present mixed commercialization prospects. Silicon photonic sensors could detect magnetic fields, accelerations, or chemical signatures with quantum-enhanced precision. However, competing platforms often better suit specific sensing applications. Nitrogen-vacancy centers in diamond excel for magnetometry, atoms provide ultimate gravimeters, and specialized materials optimize chemical detection. Silicon photonics may find niches in integrated sensing where size and power matter more than ultimate sensitivity—such as biomedical implants or distributed sensor networks.

Quantum computing faces the longest commercialization timeline but largest potential market. Photonic quantum computers must scale to millions of qubits for practical advantage over classical systems. Current demonstrations remain below 100 photonic qubits equivalent. The technological challenges—efficient sources, low-loss circuits, fast detectors—require continued innovation. However, photonic approaches offer unique advantages: room-temperature operation for some components, natural error correction schemes, and potential for extreme scaling using foundry manufacturing.

The quantum computing market will likely segment by application. Gaussian boson sampling and other photonic-native algorithms may find near-term uses in drug discovery or materials design. These applications leverage what photons do naturally rather than forcing them to emulate other qubit types. Fault-tolerant photonic quantum computers remain 10-15 years from commercial viability but could ultimately dominate due to manufacturing scalability. Early revenue from specialized applications funds the longer journey to universal quantum computing.

Hybrid classical-quantum applications may provide stepping stones to full quantum systems. Silicon photonics excels at generating and manipulating classical light for machine learning accelerators. Adding quantum light sources and detectors enables quantum-enhanced machine learning with nearer-term commercial viability. Similarly, quantum random number generators using silicon photonics provide certified randomness for cryptography and simulations. These boundary applications generate revenue while advancing quantum technology.

Market adoption depends critically on competing technologies. Quantum communication competes with post-quantum cryptography—mathematical approaches that resist quantum attack. If standardization bodies mandate hardware-based quantum security, silicon photonics wins. If software solutions suffice, the market remains niche. For quantum computing, superconducting and trapped ion systems have 5-10 year leads. Silicon photonics must offer compelling advantages—likely in scaling or networking—to capture significant market share.

Geographic factors influence commercialization timelines. China‘s quantum communication network drives domestic demand for integrated photonic components. European quantum initiatives fund pre-commercial deployments. The US focuses on quantum computing with communication as supporting infrastructure. Companies must navigate these regional differences in market priorities and regulatory environments. Silicon photonics’ compatibility with existing semiconductor supply chains provides advantages in all regions.

The most likely scenario sees quantum communication achieving modest commercial success within 5 years, providing revenue to sustain the industry. Specialized quantum computing applications emerge in 5-10 years for problems naturally suited to photonics. General-purpose photonic quantum computers remain 15+ years from commercialization but could ultimately represent the largest market. Success requires patient capital, realistic expectations, and focus on nearer-term applications that fund long-term development. Silicon photonics companies that balance vision with pragmatism—pursuing grand challenges while capturing interim markets—will most likely survive the journey from laboratory to marketplace.

9. What are the key partnerships between silicon photonics startups and established semiconductor or telecom companies?

Strategic partnerships between quantum photonics startups and established industry players have become crucial for transforming laboratory demonstrations into commercial products. These collaborations combine startup innovation with incumbent resources—manufacturing capacity, distribution channels, and customer relationships. The structure and success of these partnerships often determine which technologies reach market and which remain academic curiosities. Understanding these alliances reveals the emerging quantum photonics ecosystem.

The PsiQuantum-GlobalFoundries partnership represents the most ambitious collaboration in quantum photonics. PsiQuantum designs photonic quantum processors manufactured in GlobalFoundries’ semiconductor fabs, leveraging billions of dollars of existing infrastructure. This partnership solves multiple challenges: PsiQuantum avoids capital-intensive fab construction, GlobalFoundries diversifies beyond classical semiconductors, and both share the risks of developing novel technologies. The collaboration extends beyond simple foundry services to process co-development, with teams working together to adapt manufacturing for quantum requirements.

Intel‘s collaboration with QuTech exemplifies the research partnership model. Rather than acquiring quantum expertise, Intel funds collaborative research while contributing semiconductor manufacturing knowledge. QuTech provides quantum algorithms and architectures; Intel develops the silicon hardware. This structure allows both partners to focus on core competencies while sharing intellectual property and publications. The decade-long commitment demonstrates the patient capital required for quantum development.

Telecommunications companies pursue different partnership strategies focused on quantum networking. British Telecom collaborates with multiple quantum startups to test quantum key distribution over existing fiber infrastructure. These partnerships are typically non-exclusive, with telecoms hedging bets across multiple technical approaches. Toshiba‘s partnership with BT has progressed furthest, with commercial QKD services operating on dedicated fibers. The telecom industry’s experience with photonics and networks provides natural synergy with quantum communication.

Manufacturing partnerships extend beyond pure foundries. IMEC, the Belgian semiconductor research center, operates an open innovation model where startups access advanced fabrication tools and expertise. Companies like Miraex and SiPhox use IMEC facilities to prototype quantum photonic devices without massive capital investment. This shared infrastructure model accelerates development while spreading costs across multiple users. IMEC benefits by staying at the forefront of emerging technologies.

Equipment suppliers form another category of strategic partners. Applied Materials, traditionally focused on semiconductor manufacturing equipment, partners with quantum companies to develop specialized tools for quantum device fabrication. These partnerships often involve co-development agreements where equipment is optimized for quantum requirements—such as ultra-low damage etching for maintaining qubit coherence. Equipment suppliers gain early access to emerging markets while quantum companies get customized tools.

The partnership landscape reveals different strategic approaches. Exclusive partnerships provide deeper collaboration but limit flexibility—PsiQuantum cannot easily switch from GlobalFoundries without massive redesign. Non-exclusive collaborations preserve options but may lack commitment for difficult development challenges. Joint ventures create aligned incentives but require complex governance. Each structure suits different stages of technology development and market maturity.

Asian partnerships show distinct characteristics shaped by government industrial policy. Samsung’s collaboration with quantum startups receives government backing as part of South Korea’s quantum initiative. These partnerships often include technology transfer requirements and domestic manufacturing commitments. Chinese partnerships between startups and state-owned enterprises blur the lines between commercial and strategic national interests. Understanding these dynamics is crucial for international quantum photonics companies.

Recent partnership trends show movement toward vertical integration. Rather than pure supplier relationships, companies seek partnerships spanning the quantum stack—from materials through devices to systems. IBM’s Quantum Network includes quantum photonics companies developing components for future quantum systems. These ecosystem partnerships recognize that quantum advantage requires innovation at every level, not just individual components.

Failed partnerships provide cautionary lessons. Several early quantum-classical company alliances dissolved when timeline mismatches became apparent—startups needing patient development capital while corporations demanded quarterly progress. Cultural differences between move-fast startup environments and risk-averse corporate cultures created friction. Successful partnerships now include explicit provisions for long development timelines and staged milestones.

The future of quantum photonics depends critically on these partnerships evolving from transactional relationships to true collaborations. As technologies mature, we expect consolidation through acquisitions and deeper integration. Established companies that successfully partner with quantum innovators position themselves for the next computing revolution. Startups that navigate these relationships while maintaining technological differentiation can access resources impossible to develop independently. The partnership dance between quantum visionaries and industry giants will ultimately determine which quantum futures become reality.

10. What are the philosophical implications of using photons (which experience no proper time) as carriers of quantum information between silicon-based qubits that exist in time?

The use of photons as quantum information carriers between time-bound matter qubits creates a profound philosophical puzzle at the intersection of relativity and quantum mechanics. Photons, traveling at the speed of light, experience no passage of proper time according to special relativity—their worldlines have zero proper length. Yet these timeless entities successfully mediate quantum information transfer between silicon qubits firmly embedded in the temporal flow. This paradox illuminates deep questions about the nature of time, information, and quantum reality.

From the photon’s reference frame (technically undefined but conceptually approached in the limit), emission and absorption occur simultaneously. The photon does not experience a journey through space or duration in time—it simply connects two spacetime events. This raises the question: how can something that exists outside time carry time-dependent quantum information? The phase relationships and coherence properties essential to quantum information seem to require temporal evolution, yet the carrier experiences no time in which this evolution could occur.

The resolution may lie in recognizing that quantum information exists in the correlations between reference frames rather than in any single perspective. The silicon qubits and measurement apparatus exist in timelike reference frames where the photon’s journey has definite duration. The quantum information is encoded not in the photon’s non-existent experience but in the relationships between emission and absorption events as observed from timelike frames. The photon serves as a bridge between temporal domains without itself participating in temporal flow.

This perspective suggests profound implications for quantum information theory. If information can be carried by entities experiencing no proper time, then information itself may be more fundamental than temporal evolution. The quantum state of the photon—its polarization, frequency, and phase—remains well-defined in timelike reference frames even though the photon itself has no temporal experience. This hints at a block universe view where quantum correlations exist as timeless relationships in spacetime rather than evolving dynamical properties.

The interface between timeless photons and temporal qubits occurs at the moment of emission and absorption. A silicon qubit preparing a photonic state imprints its temporal quantum evolution onto something that will experience no time. Upon absorption, another temporal qubit extracts this information, reintegrating it into the timelike domain. These boundary events where time meets timelessness may be fundamental to understanding quantum mechanics’ relationship with relativity.

Consider quantum entanglement distributed via photons. Two silicon qubits become entangled through photon exchange, creating correlations that persist regardless of spatial separation. From the photon’s perspective, it simultaneously connects both qubits, suggesting entanglement is not maintained “during” the photon’s flight but exists as an atemporal fact about the quantum state. This view aligns with interpretations where entanglement represents information about global quantum states rather than physical connections.

The practical success of photonic quantum information transfer despite these conceptual puzzles suggests our intuitions about time may be incomplete. Perhaps the flow of time emerges from quantum information processing rather than serving as its backdrop. Silicon qubits process information through temporal evolution; photons connect information across spacetime without temporal experience. Together they form a complete quantum information system that transcends purely temporal or atemporal descriptions.

These philosophical questions have practical implications for quantum technology development. If photonic links truly enable atemporal quantum connections, then quantum networks might achieve forms of coordination impossible in purely timelike systems. Quantum error correction using photonic links might leverage this temporal transcendence to protect information in ways matter-based systems cannot. The design of quantum repeaters and memories must account for the transition between temporal and atemporal information encoding.

The deeper question remains whether this temporal dichotomy reveals something fundamental about reality or merely reflects the limitations of our theoretical frameworks. As we build increasingly sophisticated quantum systems combining matter and light, we serve as empirical philosophers, probing the nature of time and information through technological exploration. The success or failure of different quantum architectures may ultimately inform our understanding of time itself—making quantum engineering a form of experimental metaphysics.

11. When we entangle photons with matter-based qubits in silicon, what does this hybrid light-matter system tell us about the unity or disunity of physical phenomena?

The creation of entanglement between photons and silicon-based qubits represents one of the most profound demonstrations of quantum mechanics’ unifying power. These hybrid systems bridge seemingly disparate physical realms—massless bosons traveling at light speed entangled with massive fermions bound in crystal lattices. The successful generation and manipulation of such light-matter entanglement reveals deep truths about the underlying unity of physical phenomena while highlighting persistent dualities in our understanding of nature.

At the operational level, photon-matter entanglement demonstrates the universality of quantum mechanics. The same mathematical framework—Hilbert spaces, unitary evolution, measurement postulates—describes both photonic and electronic systems. When a silicon quantum dot emits a photon entangled with its spin state, the combined system exists in a superposition that cannot be factored into separate light and matter components. This inseparability at the quantum level suggests a fundamental unity beneath the apparent diversity of physical phenomena.

Yet the details of achieving light-matter entanglement reveal persistent dualities. Photons and electrons follow different statistics—bosonic versus fermionic. They couple to different fields—electromagnetic versus electronic. They exhibit different relationships with spacetime—null versus timelike worldlines. Creating entanglement requires careful engineering of interaction Hamiltonians that bridge these differences. The necessity of such engineering might suggest that unity is imposed rather than discovered.

The hybrid nature of these systems illuminates the contextual nature of physical properties. A silicon spin qubit exhibits properties we associate with matter: mass, locality, temporal persistence. The entangled photon shows characteristics of radiation: masslessness, wavelike propagation, polarization. Yet in the entangled state, these properties become intertwined. Measuring the photon’s polarization instantaneously determines the electron’s spin, suggesting that “matter-like” and “light-like” are not fundamental categories but contextual descriptions.

Consider the implications for quantum field theory, our most fundamental description of nature. In QFT, both photons and electrons arise as excitations of underlying quantum fields. The electromagnetic and electron fields are distinct but can interact through coupling terms in the Lagrangian. Light-matter entanglement manifests these couplings, showing that the fields are not independent but form an interconnected whole. The apparent separation between light and matter emerges only in limiting cases where coupling can be neglected.

The technological success of hybrid systems provides empirical evidence for theoretical unification. Quantum transduction—converting quantum information between different physical carriers—works precisely because quantum mechanics provides a common language. A quantum state can be mapped from electron spin to photon polarization to nuclear spin without loss of quantum coherence (in principle). This fungibility of quantum information across physical platforms suggests information, not matter or energy, as the fundamental currency of physics.

However, the practical challenges of maintaining entanglement across light-matter interfaces reveal remaining disunities. Decoherence rates differ dramatically—photons maintain coherence over kilometers while electron spins decohere in microseconds. Energy scales vary by orders of magnitude. The engineering required to bridge these differences—cavities for enhanced coupling, cryogenic environments for noise suppression, precise timing control—highlights the work needed to reveal underlying unity.

The philosophical implications extend to interpretations of quantum mechanics. In Many Worlds, light-matter entanglement creates branching across different substrates, suggesting the universality of quantum splitting. In collapse theories, the different masses of photons and electrons might trigger different collapse dynamics, revealing disunity. Relational interpretations emphasize that properties emerge through interactions, making light-matter entanglement the fundamental reality rather than separate light and matter entities.

These hybrid quantum systems serve as experimental probes of nature’s deep structure. Each successful demonstration of light-matter entanglement provides evidence for unification while revealing new challenges. As we push toward more complex hybrid systems—multiple photons entangled with multiple matter qubits, continuous variable entanglement, hybrid error correction—we explore the boundaries of quantum unity. The ultimate question remains whether perfect unification is achievable or whether irreducible dualities persist at nature’s foundations. Our silicon-photonic quantum devices serve as both technological tools and philosophical instruments, probing the unity of physical law through the practical act of quantum engineering.

12. Does the wave-particle duality of photons in silicon waveguides provide new insights into the measurement problem, especially when photons mediate quantum state transfer?

The behavior of photons in silicon waveguides provides a unique experimental platform for probing wave-particle duality and its relationship to quantum measurement. As photons propagate through carefully engineered silicon structures, they exhibit wavelike interference in Mach-Zehnder interferometers while manifesting particle-like detection events at single-photon detectors. When these same photons mediate quantum state transfer between silicon qubits, they embody the measurement problem in its full complexity—serving simultaneously as quantum systems and measurement apparatus.

In silicon waveguides, the wave nature of photons manifests through their electromagnetic mode structure. The photon’s wavefunction extends across the waveguide cross-section, with evanescent tails penetrating the surrounding medium. This delocalized wave nature enables phenomena like directional coupling, where photons tunnel between adjacent waveguides with probability determined by wave overlap. Yet when detected, the photon registers as a localized click at a specific detector, never partially at multiple locations. This transition from extended wave to point detection occurs within the same integrated device, bringing the measurement problem into sharp focus.

The mediation of quantum state transfer adds new dimensions to wave-particle considerations. When a silicon qubit emits a photon carrying its quantum state, the photon must maintain wave-like coherence to preserve quantum information. Phase relationships encode qubit states in photonic degrees of freedom—polarization, time-bin, or path. Any measurement that reveals particle-like properties (which path, arrival time) destroys these phase relationships and breaks entanglement. The photon must remain in wave-like superposition until absorbed by the target qubit.

Silicon photonic circuits make this duality experimentally accessible with unprecedented control. Beam splitters create superposition states where single photons exist in wavelike superposition across multiple paths. Phase shifters manipulate these superpositions, enabling quantum gates. Yet single-photon detectors register discrete clicks, not continuous fields. By adjusting the measurement basis—choosing which outputs to detect—experimenters select whether to observe wave-like interference or particle-like which-path information. The same physical system exhibits either aspect depending on measurement choice.

The integrated nature of silicon photonics provides new perspectives on the measurement problem’s locality. In traditional optics, measurement devices are macroscopic and separated from the quantum system. In silicon photonics, nanoscale detectors integrate directly with waveguides on the same chip. The boundary between quantum system and measurement apparatus becomes blurred—both are quantum mechanical structures differing only in coupling to the environment. This suggests measurement emerges from environmental entanglement rather than fundamental system-apparatus distinction.

Recent experiments demonstrate measurement backaction in silicon photonic systems with striking clarity. Weak measurements that extract partial which-path information show proportional reduction in interference visibility. The more particle-like information gained, the less wave-like interference remains. This trade-off appears fundamental, quantified by information-theoretic bounds. Silicon photonics enables precise control of measurement strength through tunable coupling, mapping the entire spectrum from weak to projective measurement.

The role of photons in mediating entanglement between matter qubits illuminates measurement from another angle. The photon carries quantum information between silicon qubits without itself being measured. Only upon absorption does measurement occur, collapsing the joint light-matter state. This delayed choice aspect—where measurement timing determines whether wave or particle properties manifest—suggests the photon exists in genuine superposition until interaction forces definiteness.

Silicon photonics also enables exploration of exotic measurement schemes. Quantum non-demolition measurements detect photon presence without destroying the quantum state. Homodyne detection measures continuous variables of the electromagnetic field. These alternatives to simple photon counting reveal the richness of quantum measurement beyond binary wave-particle choices. The measurement problem appears not as a single puzzle but a spectrum of quantum-classical transitions.

The technological implications are profound. Quantum state transfer fidelity depends critically on maintaining wave-like coherence while enabling particle-like state mapping. This requires engineering measurement processes that extract maximum information while minimizing disturbance. Silicon photonics provides the platform for optimizing this trade-off through device design. The philosophical puzzle of wave-particle duality becomes an engineering challenge with practical consequences.

Perhaps the deepest insight from silicon photonic systems is that wave-particle duality and the measurement problem are not abstract interpretational issues but concrete engineering challenges. Every quantum device embodies assumptions about measurement and duality in its design. Success or failure in building quantum technologies provides empirical feedback on our understanding of quantum foundations. Silicon waveguides don’t resolve the measurement problem philosophically but transform it into a practical question whose answer emerges through technological progress. In this view, building better quantum devices and understanding quantum mechanics become inseparable pursuits.

Final Thoughts

We stand at a remarkable juncture in the evolution of information technology. The same silicon that enabled the digital revolution now promises to host its quantum successor—not through computational brute force but through the delicate manipulation of individual photons and spins. This convergence of the classical and quantum, the electronic and photonic, the practical and profound, represents more than technological progress. It embodies humanity’s deepening dialogue with nature at its most fundamental level.

The journey mapped in these pages—from wrestling with silicon’s indirect bandgap to contemplating photons that experience no time—reveals how quantum engineering has become a form of applied philosophy. Each breakthrough in coupling spins to photons, each successful entanglement between light and matter, provides not just technological capability but empirical insight into reality’s structure. We’re not merely building devices; we’re constructing experiments that probe the universe’s quantum foundations.

Yet perhaps the most profound lesson from silicon quantum photonics is how constraints catalyze innovation. Silicon’s apparent unsuitability for quantum optics forced researchers to develop novel approaches—defect engineering, cavity enhancement, hybrid integration—that may prove more scalable than “ideal” alternatives. The field’s rapid progress suggests that perceived limitations often mask unexplored possibilities, waiting for human ingenuity to reveal them.

As quantum technologies transition from laboratory curiosities to commercial realities, silicon photonics offers a sobering reminder: the path from quantum mechanics to quantum products is neither straight nor certain. It demands patience from investors, creativity from engineers, and intellectual courage from all who venture into this quantum frontier. But for those willing to embrace both the technical challenges and philosophical implications, silicon quantum photonics promises rewards that extend far beyond computational speedup or secure communication.

The photons racing through silicon waveguides carry more than quantum information—they bear witness to humanity’s extraordinary ability to transform the most common element in Earth’s crust into a platform for exploring the universe’s deepest mysteries. In that transformation lies both the achievement and the promise of silicon quantum photonics.

Thanks for reading!