Quantum computing, secure communication, and high-precision sensing rely fundamentally on the ability to generate individual photons or entangled photon pairs with precise, predictable properties. Because photons do not naturally interact with one another, researchers rely on nonlinear optical processes to split a high-energy laser beam into pairs of lower-energy, highly correlated photons.
Two primary mechanisms drive this field: Spontaneous Parametric Down-Conversion (SPDC) in nonlinear crystals and Spontaneous Four-Wave Mixing (SFWM) in optical fibers. By understanding and engineering these processes, researchers can tailor the polarization, wavelength, and temporal properties of light to build scalable quantum architectures.
The Core Mechanism: Splitting and Mixing Light
At the heart of parametric photon generation is the conservation of energy and momentum. In a nonlinear medium, these conservation laws manifest as phase-matching conditions, which dictate exactly how incoming pump photons transform into outgoing "signal" and "idler" photon pairs.
Spontaneous Parametric Down-Conversion (SPDC):
[ Pump Photon (2ω) ] ---> [ Nonlinear Crystal ] ---> [ Signal (ω) ] + [ Idler (ω) ]
Spontaneous Four-Wave Mixing (SFWM):
[ Pump (ω) ] + [ Pump (ω) ] ---> [ Nonlinear Fiber ] ---> [ Signal (ω+Δ) ] + [ Idler (ω-Δ) ]
In SPDC, a single high-energy pump photon interacts with a crystal's second-order ($\chi^{(2)}$) nonlinearity, spontaneously splitting into two lower-energy photons. Because they are born from the same parent photon, their properties are intrinsically linked. Jeong-Hoon Ko et al. demonstrated this clean correlation in Another Quantum Eraser Experiment with Two-photon States of Light, using SPDC-generated pairs to perform high-visibility path-erasure experiments in a Hong-Ou-Mandel interferometer. This tight correlation also extends to the time domain; Bhashyam Balaji et al. leveraged these intrinsic temporal links in Target Detection Aided by Quantum Temporal Correlations: Theoretical Analysis and Experimental Validation to drastically reduce the measurement time required to identify targets in high-noise environments.
Alternatively, SFWM utilizes a material's third-order ($\chi^{(3)}$) nonlinearity, typically inside an optical fiber. Here, two pump photons are annihilated to simultaneously create a signal and an idler photon.
Heralding and the Quest for Spectral Purity
A major challenge in quantum optics is that parametric generation is inherently probabilistic; we cannot predict the exact microsecond a photon pair will emerge. To circumvent this, researchers use a technique called heralding. Because photons are always generated in pairs, detecting the idler photon acts as a trigger—a "herald"—confirming that the signal photon exists and is ready for interaction.
Thomas Roger et al. utilized this framing in Coherent perfect absorption in deeply subwavelength films in the single-photon regime, employing heralded photons to force a deterministic, unitary coupling between a single excitation and an ultra-thin metamaterial. Similarly, Wang Lian et al. showed in Correction to: New scheme for measurement-device-independent quantum key distribution that passive heralded single-photon sources can form a reliable foundation for quantum key distribution, removing the need for complex active modulation systems.
However, for heralded photons to interfere properly in quantum networks, they must be indistinguishable and spectrally pure, meaning they cannot be entangled with the residual pump noise. Researchers achieve this by engineering the geometry and material properties of the nonlinear medium:
- Birefringence Engineering: Jasleen Lugani et al. demonstrated in Spectrally pure single photons at telecommunications wavelengths using commercial birefringent optical fiber that aligning pump light along a fiber's slow axis allows the inherent birefringence to naturally handle phase matching. This cross-polarized four-wave mixing (XFWM) forces the generated photons into orthogonal polarizations, making it simple to filter out pump noise.
- Dispersion Engineering: In Photon-pair generation in photonic crystal fibre with a 1.5GHz modelocked VECSEL, Oliver J. Morris et al. showed how the microscopic air-hole structures of photonic crystal fibers (PCFs) can be customized to manipulate the dispersion profile. This allows precise control over the output wavelengths, tailoring the source to match efficient single-photon detectors.
Tailoring Entanglement and Wavelengths
Depending on the application, photon pairs must be generated at specific wavelengths or embedded in specific entangled states.
For long-distance fiber networks, the 1.55 µm telecommunications band is ideal due to its low attenuation. Joseph B. Altepeter et al. achieved this by embedding a dispersion-shifted fiber inside a Sagnac loop in Demonstration of a Quantum Controlled-NOT Gate in the Telecommunications Band, creating a highly stable, natively compatible telecom photon source to drive a linear-optical CNOT gate.
Conversely, free-space daylight communication suffers from solar background noise. To bypass this, Adetunmise C. Dada et al. shifted the generation to the mid-infrared spectrum in Near-Maximal Two-Photon Entanglement for Optical Quantum Communication at 2.1 µm. By utilizing periodically poled lithium niobate (PPLN) crystals, the authors achieved near-maximal entanglement at 2.1 µm, where atmospheric transmission is high and solar noise is significantly reduced.
[ Single Pump Polarization ]
│
▼
┌──────────────────────────────────────┐
│ AlGaAs Waveguide │
│ ├── Type-0 SPDC ──> Co-polarized ───┼─> Superposition State
│ └── Type-I SPDC ──> Co-polarized ───┼─> (Bell State)
└──────────────────────────────────────┘
To make these sources practical, researchers are moving away from bulky laboratory optics toward integrated, on-chip solutions. Dongpeng Kang et al. demonstrated a monolithic semiconductor chip in Two polarization-entangled sources from the same semiconductor chip that simultaneously drives Type-0 and Type-I SPDC processes within a single AlGaAs waveguide. By overlapping these processes in the spatial and spectral domains, they generated high-quality Bell states directly on-chip without requiring external compensation optics.
Scaling Up to Networks and Systems
Once high-fidelity photon pairs can be reliably engineered, they can be scaled into multi-photon states or distributed across large-scale networks.
For quantum computing, generating pairs is only the first step; these pairs must be fused into larger entangled structures. Chao‐Yang Lu et al. demonstrated this scalability in Experimental entanglement of six photons in graph states by combining multiple SPDC-generated photon pairs using polarizing beam splitters. This fusion operation allowed them to successfully build complex six-photon Greenberger-Horne-Zeilinger (GHZ) and cluster states.
Pair 1 (SPDC) ───┐
├──> [ Polarizing Beam Splitter ] ───> Multi-photon Graph State
Pair 2 (SPDC) ───┘ (Fusion Operation)
For quantum sensing, these correlated states can be distributed over long distances to beat classical measurement limits. In Field Demonstration of Distributed Quantum Sensing without Post-Selection, Si-Ran Zhao et al. distributed SPDC photon pairs across a 240-meter network, leveraging the quantum correlations between separated sensors to unconditionally surpass the classical shot-noise limit.
Looking toward massive parallelization, Zhibin Ye proposed a method in Quantum illumination via correlation-to-displacement conversion with cavity-enhanced mode selection to pump an SPDC source with a single-frequency laser, generating millions of frequency-mode entangled pairs simultaneously to optimize quantum illumination protocols in high-noise environments.
Finally, long-distance quantum internet architectures require a bridge between flying photonic qubits and stationary quantum memory. Haole Jiao et al. addressed this in Generation of highly retrievable atom photon entanglement with a millisecond lifetime via a spatially multiplexed cavity by integrating a ring cavity with a polarization interferometer. This setup maps parametric quantum states into two distinct spin-wave modes within an atomic ensemble, achieving a spatially multiplexed atom-photon interface with a millisecond lifetime.
Key Foundational Work
Experimental entanglement of six photons in graph states (2007) — contributed 2 concept elements to this direction.
Another Quantum Eraser Experiment with Two-photon States of Light (2002) — contributed 1 concept elements to this direction.