🔬 optics

Enhanced Third-Harmonic Generation in Diamond Photonic Crystal Slabs via Doubly Resonant Quasi-Bound States in the Continuum

This paper proposes and numerically demonstrates a design for a diamond photonic crystal slab that achieves enhanced third-harmonic generation by simultaneously resonating the fundamental and third-harmonic modes via a doubly resonant quasi-bound state in the continuum, resulting in high conversion efficiency and tunability across visible to deep-UV wavelengths.

Original authors: Sangmin Ji, Satoshi Iwamoto, Hideki Gotoh

Published 2026-07-20

📖 4 min read☕ Coffee break read

CC BY 4.0

Original authors: Sangmin Ji, Satoshi Iwamoto, Hideki Gotoh

Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

Imagine light not just as a beam you see, but as a musical note. In the world of physics, there's a special trick called "nonlinear optics" where you can smash two or more of these light notes together to create a brand new, higher-pitched note. It's like taking three low, rumbling bass sounds and magically fusing them into one sharp, high-pitched whistle. This is called "third-harmonic generation." Scientists love this because it lets them turn invisible infrared light (the kind used in fiber-optic internet cables) into visible light or even ultraviolet light, which is crucial for things like quantum computers and advanced medical imaging.

However, doing this is incredibly hard. Light usually just passes right through materials without mixing. To make them mix, you need a "trap" to hold the light still and make it bounce around so much that it gets excited. This is where "photonic crystals" come in. Think of them as a microscopic honeycomb made of a super-clear material called diamond. By carving tiny holes into this diamond, scientists can create a cage that traps light perfectly. But there's a catch: diamond is a very symmetrical material, which usually acts like a bouncer that refuses to let the light notes mix at all. The big question has been: How do we trick this symmetrical diamond into letting the light notes mix efficiently, without needing a massive, clumsy machine?

This paper proposes a clever solution using a diamond "honeycomb" with a twist. The researchers, Sangmin Ji and his team, designed a special diamond slab where they trap two different "notes" of light at the same time: the original low note (the fundamental) and the new high note (the third harmonic). Usually, trapping both at once is a nightmare because they want to live in different parts of the material. But the team found a way to engineer the diamond so that both notes can hang out in the same tiny cage.

The secret sauce is in the shape of the holes. In a perfect circle, the diamond's symmetry blocks the mixing. So, the team changed the holes from circles into equilateral triangles. Imagine a round table where everyone is too polite to talk to each other; by switching to a triangular table, the "bouncer" steps aside, and the light notes finally start chatting and mixing. This shape-shifting breaks the rules just enough to let the magic happen, while also keeping the diamond slab thin enough to be practical.

Using powerful computer simulations (not a physical lab experiment yet, but a very detailed digital model), the team showed that this design works beautifully. They found that with their triangular-hole diamond cage, the normalized efficiency of turning the low light into high light is $2.7 \times 10^{-7} \text{ W}^{-2}$ under moderate quality factors. But here's the exciting part: the paper projects that if the device can be built with higher precision—specifically reaching a fabrication-limited quality factor of about 200,000—the normalized efficiency could reach approximately $0.034 \text{ W}^{-2}$ . That would be a massive improvement, potentially making this the most efficient way to do this specific light trick on a tiny chip.

Why does this matter? Because diamond is transparent to almost everything, from deep ultraviolet to visible light and beyond. This means the same diamond design could be scaled up or down to work with different colors of light, making it a universal tool for future quantum networks and super-fast computers. The paper doesn't claim to have built the final device yet, but it provides a solid, mathematically backed blueprint showing that a simple change in hole shape could unlock a new era of efficient light manipulation on a diamond chip.

Technical Summary: Enhanced Third-Harmonic Generation in Diamond Photonic Crystal Slabs via Doubly Resonant Quasi-Bound States in the Continuum

Problem Statement
Diamond is an exceptional platform for on-chip photonics due to its ultra-wide transparency window, high thermal conductivity, and hosting of optically active color centers (e.g., NV⁻, SiV⁻). However, a significant spectral mismatch exists: spin-photon interactions occur in the visible range (e.g., 637 nm), while low-loss fiber networks operate in the telecommunication bands (1.3–1.55 µm). Bridging this gap requires efficient, monolithic wavelength conversion within diamond. While diamond possesses a high nonlinear refractive index, its centrosymmetric structure restricts optical nonlinearity to the third-order susceptibility ( $\chi^{(3)}$ ), which is intrinsically weak. Consequently, efficient frequency conversion typically demands strong field enhancement via optical cavities.

A specific challenge arises in achieving doubly resonant third-harmonic generation (THG), where both the fundamental harmonic (FH, $\omega$ ) and the third harmonic (TH, $3\omega$ ) are simultaneously confined. In photonic crystal (PhC) slabs, the FH and TH frequencies often span different photonic bandgaps; the TH mode frequently lies deep within the radiative light cone, preventing conventional bandgap confinement. Furthermore, in centrosymmetric materials like diamond, symmetry constraints can cause the nonlinear coupling coefficient ( $\beta_3$ ) to vanish, rendering the overlap between the FH-induced nonlinear polarization and the TH mode field zero. Existing solutions, such as microrings, face stringent phase-matching requirements, while previous BIC-based approaches have largely focused on non-centrosymmetric materials or plane-wave illumination rather than localized cavity pumping.

Methodology
The authors propose a design strategy combining band engineering, symmetry breaking, and coupled-mode theory (CMT) to overcome these limitations.

Theoretical Framework: The study derives a closed-form expression for THG efficiency ( $\eta$ ) using temporal CMT for a doubly resonant cavity. The efficiency is shown to scale as $\eta \propto Q_1^3 Q_3 |\beta_3|^2$ , where $Q_1$ and $Q_3$ are the quality factors of the FH and TH modes, respectively, and $\beta_3$ is the nonlinear coupling coefficient. This expression serves as a Figure of Merit (FOM) to guide the design.
Band Engineering and Symmetry Breaking:
- FH Mode: The design utilizes a K-point band-edge mode of a hexagonal-lattice PhC slab. This mode is chosen for its strong field confinement in the dielectric and its ability to achieve high $Q$ factors within the heterostructure mode gap.
- TH Mode: The TH mode is identified as a $\Gamma$ -point quasi-bound state in the continuum (quasi-BIC).
- Symmetry Breaking: In a standard circular-hole hexagonal lattice, the symmetry of the TH mode (specifically "Mode II") results in a vanishing nonlinear overlap ( $\beta_3 \approx 0$ ) with the FH mode. To rectify this, the authors deform the circular air holes into equilateral triangular holes while maintaining constant area. This deformation breaks the in-plane symmetry, converting the symmetry-forbidden overlap into a finite, non-zero value. Additionally, this geometric change lowers the TH frequency, facilitating the satisfaction of the frequency-tripling condition ( $3\omega_1 \approx \omega_3$ ) at a practical slab thickness.
Optimization: The unit cell geometry (hole size $d$ and slab thickness $t$ ) is optimized to maximize the FOM ( $Q_3|\beta_3|^2$ ). The optimal design is then embedded into a PhC heterostructure cavity consisting of core, transition, and cladding layers. The hole sizes are graded ( $d_{core} > d_{trans} > d_{clad}$ ) to pull the cladding bands to lower frequencies, creating a mode gap that confines both FH and TH modes in the core.

Key Contributions

First Doubly Resonant THG in Diamond: The paper presents the first proposal and numerical demonstration of a doubly resonant THG scheme in a centrosymmetric diamond PhC slab, addressing the specific symmetry constraints of $\chi^{(3)}$ processes.
Symmetry-Guided Design: The work demonstrates that breaking lattice symmetry via triangular air holes is a necessary and effective method to activate nonlinear coupling in centrosymmetric materials, enabling efficient THG without requiring $\chi^{(2)}$ materials.
Closed-Form FOM: The derivation of a material-independent, closed-form efficiency expression provides a rigorous metric for optimizing cavity parameters, specifically balancing the trade-off between $Q$ factors and nonlinear overlap.
Scalable Architecture: The design relies on lattice symmetry rather than specific material dispersion, offering a scalable recipe applicable across various wavelengths and integrated photonic platforms.

Results
Three-dimensional finite-difference time-domain (3D-FDTD) simulations were performed on the optimized diamond PhC heterostructure cavity ( $a = 0.584$ µm, $t = 0.55a$ ).

Resonance Characteristics: The cavity successfully co-confines the FH mode ( $\lambda_1 \approx 1.541$ µm, $Q_1 \approx 4,000$ ) and the TH mode ( $\lambda_3 \approx 514$ nm, $Q_3 \approx 1,622$ ). The frequency ratio deviates by only 0.01% from the exact tripling condition.
Efficiency: Under moderate quality factors ( $Q_1 = 4,000$ ), the normalized THG efficiency is calculated as $\eta = 2.7 \times 10^{-7}$ W $^{-2}$ .
Fabrication-Limited Projection: Extrapolating to fabrication-limited quality factors ( $Q_1 \approx 200,000$ ), the projected efficiency reaches $\eta \approx 0.034$ W $^{-2}$ .
Benchmarking: When compared to related works (including silicon metasurfaces and AlN microrings), the proposed diamond design, even at conservative $Q$ values, outperforms most reported structures. At the projected fabrication limit, it exceeds the record-high THG performance reported to date by two orders of magnitude, despite diamond's $\chi^{(3)}$ being two orders of magnitude smaller than silicon's.

Significance
The authors claim that this work establishes a robust route toward efficient, monolithic on-chip frequency conversion in an all-diamond platform. By leveraging the ultra-wide transparency of diamond and the unique properties of quasi-BICs, the design enables frequency conversion from telecommunication bands to visible and deep-UV outputs using a single geometry. This capability is critical for quantum photonics, specifically for interfacing visible-wavelength spin-photon interactions (color centers) with telecommunication-band fiber networks, thereby enabling distributed quantum networking. The approach is presented as a general, scalable solution for enhancing $\chi^{(3)}$ interactions in integrated photonics.

More like this