Diamond Integrated Optomechanical Circuits

• Diamond integrated optomechanical circuits are nanophotonic platforms that merge optical resonators with nanomechanical elements in diamond for high-precision sensing and quantum applications.
• They employ high-Q optical cavities, waveguides, and free-standing resonators fabricated via advanced techniques to achieve GHz-range mechanical modes and efficient photon–phonon coupling.
• Recent advancements in fabrication and integration enhance performance metrics, enabling scalable hybrid devices with strong spin–photon–phonon interactions for quantum information processing.

Diamond integrated optomechanical circuits are monolithic or hybrid nanophotonic platforms that co-localize and interface optical and mechanical modes in diamond, enabling highly efficient signal transduction, precision sensing, and quantum photonic applications. Leveraging diamond’s high Young’s modulus, exceptional thermal conductivity, and broadband optical transparency, these circuits incorporate photonic waveguides, high-Q optical cavities, and free-standing nanomechanical resonators. Recent advancements combine polycrystalline and single-crystal diamond thin films, ion-implanted membranes, and semiconductor-on-diamond heterostructures to realize GHz-range mechanical modes, deep sideband resolution, and integration with color centers for hybrid quantum technologies (Ummethala et al., 2014, Rath et al., 2013, Burek et al., 2015, Oh et al., 8 Aug 2025, Mitchell et al., 2015, Huang et al., 11 May 2026).

1. Diamond as a Material System for Optomechanics

Diamond’s high Young’s modulus ($E \approx 1050$ GPa), density ($\rho \approx 3.52$ g/cm³), wide bandgap ($E_g = 5.47$ eV), and ultrahigh thermal conductivity ($k \approx 2000$ W/mK) make it uniquely suited for integrated optomechanical circuits (Rath et al., 2016). Compared to silicon and Si$_3$N$_4$, diamond enables higher mechanical resonance frequencies for given device dimensions and avoids nonlinear absorption at telecom and visible wavelengths due to negligible multi-photon or free-carrier effects (Rath et al., 2013, Rath et al., 2016). Its broadband optical transparency from $\approx 220$ nm to $>$10 µm allows device operation across UV–mid-IR. High mechanical $Q$ factors ($Q_\mathrm{m} > 10^6$ for SCD, $\rho \approx 3.52$0–$\rho \approx 3.52$1 for PCD) are achievable at room and cryogenic temperatures (Oh et al., 8 Aug 2025, Rath et al., 2016, Ummethala et al., 2014). The presence of optically addressable color centers (NV, SiV, etc.) enables integration with spin qubits for quantum applications (Burek et al., 2015, Oh et al., 8 Aug 2025, Kim et al., 2023).

2. Fabrication Strategies and Device Architectures

Multiple approaches have been developed for diamond integrated optomechanical circuits, including:

• Polycrystalline diamond-on-insulator: CVD-grown PCD films on SiO$\rho \approx 3.52$2/Si wafers, followed by chemo-mechanical polishing (to sub-3 nm RMS), lithography (e-beam, HSQ resist), and RIE etching to define waveguides, MZI circuits, and free-standing H-resonators (Ummethala et al., 2014, Rath et al., 2013).
• Single-crystal diamond microdisks/nanobeams: Bulk SCD chips are patterned using hard masks (e.g., Si$\rho \approx 3.52$3N$\rho \approx 3.52$4), optimized anisotropic and isotropic O$\rho \approx 3.52$5 ICP-RIE to form undercut microdisks or nanobeams with ultra-smooth sidewalls and sub-100 nm pedestals for mechanical isolation (Mitchell et al., 2018, Khanaliloo et al., 2015).
• Membrane and smart-cut techniques: Ion implantation (e.g., He$\rho \approx 3.52$6) creates damage layers, followed by CVD overgrowth and selective etching to produce SCD membranes with high uniformity, used for high-Q optomechanical crystals integrated with color centers (Oh et al., 8 Aug 2025).
• Semiconductor-on-diamond hybridization: Transfer and patterning of GaP nanobeams on diamond substrates enable integration of non-undercut optomechanical crystal cavities, facilitating strong photon–phonon–spin coupling to subsurface diamond color centers (Ma et al., 2023).
• Femtosecond laser and plasma processes: Fs-laser stress-relief and cutting for mm-scale membranes, deep-etching for form-birefringent photonic crystals, and advanced undercut geometries have been demonstrated for both pure photonic and optomechanical devices (Huang et al., 11 May 2026).

Common device architectures include:

Device Type	Optical Resonator	Mechanical Mode	Integration Aspect
H-resonator in MZI	Ridge/slot waveguide	In-plane flexural	Embedded in interferometer
Nanobeam OMC	1D PhC cavity	GHz breathing/flapping	Integrated with waveguides, color centers
Microdisk	WGM resonator	GHz RBM	Side-coupled bus, phononic shield
Disk-on-membrane	Whispering gallery	Suspended disk modes	mm-scale platform
GaP-on-diamond	Hybrid PhC cavity	Surface acoustic	No undercut, spin addressing

Surface and etch quality strongly affect $\rho \approx 3.52$7 and $\rho \approx 3.52$8. CMP-polishing, optimized mask stacks, sidewall passivation, and post-etch annealing are crucial for loss minimization (Ummethala et al., 2014, Mitchell et al., 2018).

3. Optomechanical Coupling and Theoretical Framework

The canonical optomechanical interaction Hamiltonian is: $\rho \approx 3.52$9 where $E_g = 5.47$0 ($E_g = 5.47$1) is the photon (phonon) annihilation operator of the optical (mechanical) mode and $E_g = 5.47$2 is the vacuum coupling rate. $E_g = 5.47$3 arises from both moving boundary (MB) and photoelastic (PE) effects: $E_g = 5.47$4 For nanobeam photonic crystals, $E_g = 5.47$5 between 100–300 kHz (microdisk), 136–234 kHz (OMCs), and up to 215 kHz (smart-cut membrane OMCs with color centers) have been reported (Burek et al., 2015, Oh et al., 8 Aug 2025, Mitchell et al., 2015, Mitchell et al., 2018). Theoretical expressions involve FEM-computed derivatives of the optical resonance with respect to mechanical displacement and the zero-point motion amplitude $E_g = 5.47$6.

High optomechanical cooperativity $E_g = 5.47$7 is achieved for $E_g = 5.47$8 up to $E_g = 5.47$9 in nanobeams and microdisks, with $k \approx 2000$0 at cryogenic temperatures reported for $k \approx 2000$1 (Oh et al., 8 Aug 2025).

In the dissipative (waveguide) regime, coupling is quantified by the derivative of the fiber–nanobeam coupling coefficient, $k \approx 2000$2, exceeding $k \approx 2000$3 GHz/nm and yielding displacement measurement sensitivity approaching $k \approx 2000$4 fm/$k \approx 2000$5 (Khanaliloo et al., 2015).

4. Experimental Performance Metrics and Demonstrated Functionality

Diamond integrated optomechanical circuits achieve state-of-the-art figures of merit in both optical and mechanical domains:

• Optical $k \approx 2000$6 factors: $k \approx 2000$7 (microdisks, nanobeams, OMCs), $k \approx 2000$8 up to $k \approx 2000$9 (microdisks) (Mitchell et al., 2018)
• Mechanical $_3$0 factors: $_3$1–$_3$2 (PCD H-resonators, MZI circuits) at MHz; $_3$3 in SCD OMCs at 6 GHz at 4 K (Ummethala et al., 2014, Oh et al., 8 Aug 2025)
• Sideband resolution: $_3$4 approaching and exceeding unity for GHz modes (e.g., $_3$5 in (Oh et al., 8 Aug 2025), $_3$6–$_3$7 in (Burek et al., 2015))
• Sensitivity: Displacement sensitivity $_3$8 fm/$_3$9, force sensitivity limited by $_4$0 and $_4$1 (Ummethala et al., 2014, Khanaliloo et al., 2015)
• Actuation and control: Both optical gradient force and electrostatic actuation are demonstrated with tunable nonlinearity (Duffing, geometric stiffening/softening) (Rath et al., 2014, Rath et al., 2013)
• Readout: High-extinction, low-loss MZI readout and on-chip interferometry; superconducting nanowire single-photon detectors (SNSPDs) monolithically integrated for quantum photonic circuits (Rath, 2017, Rath et al., 2016)

These systems support resonances from a few MHz (H-resonators, disk supports) to tens of GHz (OMCs, hybrid nanobeams), with large-scale wafer-scale integration possible via standard lithographic patterning on diamond-on-insulator or SCD membranes (Ummethala et al., 2014, Huang et al., 11 May 2026).

5. Circuit Integration, Signal Routing, and Hybrid Interfaces

Diamond optomechanical circuits support complex routing and scaling:

• Photonic routing: Single-mode diamond waveguides connect OMCs, microdisks, and MZI arms, supporting both C/L band and visible operation (Burek et al., 2015, Rath et al., 2016).
• Phononic routing: Phononic-crystal waveguides (GHz bandgaps), diamond “buses,” and acoustic shields transfer mechanical signals between circuit elements, enabling phonon-mediated coupling (Burek et al., 2015, Mitchell et al., 2015, Oh et al., 8 Aug 2025).
• Multiplexing: Wavelength-division (photonic) and frequency-division (mechanical) multiplexing, supported by integrated ring filters and arrays of OMCs with distinct $_4$2 (Burek et al., 2015).
• Hybrid spin–photon–phonon coupling: NV and SiV centers addressed via local strain and cavity Purcell effect, enabling coherent quantum state transfer and quantum networking (Oh et al., 8 Aug 2025, Burek et al., 2015, Kim et al., 2023).
• Alternative integration: Semiconductor-on-diamond (e.g., GaP nanobeams) enables cavities and OMCs without undercut, with improved robustness and direct access to color centers in the substrate (Ma et al., 2023).
• Microwave-to-optical transduction: Piezoelectric (AlN) actuators integrated with OMCs provide microwave–phonon–photon connectivity at the chip scale (Kim et al., 2023).

These integration strategies support the assembly of large-scale, multi-node quantum photonic architectures on a diamond platform.

6. Key Challenges and Optimization Approaches

Despite substantial progress, diamond integrated optomechanical circuits face several technical challenges:

• Optical absorption and scattering: Sidewall roughness and plasma-induced damage remain dominant contributions to $_4$3 limitations; optimized etch chemistry, post-fabrication cleaning/annealing, and high-aspect-ratio etch processes are used to mitigate (Mitchell et al., 2018, Huang et al., 11 May 2026).
• Mechanical loss mechanisms: Anchor (clamping) losses, TLS (two-level-system) dissipation, and surface contamination reduce $_4$4. Strategies include phononic shields, surface treatment, beam/support geometry optimization, and vacuum or cryogenic operation (Huang et al., 11 May 2026, Oh et al., 8 Aug 2025).
• Thermal management: High intracavity powers create temperature shifts via residual absorption. Diamond's thermal conductivity is a key advantage, but cryogenic operation and active stabilization are used for quantum-limited performance (Burek et al., 2015, Oh et al., 8 Aug 2025).
• Fabrication complexity and yield: Achieving uniformity in sub-100-nm features (membrane thickness, hole roughness), precise lithographic alignment, and scalable hybrid bonding remain bottlenecks for mass production (Burek et al., 2015, Ma et al., 2023).
• Integration with electronics and detectors: On-chip integration of SNSPDs, photodiodes, and control/readout circuits requires compatibility of subsequent processing routes and low-loss interconnects (Rath, 2017, Rath et al., 2016).

Ongoing research addresses these issues with improved membrane formation (e.g., smart-cut + CVD overgrowth), optimized integration of piezoelectric and GaP/diamond stacks, and advanced post-processing methodologies (Oh et al., 8 Aug 2025, Kim et al., 2023).

7. Applications and Outlook

Diamond integrated optomechanical circuits address a range of classical and quantum functionalities:

• Precision sensing: Force, mass, and displacement sensing with resolution below $_4$5 fm, using high-$_4$6 mechanical and optical elements (Ummethala et al., 2014, Rath et al., 2013).
• Signal transduction: GHz-range optomechanical and optoelectronic signal processing for RF photonics and microwave-to-optical interfacing (Burek et al., 2015, Kim et al., 2023).
• Quantum information processing: Memory interfaces and transduction between spins (NV, SiV), photons, and phonons, including ground-state cooling and phonon-mediated entanglement (Oh et al., 8 Aug 2025, Burek et al., 2015, Ma et al., 2023).
• Large-scale integration: Wafer-scale, lithographically defined platforms capable of supporting hundreds to thousands of optomechanical nodes, with prospects for distributed quantum repeaters and hybrid processors (Ummethala et al., 2014, Oh et al., 8 Aug 2025).
• Hybrid circuit modules: Integration of mechanics, photonics, and single-photon detection on monolithic diamond chips (Rath, 2017, Rath et al., 2016).

Further advances are expected in high-cooperativity quantum protocols, e.g., quantum-limited transducers ($_4$7), scalable microwave-to-optical conversion, and on-chip entanglement of color center qubits. Progress in diamond nanofabrication and photonics will continue to push fidelity, efficiency, and scalability toward practical quantum networks and advanced nanomechanical sensing (Kim et al., 2023, Oh et al., 8 Aug 2025).