Release-free electro-optomechanical crystal modulator

This paper demonstrates a release-free electro-optomechanical transducer that integrates silicon optomechanical crystals with lithium niobate via micro-transfer printing to achieve quantum-compatible coupling rates, thereby overcoming thermal noise limitations and advancing practical microwave-optical interfaces for quantum technologies.

The Gist

Imagine you have two very different languages: Microwave (the language of super-fast computers and quantum processors) and Optical Light (the language of fiber-optic internet cables). These two languages speak at completely different speeds and frequencies, making it nearly impossible for them to talk to each other directly.

This paper introduces a new "translator" device that helps these two languages understand each other. Here is the breakdown of how it works, using simple analogies:

1. The Problem: The "Floating" Translator

Scientists have been trying to build these translators for years. The best previous versions were like suspended bridges. They worked well because they were isolated, but they had a major flaw: they were thermally "floating."

• The Analogy: Imagine trying to cool a hot cup of coffee by holding it in your hand while standing in a blizzard. If the cup is floating in the air (suspended), the cold air can't touch the bottom to cool it down efficiently. In these old devices, the heat generated by the laser couldn't escape easily, creating "thermal noise" (static) that ruined the delicate quantum conversation.

2. The Solution: The "Grounded" Translator

The team at Chalmers University built a new type of translator that is release-free.

• The Analogy: Instead of a floating bridge, they built a solid road that is firmly attached to the ground.
• How it works: They took a silicon chip (the ground) and glued a thin slice of a special crystal called Lithium Niobate on top of it. Because the device is still attached to the silicon "ground," heat can flow away easily, just like a hot pan cooling down on a metal stove. This keeps the device quiet and stable.

3. The Mechanism: The "Middleman"

The device doesn't translate Microwaves directly to Light. It uses a mechanical vibration (a tiny, invisible shaking) as a middleman.

• Step 1 (Microwave to Shake): A microwave signal hits a special part of the chip (made of Lithium Niobate), which acts like a piezoelectric speaker. It turns the electrical signal into a tiny, high-speed vibration (a phonon).
• Step 2 (The Shake): This vibration travels through the silicon.
• Step 3 (Shake to Light): The vibration hits a laser beam trapped in the silicon. The shaking changes the laser's properties, effectively "imprinting" the microwave message onto the light.

4. The Innovation: Micro-Transfer Printing

How did they stick the Lithium Niobate onto the silicon so perfectly?

• The Analogy: Think of it like microscopic stamping. They printed the Lithium Niobate pattern onto a soft rubber stamp (PDMS) and then gently pressed it onto the silicon chip, like stamping a piece of paper. This allowed them to combine the best properties of two different materials without melting or damaging them.

5. What They Actually Achieved

The paper reports on a "proof-of-concept" experiment. They didn't build a commercial product, but they proved the idea works:

• The Test: They sent a microwave signal in and successfully detected a corresponding signal in the light coming out.
• The Data: They showed they could send a simple digital message (a string of 1s and 0s) through this translator. When they sent a "square wave" (a digital signal), the light output showed the same pattern, proving the device can carry information.
• The Limitation: The current version is a bit "noisy" and inefficient compared to what the math predicted. The authors admit that the physical device they built was slightly different in size than the computer design, which affected the performance. However, the fact that it works at all is a major step forward.

Summary

This paper demonstrates a new, grounded translator that connects superconducting quantum computers to fiber-optic networks. By keeping the device firmly attached to a silicon base, they solved the heat problem that plagued previous designs. While the current version is a laboratory prototype with some imperfections, it successfully proved that you can use a "grounded" design to translate microwave signals into light, paving the way for future quantum networks.

Technical Summary

Technical Summary: Release-free Electro-optomechanical Crystal Modulator

Problem Statement
Electro-optic modulation is essential for classical communications and emerging quantum technologies, specifically for interfacing superconducting qubits with optical fibers. While high-confinement suspended optomechanical crystals (OMCs) have enabled strong optomechanical interactions and quantum experiments, they suffer from a critical limitation: thermal noise generated by optical absorption. Because suspended devices are thermally disconnected from the substrate, they have poor thermal anchoring, which restricts power handling and introduces noise that hinders quantum-level operation. Conversely, "release-free" devices (where the active layer remains attached to the substrate) offer superior thermal anchoring but have historically struggled to confine and couple mechanical and optical waves effectively due to the presence of the substrate. Furthermore, integrating these release-free structures with piezoelectric materials for microwave-to-optical transduction has not yet been demonstrated.

Methodology
The authors present a chip-scale, release-free electro-optomechanical transducer designed to overcome these limitations. The device is fabricated on a silicon-on-insulator (SOI) platform (220 nm silicon on SiO2) and heterogeneously integrated with a thin-film lithium niobate (LiNbO3) layer via micro-transfer printing.

• Design Strategy: The device utilizes a "release-free" architecture where the silicon device layer remains attached to the SiO2 substrate. To achieve mechanical confinement and optomechanical phase-matching despite the substrate, the design targets mechanical modes near the X-point of the Brillouin zone, which are phase-protected from the radiation continuum. The optical wavevector is targeted at half the mechanical wavevector ($k_m = 2k_o$) to enable coupling between a mechanical wave and counter-propagating optical waves.
• Structure: The transducer consists of two distinct regions:
  1. An optomechanical region (silicon with elliptical holes) where optical photons and phonons interact via spontaneous parametric up/down-conversion.
  2. An electromechanical region featuring a thin-film LiNbO3 layer on top of the silicon, patterned with a short interdigitated transducer (IDT) to convert microwave signals into phonons via the piezoelectric effect.
• Fabrication: The process involves patterning the LiNbO3 film, suspending it, and transfer-printing it onto the SOI chip. Subsequent lithography defines the silicon photonic crystal and aluminum electrodes.
• Characterization: The device was characterized at room temperature using a tunable laser coupled via a grating coupler and microwave tones applied to the aluminum pads. Measurements included optical resonance spectroscopy, mechanical linewidth analysis, and microwave-to-optical scattering parameter ($S_{oe}$) measurements.

Key Contributions

1. First Release-free Piezo-optomechanical Transducer: This work demonstrates the first integration of a high-confinement optomechanical crystal with a piezoelectric transducer in a release-free configuration, combining the strong optomechanical interactions of silicon with the efficient piezoelectricity of lithium niobate.
2. Thermal Anchoring: By keeping the device layer attached to the substrate, the design improves thermal anchoring and power handling compared to suspended OMCs, addressing the thermal noise bottleneck.
3. Quantum-Ready Coupling Rates: The device achieves electro- and optomechanical coupling rates that are compatible with quantum-level operation when co-integrated with superconducting microwave circuits.
4. Classical Signal Transmission: The authors demonstrate the device's functionality as an electro-optic interface by successfully transmitting classical digital data (bit arrays) through the transducer.

Results

• Optical and Mechanical Properties: The device exhibits an optical resonance at 194.9 THz with an intrinsic quality factor ($Q_{o,i}$) of $1.7 \times 10^5$. The primary mechanical mode is located at 4.32 GHz, placing the system in the resolved-sideband regime required for quantum transduction.
• Coupling Rates: The measured vacuum optomechanical coupling rate is $g_{om}/(2\pi) = 130$ kHz. The intrinsic mechanical linewidth is $\gamma_{m,i}/(2\pi) = 8.4$ MHz.
• Electromechanical Conversion: The electromechanical decay rate into a 50 $\Omega$ feedline was measured at $\gamma_{m,e}/(2\pi) = 58$ Hz, corresponding to an electromechanical conversion efficiency ($\eta_{em}$) of $7 \times 10^{-6}$.
• Total Efficiency: The total microwave-to-optical conversion efficiency ($\eta_{tot}$) was measured at $1.5 \times 10^{-7}$. The authors note that this measurement utilized a blue-detuned pump, which stimulates parametric pair generation rather than direct photon conversion (which requires red detuning).
• Discrepancies: Measured coupling rates and resonant frequencies deviated from initial simulations. The authors attribute this to fabrication deviations (e.g., hole sizes, sidewall angles, electrode thickness) and refined their simulations using SEM data to achieve qualitative agreement with the observed spectra.
• Classical Transmission: The device successfully transmitted a 48-bit non-return-to-zero (NRZ) array. Eye diagrams were recorded at bit rates up to 10 Mbit/s, limited by the mechanical ring-up and ring-down dynamics ($\gamma_m \approx 9.25$ MHz).

Significance and Claims
The paper claims that this work represents a significant step toward practical microwave-optical interfaces for quantum technologies. By demonstrating a release-free architecture, the authors show that it is possible to achieve strong optomechanical interactions while maintaining improved thermal anchoring, a crucial factor for reducing thermal noise in quantum transducers.

The authors modestly state that while the current electromechanical efficiency is limited by intrinsic mechanical loss and coupling strength, the platform is viable for quantum applications. They argue that coupling the transducer to a high-impedance microwave resonator or a superconducting qubit (specifically a transmon) could dramatically boost the electromechanical cooperativity ($C_{em}$), potentially enabling Rabi swapping operations if cryogenic mechanical loss rates are sufficiently low. Furthermore, the platform offers a route for energy-efficient classical electro-optic modulation and readout of superconducting circuits. The authors conclude that refining the fabrication process to match the simulated design dimensions and operating the device under cryogenic conditions are necessary next steps to realize single-phonon level performance and full quantum transduction.