High bandwidth traveling wave electro-optic modulator at 1{\mu}m on thin-film lithium tantalate

This paper presents the first experimental demonstration of a high-bandwidth traveling wave electro-optic modulator on thin-film lithium tantalate operating at 1 μm, achieving a Vπ of 2.4 V, stable DC bias, and less than 2 dB electro-optic roll-off up to 50 GHz.

The Gist

Imagine you are trying to send a message using light through a special, super-efficient "hollow glass tube" (a hollow-core fiber). These tubes are amazing because they let light travel with almost no loss and can handle huge amounts of power, like a highway that never gets jammed. However, to use them effectively, you need a "traffic controller" (a modulator) that can turn the light on and off incredibly fast to encode data.

The problem is that the traffic controllers we usually use (made of Silicon or Indium Phosphide) are like old-fashioned traffic lights: they break down or get confused when the light is a specific color (around 1 micrometer, which is near-infrared). They simply aren't designed for this "wavelength window."

The Solution: A New Kind of "Smart Glass"
The researchers at Nokia Bell Labs have built a new, super-fast traffic controller using a material called Thin-Film Lithium Tantalate (TFLT). Think of this material as a "smart glass" that reacts instantly when you apply electricity, changing how it lets light pass through.

Here is the breakdown of their breakthrough in simple terms:

1. The "Speed Demon" Modulator

They created a device that can switch light on and off 50 billion times per second (50 GHz).

• The Analogy: Imagine a camera shutter that can open and close 50 times in a single second. Now, imagine that same shutter doing it 50 billion times. That is how fast this device is.
• The Efficiency: It does this with very little energy (only 2.4 Volts). It's like a high-performance sports car that gets incredible gas mileage.

2. The "Ghost" Problem (The Photo-Refractive Effect)

This is the most critical part of their discovery.

• The Old Problem: The previous version of this "smart glass" (made of Lithium Niobate) had a nasty side effect called the "Photo-Refractive effect." Imagine if you shined a flashlight on a piece of paper, and the paper slowly started to change color or get sticky, making it hard to write on later. In the old devices, the light itself would cause the material to get "sticky" with electrical charges, ruining the signal over time. This made them unstable and useless for long-term use.
• The New Fix: The team switched to Lithium Tantalate. Think of this as swapping that sticky paper for a non-stick Teflon pan. Even when you shine the light on it, it stays clean and stable. They proved that their new device can run for over an hour without drifting or getting confused, even at the tricky 1-micrometer wavelength.

3. The "Pulse" Test

To prove their device was truly stable, they didn't just check if it worked; they asked it to draw sharp, perfect squares with light.

• The Analogy: If you try to draw a sharp square with a wet, dripping marker (the old unstable material), the corners get blurry and the lines bleed. If you use a fine-tip, dry-erase marker (the new TFLT), you get crisp, perfect corners.
• The Result: Their device drew perfect, sharp light pulses without any blurring, proving it can handle the "traffic" without getting clogged up by electrical charges.

Why Does This Matter?

This discovery is like finding the missing key to unlock a new world of communication.

• Hollow-core fibers are the future of super-fast, high-power internet and data centers.
• But until now, we didn't have a reliable "traffic controller" that worked well with the specific colors of light these fibers use best.
• By showing that Lithium Tantalate works perfectly at 1 micrometer, the researchers have opened the door for using these super-fibers in real-world networks. This could mean faster internet, fewer signal boosters needed, and the ability to send massive amounts of data with less energy.

In a nutshell: They built a super-fast, energy-efficient light switch that doesn't get "sticky" or confused when exposed to specific colors of light, finally making the next generation of ultra-fast fiber optic networks possible.

Technical Summary

1. Problem Statement

The emergence of hollow-core fibers (HCFs) has opened new opportunities for optical communication using wavelengths beyond the standard C-band, specifically in the visible and near-infrared (NIR) ranges (e.g., 780 nm, 850 nm, and 1 µm). HCFs offer lower optical loss and significantly higher power-handling capabilities due to weak nonlinearity. However, existing material platforms face critical limitations in this spectral region:

• Silicon (Si) and Indium Phosphide (InP): Their bandgaps (1.1 µm and 0.9 µm, respectively) make them unsuitable for wavelengths below the O-band due to absorption.
• Thin-Film Lithium Niobate (TFLN): While TFLN offers a strong electro-optic (EO) effect and broadband transparency, it suffers severely from the photo-refractive (PR) effect. At shorter wavelengths (visible/NIR), defect-related trap centers (e.g., Fe ions) are easily photo-excited, causing DC-bias instability. This necessitates high-power thermal control, rendering TFLN impractical for large-scale, stable communication circuits at 1 µm.

There is a critical need for a material platform that combines low optical loss, a strong EO effect, high power handling, and stable DC-bias operation at 1 µm wavelengths.

2. Methodology

The authors propose and demonstrate a high-speed electro-optic modulator based on Thin-Film Lithium Tantalate (TFLT), which is known to have a weaker PR effect than TFLN.

• Device Architecture:
  • Platform: A standard 600 nm thick TFLT wafer (4-inch) with a 4.7 µm insulating oxide layer, fabricated by NanoLN.
  • Topology: A traveling-wave Mach-Zehnder Interferometer (MZI) consisting of three cascaded MZIs. The first and third MZIs act as tunable splitters to optimize beam splitting ratios across a wide wavelength range (1.5 µm down to ~1 µm).
  • Electrodes: Designed for 50 Ω impedance matching with on-chip termination. The electrode gap is 5 µm.
  • Thermal Tuning: Integrated NiCr heaters are used to control the phase of the tunable splitters and bias the modulator.
• Fabrication: Utilized a wafer-level fabrication process flow (previously detailed in the authors' prior work) involving standard lithography and etching techniques.
• Characterization:
  • Tested at four wavelengths: 984 nm, 1071 nm, 1311 nm, and 1551 nm.
  • DC Stability: Monitored output power drift over time at the quadrature point.
  • RF Performance: Measured S11 (reflection) and S21 (transmission) to extract microwave phase indices and EO bandwidth.
  • Pulse Generation: Tested square-wave modulation at low frequencies (10 Hz – 1 kHz) to observe charge-related distortions.

3. Key Contributions

• First Demonstration: This is the first experimental demonstration of a high-bandwidth (>50 GHz) TFLT electro-optic modulator operating at 1 µm.
• Superior Stability: The device demonstrates DC-bias stability at 1 µm, a regime where TFLN typically fails due to the photo-refractive effect.
• High Performance Metrics: Achieved a half-wave voltage ($V_\pi$) of 2.4 V with less than 2 dB electro-optic roll-off up to 50 GHz.
• Pulse Fidelity: Successfully generated sharp optical pulses without distortion, proving the negligible impact of charge dynamics at this wavelength.

4. Key Results

• Electro-Optic Performance:
  • $V_\pi$: Measured at 2.4 V at 1 µm. The $V_\pi L$ product scales inversely with wavelength, confirming the advantage of shorter wavelengths for tighter mode confinement and lower drive voltages.
  • Bandwidth: The EO response shows less than 2 dB roll-off up to 50 GHz at both 1071 nm and 1551 nm. This indicates ultra-flat dispersion, enabling simultaneous high-bandwidth operation across a wide spectral range.
  • Extinction Ratio: Achieved approximately 30 dB extinction, limited only by the dynamic range of the photodetector used.
• DC Bias Stability:
  • When biased at the quadrature point with ~0 dBm on-chip power, the modulator showed negligible drift over more than one hour. This contrasts sharply with TFLN, which requires active thermal stabilization to maintain bias.
• Pulse Generation:
  • The device faithfully tracked input square-wave signals at 10 Hz, 100 Hz, and 1 kHz.
  • Output waveforms preserved sharp edges with no observable distortion, confirming that charge-related effects (the primary cause of PR instability) are effectively suppressed in TFLT at 1 µm.
• Dispersion: Simulations and measurements confirmed that TFLT waveguides offer flatter dispersion compared to other platforms, supporting the wide operational bandwidth.

5. Significance

This work addresses a major bottleneck in the deployment of hollow-core fiber networks. By demonstrating a stable, high-speed modulator at 1 µm, the authors validate Thin-Film Lithium Tantalate (TFLT) as a superior material platform for next-generation optical transceivers.

• Enabling HCFs: It allows optical systems to fully leverage the low-loss and high-power capabilities of hollow-core fibers in the 1 µm band.
• Scalability: The elimination of the photo-refractive effect removes the need for complex thermal control systems, making TFLT a viable candidate for large-scale photonic integrated circuits (PICs).
• Future Applications: The technology paves the way for high-power laser delivery, quantum photonics, and advanced communication systems operating in non-conventional wavelength bands where traditional materials (Si, InP, TFLN) are ineffective or unstable.