🔬 optics

Few-cycle THz Pulse Generation in DSTMS Crystal Pumped by a 8.3-MHz Amplified Mamyshev Oscillator

This paper demonstrates the first generation of few-cycle terahertz pulses at an 8.3-MHz repetition rate by optically rectifying an amplified Mamyshev oscillator in a DSTMS organic crystal, achieving 40 µW of average power and a 4 THz bandwidth that significantly outperforms comparable inorganic GaP crystal setups.

Original authors: Francesco Canella, Dario Giannotti, Sara Pizzurro, Riccardo Gotti, Lorenzo Mosesso, Stefano Lupi, Antonio Agnesi, Federico Pirzio, Gianluca Galzerano

Published 2026-01-29

📖 4 min read☕ Coffee break read

CC BY 4.0

Original authors: Francesco Canella, Dario Giannotti, Sara Pizzurro, Riccardo Gotti, Lorenzo Mosesso, Stefano Lupi, Antonio Agnesi, Federico Pirzio, Gianluca Galzerano

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 you are trying to create a very specific kind of "sound" using light. In this case, the "sound" is a pulse of Terahertz (THz) radiation—a type of energy that sits between microwaves and visible light on the spectrum. Scientists use these pulses for things like seeing through packaging or analyzing materials, but making them is usually tricky.

Here is a simple breakdown of what the researchers in this paper did, using everyday analogies:

1. The Engine: A "Mamyshev" Laser

Think of the laser used in this experiment as a high-performance race car engine.

The Mamyshev Oscillator: This is a special type of fiber laser that acts like a "pulse compressor." It takes light and squeezes it into incredibly short, powerful bursts.
The Result: The researchers created light pulses that last only 31 femtoseconds. To put that in perspective: if one femtosecond were a second, a second would be about 31 million years. These pulses are so short they are almost instantaneous.
The Power: They amplified these pulses to about 1 Watt of average power, which is a lot for such tiny, fast bursts.

2. The Factory: Turning Light into THz

Once they had these super-fast light pulses, they needed to turn them into THz radiation. They did this using a process called Optical Rectification.

The Analogy: Imagine hitting a drum with a hammer. The hammer is your laser pulse, and the drum is a special crystal. When the hammer hits, the drum vibrates and makes a sound. In this experiment, the "sound" is the THz pulse.
The Drum (The Crystal): They used two different types of "drums" to see which one worked better:
1. GaP (Gallium Phosphide): A standard, inorganic crystal (like a common wooden drum).
2. DSTMS: An organic crystal (like a high-tech, custom-made drum made of special materials).

3. The Race: Organic vs. Inorganic

The researchers set up a head-to-head comparison between the two crystals to see which one could generate the strongest THz signal.

The Winner: The DSTMS organic crystal won by a huge margin.
The Score: The DSTMS crystal produced 20 times more power than the GaP crystal.
The Catch: The DSTMS crystal is very sensitive to heat. If you hit it too hard or too often without letting it cool down, it can get damaged (like a drum skin tearing from too much heat). The researchers had to be careful with how they "chopped" (interrupted) the laser beam to keep the crystal cool.

4. The Results: A Perfect "Few-Cycle" Pulse

The goal was to create a "few-cycle" pulse.

The Analogy: Think of a wave in the ocean. A "few-cycle" pulse is like a wave that goes up and down only two or three times before stopping, rather than a long, rolling swell. This is ideal for getting a sharp, clear picture of whatever you are studying.
The Outcome: Using the DSTMS crystal and their powerful laser, they successfully created these sharp, short THz pulses.
- Speed: They generated these pulses 8.3 million times every second (8.3 MHz).
- Strength: They measured an average power of 40 microwatts (very small in human terms, but significant for this type of experiment).
- Range: The pulses covered a wide "bandwidth" (a range of frequencies) of over 4 THz, which is very broad and useful.

5. Why This Matters (According to the Paper)

Before this experiment, no one had successfully used this specific type of "Mamyshev" laser to generate THz waves.

The Breakthrough: They proved that you can combine this new, powerful laser technology with organic crystals to make THz waves very efficiently.
The Advantage: Because the laser fires so many times per second (multi-MHz), it allows for faster data collection. The paper suggests this setup is a great step toward building compact, portable devices that can analyze materials quickly, without needing the massive, complex equipment usually required for this kind of work.

In summary: The team built a super-fast laser engine, used it to hit a special organic crystal, and found that this combination creates a very strong, sharp, and fast "THz sound" much better than traditional methods. They did this without adding extra noise, proving it's a clean and efficient way to generate these waves.

Technical Summary: Few-cycle THz Pulse Generation in DSTMS Crystal Pumped by an 8.3-MHz Amplified Mamyshev Oscillator

Problem Statement
While Mamyshev oscillators (MOs) have emerged as a cutting-edge class of ultrafast fiber lasers capable of generating exceptionally broadband spectra and few-femtosecond pulses, their application in nonlinear frequency conversion for terahertz (THz) generation has not been demonstrated. Although MOs offer high pulse energies and compressibility down to ~30 fs, the initiation of mode-locking and the transfer of these features to high-energy applications via Gain-Managed Nonlinear Amplification (GMNA) present specific challenges. Furthermore, existing THz sources face a trade-off between bandwidth, average power, and repetition rate. Photoconductive antennas struggle with carrier-induced heating at MHz repetition rates, limiting average power. Conversely, while $\chi^{(2)}$ nonlinear crystals are widely available, achieving high bandwidth (>4 THz) often requires thin crystals that sacrifice dynamic range, or high-energy sources that necessitate lower repetition rates, degrading signal-to-noise ratios (SNR). Organic crystals, such as DSTMS, offer high conversion efficiency due to large nonlinear coefficients but are sensitive to thermal damage at high average powers, requiring careful management of duty cycles and cooling.

Methodology
The authors constructed an experimental setup utilizing an amplified Mamyshev oscillator seeded by a sub-nanosecond passively Q-switched (PQS) external starter to ensure robust mode-locking. The system operates in the GMNA regime, which allows for spectral broadening beyond the amplifier's gain bandwidth.

Laser Source: A 976 nm laser diode pumps a Yb-doped fiber amplifier. The seed is a Mamyshev oscillator spectrum filtered by a band-pass filter (BPF) to select a ~3 nm portion, resulting in a ~2 mW seed. This is amplified to 1.1 W average power.
Pulse Compression: Post-amplification, pulses are compressed using a grating pair (1000 grooves/mm) in a Treacy configuration to a duration of 31 fs. The pulses are pre-chirped by approximately -600 fs² to empirically maximize THz generation.
THz Generation: The amplified pulses (centered at ~1040 nm) are focused into a 190-µm-thick uncoated DSTMS organic crystal for single-pass collinear optical rectification (OR). A mechanical chopper modulates the beam to manage thermal load.
Comparison: For benchmarking, the DSTMS crystal was replaced by a conventional 500-µm-thick, AR-coated Gallium Phosphide (GaP) crystal in the same optical configuration.
Characterization: THz generation was characterized via:
1. Power Measurement: Using a pyroelectric power meter (Ophir RM9-THz) with background shielding.
2. Temporal and Spectral Analysis: Electro-optical sampling (EOS) using a 500-µm GaP crystal as the sensor, a motorized delay line, and a balanced photodiode setup synchronized with a lock-in amplifier.
3. Repetition Rate Noise: Power spectrum analysis using a Schottky diode and electrical spectrum analyzer.

Key Contributions and Results
The study reports the first demonstration of THz generation using an amplified Mamyshev oscillator.

Pulse Characteristics: The system generated 31-fs pulses with a spectral bandwidth exceeding 80 nm (extending from ~1 µm to >1.2 µm) and a peak power of 4 MW.
THz Output Performance:
- DSTMS: At an 8.3 MHz repetition rate, the system produced a maximum THz average power of 40 µW with a spectral bandwidth exceeding 4 THz. The temporal profile exhibited few-cycle oscillations.
- GaP Comparison: Using the 500-µm GaP crystal under identical conditions yielded a maximum average power of 2.3 µW. This represents a ~17.4-fold (approx. 20x) increase in power for the DSTMS crystal, while maintaining comparable spectral bandwidth.
Thermal Management: Investigations into duty cycles (10% to 50%) and pump power (up to 1.02 W) confirmed that while thermal saturation is a factor for organic crystals, the system operates effectively within these constraints. The THz power showed a purely quadratic dependence on pump power, consistent with optical rectification theory.
Noise Performance: Measurements of the repetition rate peak at 8.3 MHz showed a high SNR (~33 dB) with no significant noise degradation attributed to the Mamyshev oscillator or the OR process.
Spectral Features: The THz spectra from both crystals spanned from quasi-DC to above 4 THz. The DSTMS spectrum showed modulation due to parasitic etalon effects in the sampling crystal but maintained a bandwidth comparable to the GaP reference.

Significance and Claims
The authors claim that this work demonstrates the viability of combining Mamyshev oscillators with organic crystals for efficient, high-repetition-rate, broadband THz generation. The results confirm that the GMNA regime, when seeded by an MO, provides the necessary pulse quality and energy to drive nonlinear conversion in organic materials without the complexity of optical cavities. The paper asserts that this configuration opens promising perspectives for the development of compact, multi-MHz THz time-domain spectrometers. The work highlights that organic crystals like DSTMS can significantly outperform traditional inorganic crystals (like GaP) in terms of conversion efficiency at MHz repetition rates, provided that thermal management is addressed. The study concludes that this approach offers a robust path toward high-performance THz sources suitable for both fundamental research and industrial applications.