🔬 optics

Structured Ytterbium and Erbium -doped Silica Fiber for Dual Wavelength Laser Operation

This paper reports the fabrication and characterization of novel structured-core silica fibers containing spatially separated ytterbium and erbium-doped regions, which enable dual-wavelength laser emission at 1042 nm and 1550 nm with controllable power ratios by optimizing the active fiber length.

Original authors: Ivo Barton, Pavel Peterka, Martin Grabner, Jan Aubrecht, Michal Kamradek, Ondrej Podrazky, Petr Varak, Dariusz Pysz, Marcin Franczyk, Rafal Kasztelanic, Ryszard Buczynski, Ivan Kasik

Published 2026-03-24

📖 4 min read☕ Coffee break read

CC BY 4.0

Original authors: Ivo Barton, Pavel Peterka, Martin Grabner, Jan Aubrecht, Michal Kamradek, Ondrej Podrazky, Petr Varak, Dariusz Pysz, Marcin Franczyk, Rafal Kasztelanic, Ryszard Buczynski, Ivan Kasik

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

The Big Idea: A Fiber Optic "Two-Headed" Laser

Imagine you have a standard fiber optic cable. Usually, it's like a single-lane highway carrying one specific type of traffic (light) at one specific color. Scientists want to build a "two-headed" laser that can shoot two different colors of light at the same time: Red-Infrared (1042 nm, used for cutting and welding) and Telecom-Infrared (1550 nm, used for internet data).

The problem? In the world of lasers, mixing two different "fuel types" (Ytterbium and Erbium ions) inside the same tiny glass tube is like trying to mix oil and water. Usually, they fight for energy, or one steals all the power from the other, making the laser unstable or weak.

The Solution: Instead of mixing the fuels, the scientists built a structured core. Think of it like a fruit tart. Instead of blending the strawberries and blueberries into a mush, they arranged them in distinct, separate spots within the same pastry.

How They Did It: The "Lego" Approach

The Ingredients: They made two types of glass rods. One set was doped with Ytterbium (the "1042 nm" fuel) and the other with Erbium (the "1550 nm" fuel).
The Assembly: They didn't melt them together. Instead, they used a technique called "Stack and Draw." Imagine taking 7 tiny glass straws (2 of one color, 5 of the other) and bundling them tightly inside a larger glass tube.
The Stretch: They heated this bundle and pulled it until it was as thin as a human hair. The result? A single fiber where the two types of glass ions are neighbors but spatially separated. They are close enough to share the same "room" (the fiber core) but far enough apart that they don't bump into each other and cause a fight.

They made two versions:

The 7-Rod Team: A small bundle with 2 Ytterbium rods and 5 Erbium rods.
The 19-Rod Team: A larger, more crowded bundle with 7 Ytterbium rods and 12 Erbium rods.

The "Traffic Controller" (The Length Trick)

Here is the coolest part of the discovery. Even though the fuel is separated, the laser light has to travel through the whole fiber to get out.

If the fiber is too long: The "1042 nm" light gets tired and fades out before it reaches the end, while the "1550 nm" light keeps going strong.
If the fiber is too short: The "1550 nm" light hasn't had enough time to build up power, so the "1042 nm" light dominates.

The Magic: The scientists realized they could act like a traffic controller. By simply cutting the fiber to a specific length, they could balance the power.

They found that for the 7-rod fiber, a length of 2 meters was the "Goldilocks zone" where both colors came out with almost equal strength.
For the 19-rod fiber, the sweet spot was slightly longer (2.15 meters).

Why This Matters (The "Bottleneck" Analogy)

In traditional mixed fibers, there is a problem called the "Bottleneck Effect." Imagine a busy highway where the Erbium ions are stuck in a traffic jam, waiting for the Ytterbium ions to pass them energy. If the traffic gets too heavy (high power), the jam gets worse, and the system crashes or becomes unstable.

By separating the ions into their own "lanes" (the structured core), the scientists removed the traffic jam.

No more fighting: The ions don't steal from each other.
Stability: The laser works smoothly even at different power levels.
Control: You can tune the output just by changing the length of the fiber, like turning a dial.

The Results

The team successfully built a laser that shoots two colors simultaneously:

1042 nm: Great for industrial applications.
1550 nm: Great for telecommunications.

They proved that by carefully arranging the "ingredients" and cutting the fiber to the right length, they could get a balanced, stable, dual-color laser. This opens the door for new tools in medical imaging, better internet connections, and advanced scientific sensors, all from a single, tiny piece of glass.

In a nutshell: They stopped mixing the ingredients and started arranging them like a mosaic. This allowed them to control the laser's output simply by changing the size of the canvas (the fiber length), creating a stable, dual-color light source that was previously very hard to achieve.

1. Problem Statement

Dual-wavelength fiber lasers operating simultaneously at ~1040 nm (Yb³⁺) and ~1550 nm (Er³⁺) are critical for applications in photonic component characterization, optical sensing, and microwave generation. However, traditional co-doped fibers (where Yb³⁺ and Er³⁺ ions are mixed in the same glass matrix) face significant challenges:

The "Bottleneck" Effect: In standard co-doped fibers, energy transfer from Yb³⁺ to Er³⁺ is highly efficient but difficult to control. At high pump powers, the Er³⁺ population saturates, preventing further excitation and leading to unstable operation or fiber damage.
Diffusion and Clustering: In standard silica matrices, high doping concentrations often lead to ion clustering and quenching. Furthermore, in nanostructured fibers with thousands of elements, high-temperature drawing causes diffusion of dopants (specifically Al₂O₃ and rare earths), distorting the designed spatial separation and degrading performance.
Lack of Independent Control: Existing solutions often require complex gain equalization or specific seed signal injection to achieve stable dual-wavelength lasing, particularly at lower power levels.

2. Methodology

The authors propose a novel structured-core fiber design that spatially separates Yb³⁺ and Er³⁺ doping regions within a single silica core to prevent direct ion-ion interaction while maintaining a single-mode beam profile.

Design Concept: Instead of mixing ions or using thousands of nano-elements, the core is constructed using a limited number of discrete doped rods (7 and 19 elements) arranged in a hexagonal lattice. This minimizes diffusion effects during the drawing process.
Fabrication Process:
1. Preform Preparation: Two types of preforms were created using the Modified Chemical Vapor Deposition (MCVD) method combined with nanoparticle doping. One set contained Yb³⁺ in a silica-alumina matrix, and the other contained Er³⁺ in the same matrix.
2. Etching: Preforms were etched in hydrofluoric acid to adjust the ratio of the doped core area to the silica cladding, optimizing the refractive index profile and overlap factors.
3. Stack and Draw: The etched preforms were drawn into sub-rods (~390 µm). These rods were then stacked into a silica capillary (7 rods for the first design, 19 for the second) and overcladded to form the final preform.
4. Final Drawing: The preforms were drawn into fibers at ~1940°C, resulting in a core diameter of ~7.2–8.1 µm and a 125 µm cladding.
Numerical Modeling: A rate equation model was developed to optimize the average concentration ratio of Yb³⁺ to Er³⁺ and the fiber length. The model assumed no energy transfer between the spatially separated ions, relying solely on the 980 nm pump source.
Experimental Setup: A core-pumped Fabry-Perot laser cavity was used. A 974 nm pump diode (450 mW max) was coupled into the fiber. The cavity mirrors were formed by a High-Reflectivity Fiber Bragg Grating (HRFBG) reflecting both 1042 nm and 1550 nm, and a cleaved fiber end acting as a Fresnel output coupler (~3.5% reflectivity).

3. Key Contributions

Novel Spatial Separation: The paper demonstrates a proof-of-concept for a structured-core fiber where Yb³⁺ and Er³⁺ are physically separated into distinct rods within a single core. This eliminates the "bottleneck" effect and uncontrolled energy transfer found in co-doped fibers.
Mitigation of Diffusion: By using a limited number of rods (7 and 19) rather than hundreds or thousands (as in nanostructured fibers), the authors successfully minimized the diffusion of dopants during the high-temperature drawing process, preserving the intended spatial distribution.
Tunable Power Ratio: The study proves that the output power ratio between the two wavelengths can be precisely controlled by adjusting the length of the active fiber, offering a simple mechanism for balancing dual-wavelength output without complex external control systems.
High-Quality Fabrication: The successful fabrication of two distinct fiber types (7-rod and 19-rod cores) with high rare-earth concentrations (up to ~5000 ppm) without significant quenching.

4. Results

Structural Characterization:
- SEM Analysis: Confirmed the spatial separation of the doped rods. The 7-core fiber had 2 Yb³⁺ rods and 5 Er³⁺ rods; the 19-core fiber had 7 Yb³⁺ and 12 Er³⁺ rods.
- Diffusion Observation: Rod diameters increased by ~30–48% during drawing due to thermal diffusion, but the spatial separation remained intact.
- Refractive Index: Measured NA was 0.12 (7-core) and 0.108 (19-core).
Optical Properties:
- Concentration Ratios: Measured via absorption, the total rare-earth concentration ratios ( $N_{tYb}:N_{tEr}$ ) were 52:48 for the 7-core fiber and 56:44 for the 19-core fiber.
- Fluorescence Lifetimes:
  - Er³⁺: ~10.3 ms (indicating no concentration quenching).
  - Yb³⁺: 0.84 ms. This is a critical result; a longer Yb³⁺ lifetime (compared to 0.76 ms in previous high-element nano-structures) indicates minimal clustering and successful prevention of harmful energy transfer processes.
- OH⁻ Content: Low levels were measured (3.1 ppm for 7-core, 0.72 ppm for 19-core).
Laser Performance:
- Dual-Wavelength Operation: Both fiber types successfully lased simultaneously at 1042 nm and 1550 nm.
- Optimal Length:
  - 7-core: Optimal length was 2.0 m, yielding a peak power difference of only 0.5 dB between wavelengths.
  - 19-core: Optimal length was 2.15 m, yielding a peak power difference of ~1 dB.
- Output Power & Efficiency (at 220 mW pump):
  - 7-core (2m): 1042 nm (27.7% slope efficiency), 1550 nm (16.4% slope efficiency).
  - 19-core (2.15m): 1042 nm (55.98% slope efficiency), 1550 nm (11.5% slope efficiency).
- Control: Shortening the fiber length increased the 1550 nm output relative to 1042 nm, while lengthening it favored 1042 nm.

5. Significance

This work represents a significant step forward in the development of stable, high-power dual-wavelength fiber lasers. By decoupling the Yb³⁺ and Er³⁺ gain media spatially within a single-mode fiber, the authors overcome the inherent instability and "bottleneck" limitations of traditional co-doped fibers.

The ability to control the output power ratio simply by trimming the fiber length offers a practical, low-cost solution for applications requiring specific dual-wavelength power balances, such as:

Microwave/Mid-IR Generation: Generating difference frequencies for spectroscopy.
5G Radio-over-Fiber: Transmitting millimeter waves.
Satellite Communications: Providing low SWaP (Size, Weight, and Power) amplifiers.

The methodology also opens the door for creating specialized active fibers with other rare-earth combinations (e.g., Thulium or Holmium) by simply adjusting the number and arrangement of doped rods, providing a versatile platform for future photonic devices.