🔬 atomic physics

Characterisation of a triple-species 87Rb/85Rb/133Cs magneto-optical trap towards cold-atom interferometry

This paper demonstrates the simultaneous trapping and cooling of $^{85}$ Rb, $^{87}$ Rb, and $^{133}$ Cs in a triple-species magneto-optical trap using a compact all-fibre laser system, characterizing interspecies losses to confirm the system's suitability for future multi-species cold-atom interferometry applications.

Original authors: Mal Landru, Amandine Lauret, Yannick Bidel, Alexandre Bresson, Sylvain Schwartz, Alexis Bonnin, Antoine Godard, Nassim Zahzam

Published 2026-08-20

📖 4 min read☕ Coffee break read

CC BY 4.0

Original authors: Mal Landru, Amandine Lauret, Yannick Bidel, Alexandre Bresson, Sylvain Schwartz, Alexis Bonnin, Antoine Godard, Nassim Zahzam

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

In the quiet world of atomic physics, scientists have long sought to harness the strange behavior of matter at temperatures near absolute zero. When atoms are cooled to such extremes, they slow down and begin to act less like tiny billiard balls and more like waves, allowing researchers to measure time, gravity, and rotation with incredible precision. This field, known as cold-atom interferometry, relies on trapping clouds of atoms using lasers and magnetic fields. For years, these experiments have typically used a single type of atom, such as rubidium or cesium. However, a new frontier is opening up: mixing different types of atoms together in the same trap. Doing so could allow for more sophisticated measurements, such as testing whether gravity affects different masses in exactly the same way, or building sensors that can measure motion in multiple directions at once. The challenge has been that different atoms often interfere with one another, colliding and knocking each other out of the trap, which makes keeping a mixed group stable extremely difficult.

A team of researchers at ONERA in France has now taken a significant step toward solving this problem by successfully trapping and cooling three different types of atoms simultaneously: two isotopes of rubidium and one of cesium. They built a specialized laboratory setup designed to be compact and robust, using a laser system based on telecommunications technology that is small enough to potentially fit on a moving vehicle. Instead of using bulky, complex equipment, their system relies on a network of fiber-optic cables and just four laser diodes. By carefully converting the light from these lasers, they generated the specific colors of light needed to catch and cool all three species of atoms at the same time. The result is a "magneto-optical trap" that holds a cloud containing roughly one hundred million atoms of each type, all living together in a tiny glass cell under an ultra-high vacuum.

The researchers were particularly interested in seeing how these different atoms would behave when forced to share the same space. In previous experiments with two types of atoms, scientists had observed that the atoms would sometimes collide and escape the trap, reducing the total number of atoms held. To understand if this would be a problem for their three-species mixture, the team ran a series of tests. They first loaded the trap with just one type of atom, then added a second, and finally added the third, carefully counting how many atoms remained at each stage. They discovered that the two types of rubidium atoms, which are chemically very similar, did not disturb each other at all; they coexisted peacefully without causing any loss. However, when cesium was introduced, it did cause some friction. The presence of cesium caused a small number of rubidium atoms to be knocked out of the trap, and vice versa.

Despite these collisions, the mixture remained stable. The team calculated that the loss of atoms due to these interactions was very small, amounting to less than seven percent of the total population. This means that the system is robust enough to hold a large, mixed cloud of atoms for the duration needed to perform complex measurements. The researchers also looked for evidence of more complex interactions, where three atoms might collide at the exact same time, but found no sign of such events occurring. Their measurements showed that the rate at which atoms were lost was consistent with what is known about collisions between rubidium and cesium, confirming that their laser system was working exactly as intended.

This work demonstrates that it is possible to create a stable, multi-species environment using a compact, all-fiber laser system, which is a crucial requirement for future applications. The ability to trap these three species together without losing a significant portion of them suggests that scientists can now move toward building advanced sensors that use multiple types of atoms simultaneously. Such sensors could eventually be used to measure acceleration and rotation with unprecedented accuracy, potentially leading to new navigation systems that do not rely on satellites. The study confirms that while the atoms do interact, the interactions are manageable, paving the way for a new generation of quantum instruments that can operate in the real world.

Technical Summary: Characterisation of a Triple-Species 87Rb/85Rb/133Cs Magneto-Optical Trap

Problem Statement
The development of multi-species cold atom systems is critical for advancing quantum technologies, including precision inertial sensing, fundamental constant measurements, and quantum simulation. While double-species mixtures and triple-species Magneto-Optical Traps (MOTs) have been demonstrated previously, no existing triple-species apparatus has been shown to be compatible with high-precision inertial measurements based on cold-atom interferometry. A significant barrier to onboard applications is the complexity and bulk of laser systems required to address the numerous atomic transitions of different species. Furthermore, the impact of interspecies collisions on trap stability and atom number in a triple-species configuration remains a key variable to characterize for future interferometric applications.

Methodology
The authors present an experimental apparatus designed for the simultaneous trapping and cooling of three atomic species: $^{85}$ Rb, $^{87}$ Rb, and $^{133}$ Cs.

Laser System: The system utilizes a compact, all-fibre architecture based on telecom 1.5 $\mu$ m and 2 $\mu$ m technologies. Four tunable laser diodes serve as the primary sources. Frequencies are up-converted via non-linear stages: Second Harmonic Generation (SHG) of a 1560 nm source yields 780 nm for Rubidium, while Sum Frequency Generation (SFG) of 1560 nm and 1878 nm sources yields 852 nm for Cesium. Phase modulation via Electro-Optic Modulators (EOMs) generates the necessary cooling and repumping lines for all three species.
Vacuum and Trapping: The MOTs are formed in a glass cell within an ultra-high vacuum (UHV) chamber ( $\sim 10^{-9}$ mbar) using commercial dispensers for the atomic sources. The trap employs a retro-reflected optical configuration with a magnetic quadrupole field gradient of 15 G/cm.
Imaging and Diagnostics: Fluorescence is monitored using photodiodes with narrow-band filters (780 nm and 852 nm) and a CMOS camera for spatial imaging. The density and atom numbers are estimated by fitting 2D Gaussian functions to the camera data.
Experimental Protocol: The study involves loading sequences for single-species, double-species, and triple-species MOTs. By alternating the repumping lines, the authors measure loading curves and steady-state atom numbers to quantify interspecies interactions. The data is analyzed using rate equations to extract two-body and three-body trap-loss coefficients.

Key Results

Trapping Performance: The system successfully traps and cools approximately $10^8$ atoms for each species simultaneously. The loading times are approximately 159 ms for $^{133}$ Cs, 162 ms for $^{85}$ Rb, and 168 ms for $^{87}$ Rb.
Interspecies Losses:
- Rubidium-Rubidium ( $^{85}$ Rb/ $^{87}$ Rb): No significant losses were observed between the two Rubidium isotopes in double-species or triple-species configurations. The trap-loss coefficient is bounded at $\beta_{85,87} < 1.9 \times 10^{-12}$ cm $^3$ /s.
- Rubidium-Cesium: Measurable losses occur between Rubidium and Cesium. In double-species MOTs, adding Cesium to a Rubidium MOT reduced the Rubidium atom number by 1.6% ( $^{85}$ Rb) and 6.7% ( $^{87}$ Rb). Conversely, adding Rubidium to a Cesium MOT reduced the Cesium count by 1.3% ( $^{85}$ Rb) and 6.4% ( $^{87}$ Rb).
- Coefficients: The calculated two-body interspecies trap-loss coefficients are $\beta_{87,Cs} \approx \beta_{Cs,87} = 1.1 \times 10^{-11}$ cm $^3$ /s and $\beta_{85,Cs} \approx \beta_{Cs,85} = 2.6 \times 10^{-12}$ cm $^3$ /s.
- Triple-Species Dynamics: In the triple-species MOT, the total atom number loss relative to single-species operation is less than 7% for each species. The ratio of $^{85}$ Rb to $^{87}$ Rb shifts from 60/40 (in the absence of Cesium) to 65/35, attributed to the stronger loss interaction between $^{87}$ Rb and $^{133}$ Cs.
Three-Body Collisions: No significant three-body trap-loss collisions were observed. An upper bound for the three-body coefficient was calculated as $\beta_{i,j,k} < 1.5 \times 10^{-23}$ cm $^6$ /s.

Significance and Claims
The paper claims that the demonstrated system is suitable for future multi-species cold-atom interferometry applications. The primary significance lies in two areas:

Compactness: The all-fibre laser system based on telecom wavelengths offers a robust and compact solution for onboard inertial measurements, addressing the bulkiness typically associated with multi-species laser setups.
Viability for Interferometry: The characterization of interspecies losses indicates that while heteronuclear collisions exist, they result in a maximum atom loss of approximately 7%. The authors argue that in the considered regime, these inelastic collisions are not expected to significantly impact the performance of a cold-atom interferometer. However, they note that collisions not leading to trap loss (e.g., elastic or $m_F$ -changing collisions) could still induce frequency shifts and affect accuracy, a topic requiring further study.

The work validates the operation of the original laser system for standard atom cooling and trapping and confirms that the experimental setup is compatible with the requirements for future triple-species interferometers.

Technical Summary: Characterisation of a Triple-Species 87Rb/85Rb/133Cs Magneto-Optical Trap

More like this