🔬 atomic physics

Light-induced spin-polarized desorption of Rb atoms from Co surfaces

This study demonstrates that pulsed UV-light irradiation induces non-thermal desorption of spin-polarized rubidium atoms from a spin-polarized cobalt (110) surface, confirming the occurrence of spin transfer between the surface and the adsorbate during the process.

Original authors: Kanta Asakawa, Naoki Tanabe, Oki Watanabe, Shuji Kamada, Keisuke Hara, Kaori Niki, Atsushi Hatakeyama

Published 2026-08-03

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Kanta Asakawa, Naoki Tanabe, Oki Watanabe, Shuji Kamada, Keisuke Hara, Kaori Niki, Atsushi Hatakeyama

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 a world where atoms aren't just tiny, neutral marbles, but little magnets with a secret "spin" direction, like a top spinning clockwise or counter-clockwise. In the realm of surface science, researchers study what happens when these atomic marbles land on a magnetic surface, like a giant, invisible dance floor made of cobalt. Usually, when atoms stick to a surface and then fly off, they lose their magnetic secrets, becoming a chaotic, un-polarized crowd. But what if we could zap them with a laser so fast that they fly off before they have time to forget their spin? This question matters because understanding how electrons and spins move between a surface and an atom could help us build better catalysts for making fuel or create new sources of polarized atoms for advanced technology. The key players here are "spin polarization" (atoms having a preferred spin direction), "light-induced desorption" (using light to kick atoms off a surface), and the idea that if the surface is magnetic, the atoms leaving it might carry that magnetic memory with them.

In this study, a team of scientists decided to test this idea by playing a high-speed game of "kick-off" with Rubidium (Rb) atoms and a Cobalt (Co) surface. They set up a vacuum chamber, which is like a giant, airless room, and grew a thin, magnetic film of cobalt on a crystal base. Then, they sprinkled Rubidium atoms onto this cobalt floor. Once the atoms were settled, the researchers hit the surface with a super-fast pulse of ultraviolet (UV) light, acting like a microscopic slingshot. The goal was to see if the Rubidium atoms would fly off, and if so, whether they kept their "spin" aligned with the cobalt magnet underneath.

To catch these flying atoms, the team used a clever trick involving "circularly polarized light," which is like a flashlight beam that twists as it travels. They tuned this light to match the exact frequency that Rubidium atoms love to absorb. Because of a quirk in quantum physics, Rubidium atoms with a specific spin direction will gobble up this twisting light more eagerly than those spinning the other way. By measuring how much light the flying atoms swallowed, the scientists could tell if the atoms were spinning in a coordinated way or just randomly. They also tracked how fast the atoms were moving to figure out how they were being kicked off.

The results were exciting. The team found that when the UV light hit the cobalt, the Rubidium atoms did indeed fly off, and they were spin-polarized. This means the atoms leaving the surface weren't a random mix; they were mostly spinning in the same direction, matching the magnetic orientation of the cobalt surface. The researchers calculated that the average spin value was small but definitely not zero, confirming that the atoms carried a magnetic signature from the surface.

However, the story gets more interesting when looking at how they flew off. If the atoms were just getting hot and boiling away (a thermal process), they would have moved at speeds consistent with a hot gas. Instead, the scientists observed that the atoms were moving incredibly fast—up to 500 meters per second initially—and that this speed actually slowed down over time as more atoms piled up on the surface. This behavior suggested a non-thermal mechanism. It's as if the light didn't just heat the atoms up; it triggered a specific electronic event. The team's computer simulations (using a method called Density Functional Theory) suggested that the UV light excited electrons in the cobalt, which then jumped onto the Rubidium atoms. This electron transfer neutralized the Rubidium and gave it a "kick" to fly off. Crucially, because the cobalt surface is magnetic, the electrons it gave up were spin-polarized, and this spin was transferred to the Rubidium atom just before it took flight.

The paper is careful to note that this isn't a perfect, crystal-clear picture of every single atom. The team suspects that as the Rubidium layer got thicker (forming multiple layers), the atoms on top might not have received the spin-polarized electrons directly from the cobalt, which is why the measured spin polarization wasn't as high as the maximum possible value. They also ruled out the idea that stray magnetic fields from the sample were tricking their measurements, showing that the effect was real and came from the desorption process itself.

In short, this paper demonstrates that by zapping a magnetic surface with light, you can launch atoms into the air while preserving their magnetic spin. It's like a magic trick where the magician (the cobalt surface) passes a secret code (the spin) to the assistant (the Rubidium atom) just as the assistant is launched into the spotlight. While the exact dance steps of the electrons are still being figured out, the experiment proves that light-induced desorption is a powerful way to study how spin and charge move between surfaces and atoms, opening the door to understanding magnetic catalysts and potentially creating new tools for spin-based technology.

Technical Summary: Light-Induced Spin-Polarized Desorption of Rb Atoms from Co Surfaces

Problem Statement
Interactions between gaseous atoms and magnetic surfaces often alter the spin states of both phases. While charge transfer between magnetic surfaces and adsorbates is known to modify surface magnetic properties (e.g., demagnetization or enhanced spin polarization of conduction electrons), the transfer of spin during the desorption process remains insufficiently investigated. Detecting spin transfer between a surface and desorbing adsorbates is experimentally challenging. Specifically, in non-thermal, light-induced desorption processes, desorption occurs faster than spin relaxation, theoretically allowing for the observation of spin-polarized desorbed atoms. However, a suitable detection method to confirm spin transfer during desorption from ferromagnetic surfaces has been lacking, hindering the elucidation of spin-related surface chemical reactions.

Methodology
The study employed an experimental setup combining ultra-high-vacuum (UHV) techniques with spin-selective optical detection and Density Functional Theory (DFT) calculations.

Sample Preparation: An epitaxial face-centered cubic (fcc) Co (110) thin film (10 nm) was grown on an MgO (110) substrate. The sample was magnetized along the [001] easy axis using a pulsed current source to generate a 1.7 kOe field.
Desorption Induction: Rubidium (Rb) atoms were adsorbed onto the Co surface via an effusive beam source. Desorption was induced by pulsed UV irradiation (355 nm, 5–10 ns pulses, 65.0 mJ/cm² fluence) at a 45° angle.
Spin Detection: The spin polarization of the desorbed Rb atoms was measured using the spin-selective absorption of circularly polarized light. A probe laser, tuned to the $F=3 \to F'=4$ transition of the $^{85}$ Rb D2 line, was converted into circularly polarized light (switching helicity every 20 UV pulses). The absorbance of this light by the desorbed atoms was detected via a photodetector.
Data Analysis: The number of desorbed atoms ( $I_{des}$ ) and their mean velocity ( $v_z$ ) were derived from the time-dependent absorbance. The spin polarization was quantified by the ratio of desorption intensities under left ( $\sigma^+$ ) and right ( $\sigma^-$ ) circularly polarized light ( $I_{des,\sigma^+}/I_{des,\sigma^-}$ ). This ratio was mapped to the average magnetic quantum number $\langle m_{F=3} \rangle$ using a Maxwell-Boltzmann velocity distribution model.
Theoretical Support: DFT calculations (VASP, GGA-RPBE) were performed on a Rb/Co(110) slab model to analyze the electronic structure, adsorption energy, and density of states (DOS) near the Fermi level.

Key Results

Non-Thermal Desorption Mechanism: The desorption of Rb from the Co surface commenced immediately upon Rb adsorption, contrasting with threshold behaviors seen in other systems. The mean velocity ( $v_z$ ) of desorbed atoms decreased over time (from ~500 m/s to ~357 m/s), a trend inconsistent with thermal desorption (where velocity typically increases or remains constant with coverage). This suggests a non-thermal mechanism driven by electronic transitions and charge transfer, similar to mechanisms observed in K desorption from Cr $_2$ O $_3$ .
Observation of Spin Polarization: The ratio $I_{des,\sigma^+}/I_{des,\sigma^-}$ deviated significantly from unity for the Co/MgO sample, with the direction of deviation reversing when the sample magnetization was reversed. Control experiments on a pure MgO substrate showed no such deviation, confirming the effect originated from the magnetized Co surface.
Quantified Spin State: The calculated average magnetic quantum number $\langle m_{F=3} \rangle$ was $-0.024 \pm 0.003$ for the "anti-parallel" configuration and $0.024 \pm 0.004$ for the "parallel" configuration. This indicates that the desorbed Rb atoms were spin-polarized parallel to the sample's magnetization direction. Since the majority electron spin in Co is anti-parallel to the magnetization, the desorbed atoms align with the minority-spin direction.
Electronic Structure Insights: DFT results showed that the Co 3d states near the Fermi level are polarized along the minority-spin direction. The Rb 5s orbital lies above the Fermi level, indicating electron donation from Rb to the substrate upon adsorption. The observed spin polarization suggests that during UV irradiation, spin-polarized electrons from the Co 3d states (near the Fermi level) are transferred to the Rb atoms, neutralizing them and triggering desorption.

Significance and Claims
The paper claims to provide direct experimental evidence of spin transfer between a ferromagnetic surface and adsorbates during the desorption process. By demonstrating that light-induced desorption of Rb from a Co surface yields spin-polarized atoms, the study validates the hypothesis that non-thermal desorption mechanisms can preserve spin information transferred from the substrate.

The authors state that their method offers a viable approach to detect spin transfer during desorption, addressing a long-standing gap in surface science where the spin state of desorbed atoms has remained unclear due to detection limitations. They assert that these findings provide insights into the mechanisms of adsorption and light-induced desorption from spin-polarized surfaces. While the method is limited to optically detectable atoms, the authors suggest it holds potential for addressing fundamental questions regarding spin behavior during charge transfer in adsorption/desorption processes and may elucidate spin-related surface reactions, such as the promotion of ferromagnetic catalysts by alkali metal adsorption. Additionally, they note the potential application of this phenomenon for creating polarized atomic sources.

The study acknowledges limitations, including the inability to measure hyperfine states other than $F=3$ and the possibility of multilayer adsorption effects reducing the observed polarization magnitude. The authors conclude that systematic studies comparing desorption cross-sections and velocity distributions at various wavelengths are necessary to further clarify the electron and spin dynamics involved.

More like this