Rare-Earth Spins on Surfaces


Spin qubits in QNS have to date primarily consisted of transition metal atoms of the 3d electronic configuration, such as iron, titanium and copper. These spins have the unpaired electrons in the outermost shell, which has advantages and disadvantages. On the pro-side, it offers the electrons that flow in the tunnel current of an STM a good chance to interact with the magnetic atoms. On the con-side, this same interaction with the environment also decreases the spin coherence times.

In this research effort, we are focused on the magnetic properties of rare earth spins on surfaces. Rare earth atoms contain a partially filled 4f electron shell and crucially, this 4f shell is hidden deeper inside the atom. Therefore their interaction with the environment is generally weaker. Also, because they consist of heavy atoms, the spin degree of freedom is locked to the angular momentum degree of freedom, which generally makes the interpretation of the magnetic behavior easier.

In this group, we rely on a combination of experimental and theoretical techniques. On the experimental side, we study the ensemble properties of the rare earth spins by using x-ray absorption spectroscopy, especially the magnetic technique, called X-ray Magnetic Circular Dichroism (XMCD). Our team regularly travels to the world’s best synchrotron radiation facilities to perform these experiments, both in Korea and in Europe. We also utilize STM studies to understand the absorption of the rare earth spins on surfaces. We are planning to utilize ESR-STM in the future. On the theory side we use multiplet calculations and density functional theory, which are described in more detail in the theory team’s pages.

Our goal is to identify surface-adsorbed rare earth atoms with quantum levels that are suitable for coherent manipulation and with long coherence time. In addition, we aim at engineering the occupation of their individual orbitals in order to optimize the detection of their magnetic signal using STM and ESR-STM. These atoms could be an excellent platform to test advanced pulse sequences for quantum logic and error corrections.

Longer-term Goals

• Utilize rare earth atoms to test quantum logic operations and error correction
• Obtain entanglement in multiple qubits consisting of rare-earth atoms

Near-term Goals

• Find rare-earth atoms with suitable quantum states and long coherence time
• Understand the impact of the environment (temperature, local electric and magnetic fields) on the stability of the quantum states
• Perform electron spin resonance on a single rare-earth atom on a surface

Research results to date

Research results to date

In 2021 the group proved the use of XMCD to detect the spins of rare-earth atoms resolving their electron orbital structure. Since the 4f electrons are hidden close to the nucleus, they are very well protected from external perturbations. However, when one tries to measure their spin using electric transport (such as in an STM), these 4f electrons provide a very little signal to the current, making their measurement rather difficult. However, they can be coupled to more external shells, such as 5d, 6s, or 6p, which can operate as an intermediary between the internal 4f spins and the flow of electric current across the atom. Using this newly developed orbital-resolved XMCD technique (Figure 1), we were able to map the presence or absence of spins in the different orbitals of gadolinium and holmium atoms deposited on MgO. This approach allows us to understand the mutual coupling between electrons in different orbitals and determine if an atom can provide a large signal in transport experiments. Accepted for publication at ACS Nano (2021).

Also in 2021 we identified a new single-atom rare earth magnet, dysprosium on MgO films on silver. Using STM, XMCD, and theoretical calculations, we could determine that its anisotropy barrier (one of the key properties determining the stability of magnetic orientation of an atom) is the largest ever measured for a spin on a surface, about twice as much as the previous record held by a holmium atom (Dy’s neighbor in the periodic table). The magnetic orientation of these atoms is stable for many hours at low temperatures. These atoms also show an interesting change in the charge state and magnetic stability when increasing the thickness of the oxide film onto which they are deposited. The change in charge state is due to a process called charge transfer, which occurs when the oxide film is just a few atomic layers thick. These results have been published in Nature Communications (2021) and another paper is submitted for publication.

In 2020, we determined the relaxation mechanism that limits the stability of holmium single-atom magnets. Holmium atoms have been the first atoms to show magnetic stability and are still the surface spins with the highest blocking temperature (45 K) up to date. The magnetic orientation of these atoms becomes less and less stable with increasing temperature. In this XMCD study, done in collaboration with the EPFL and ETH Zurich in Switzerland, we found that the reason for this instability is a thermally activated vibration that mostly involves the local oscillation of the Ho around its equilibrium position (Figure 2). This study is published in Physical Review Letters.


In 2019, in collaboration with the scientists at the Swiss Light Source, we unraveled the main magnetic relaxation processes that limit the magnetization stability of two very popular rare-earth-based molecules, TbPc2 and DyPc2, adsorbed on MgO films. In this study, we combined XMCD and theoretical calculations to understand the role of specific mechanisms that are well known to occur in solid-state materials, but much less understood for the cases of molecules on surfaces. The study was published in Advanced Science (2019).

In 2018, we collaborated with EPFL (Switzerland) to determine the limit of the magnetic stability of holmium atoms on MgO. Using spin-polarized STM, we could observe the switching of the magnetic orientation of individual Ho atoms as a function of time under the effect of increasing temperature and magnetic field. At the maximum available field, these atoms showed stability up to 45 K, a record-high value for a single spin on a surface. We also determined how holmium atom couples magnetically when in contact with another cobalt atom on the same magnesium oxide surface. By merging these two atoms in a so-called heterodimer, we observed for the first time intense spin-excitations on a rare-earth atomic-scale structure. By studying these excitations as a function of the magnetic field, we could understand the magnetic coupling between holmium and cobalt, and determine the quantum levels of the combined system. Based on these results we published two papers in Physical Review Letters (2018).

In 2018 and 2017, also in collaboration with the EPFL (Switzerland), we combined XMCD and multiplet calculations to determine the quantum levels of several rare-earth atoms deposited on metals and on graphene. By analyzing the X-rays spectra and comparing to simulations, we determined the number of electrons in the 4f orbitals for every element and surface in the study, identifying the interplay between intra-atomic interaction and coupling to the surface to determine the magnetic properties of rare-earth atoms. These studies have been published in two papers in Physical Review B .

*Fig.1 In orbital resolved XMCD, we can select specific X-ray transitions that can bring an electron from orbitals very close to the nucleus (3s, 3p and 3d) to outer orbitals (left). The signal coming from these transitions (right) tells how many electrons are present in each of the outer orbitals.

*Fig.2 the lifetime (τ) of Ho atoms depends on the magnetic field and temperature (top), due to oscillations of the Ho atom around its equilibrium position (bottom), which are activated when we increase the temperature.

Research Equipment

Swiss Light Source “X-treme” end-station
Paul Scherrer Institute, Villigen (Switzerland).
Key results
• Relaxation mechanisms of rare earth atoms and molecules
• Discovery of new single atom magnets
• Quantum levels of single lanthanide atoms
Temp: from 2K to 350 K
Magnet: up to 7 T (main axis), and up to 2 T (vector two axis)
Energy range: 400 – 2000 eV
In situ sample preparation
“Xtreme” end station, Switzerland (www.psi.ch)

X-treme offers very low temperature and high magnetic fields. The energy range is suitable for exploring the magnetic properties of transition metals and rare-earth atoms. A preparation chamber equipped with an STM is used to prepare and characterize samples before an XMCD experiment, and to transfer the prepared sample without breaking the vacuum. The very high signal-to-noise ratio allows measuring well-separated atoms and molecules (much less than one full layer). The group has a long-standing experience with this end-station and collected many results there.

ALBA “Boreas/Hector” end station
ALBA synchrotron, Barcelona (Spain).
Key results
• Orbital resolved XMCD of atoms on surface
Temp: from 3K to 350 K
Magnet: up to 6 T (main axis), and up to 2 T (vector three axis)
Energy range: 80 – 4000 eV
In situ sample preparation
“Hector” end station, Spain

The “Hector” end-station is part of the “Boreas” beamline. It is equipped with a 3-axis vector magnet and reaches very low temperatures. The preparation chamber equipped with an STM and other surface science diagnostics allows the preparation and characterization of clean surfaces prior to the XMCD experiments, and to transfer samples without breaking the vacuum. The special energy range of this beamline, as well as the excellent signal-to-noise ratio, allowed us to detect the spin of single rare-earth atoms and resolve them according to their orbitals.

Pohang Light Source II “6A” end station
Pohang Accelerator Laboratory, Pohang (Korea).
Key results
• Orbital resolved XMCD of molecular magnets
Temp: from 20K to 370 K
Magnet: up to 7 T (main axis)
Energy range: 250 – 3000 eV
In situ sample preparation
The “6A” endstation in Pohang.

The “6A” end-station allows measuring samples at low temperature and high magnetic field. It offers a wide energy range that allows resolving the spin properties of rare-earth-based molecular magnets. It is possible to produce a molecular film in situ using the preparation chamber. In collaboration with the local scientists, we have developed a vacuum suitcase to transport samples prepared in ultra-high vacuum chambers in QNS and measure them using X-rays. This strategy allows us to combine the characterization of the sample’s morphology and electronic properties with our low-temperature STMs and measure their magnetic properties using XMCD without contaminating them by exposure to air.