XMCD fingerprints expose altermagnetism

Altermagnets are a new class of magnetic materials that possess the useful properties of both ferromagnets and antiferromagnets, potentially making them very useful in future electronic devices based on spintronics – technology that uses the spin-state of electrons to carry information. However, identifying altermagnetism is challenging. X-ray magnetic circular dichroism (XMCD) measures the difference in absorption of left- and right-circularly polarised X-rays. In work recently published in Physical Review Letters, an international team led by Associate Professor Atsushi Hariki from Osaka Metropolitan University and Professor Peter Wadley of the University of Nottingham theoretically predicted an altermagnetism fingerprint in manganese telluride (α-MnTe), then used XMCD on Diamond’s I06 beamline to detect it experimentally. Their results demonstrate that XMCD is an effective method of identifying altermagnetic materials, a discovery that could accelerate the use of these new properties in next-generation electronic devices.

Traditionally, collinear magnetic materials have been classified as ferromagnets or antiferromagnets. Magnetism arises from electron spin, and in ferromagnets the spins all point in the same direction. In antiferromagnets, they point in opposite directions, cancelling each other out. Altermagnets also have an opposing arrangement of spins. However, unlike antiferromagnets, their crystal symmetries give rise to an electronic band structure with a spin polarisation that alternates as you pass through the energy bands.

We are familiar with the use of ferromagnets for data storage. In hard drives, for example, small particles of ferromagnetic material form tiny magnets with a magnetic field that extends outwards and can be read by a magnetic head. Writing data involves switching the alignment of the tiny magnets. There’s a limit to how closely packed the tiny magnets can be before their magnetic fields start to interfere, producing a physical limit to storage capacity. The stored data is also vulnerable to being wiped by an external magnetic field.

Antiferromagnets aren’t vulnerable to magnetic fields, so data stored in these materials is more secure. Switching the bits is potentially much quicker in antiferromagnets than ferromagnets. Antiferromagnets don’t create stray magnetic fields that can interfere and corrupt the data, so their theoretical storage capacity is higher. However, writing and reading the data back is a challenge.

Prof Jan Kuneš of Masaryk University, Czech Republic, said:

XMCD is a standard way for investigating ferromagnetic materials, which shows the difference in absorption between left- and right-circularly polarised light. In a non-magnetic sample, and in antiferromagnets, no difference is observed. However, if the spins in an altermagnet are aligned, a signal is observed.

Prof Kuneš explained:

The Mn site-resolved contributions to XMCD calculated by the LDA+DMFT AIM for the Néel vector 𝐋∥[1⁢1¯⁢00] and the light propagation vector 𝐤⟂𝐋 (a) and 𝐤∥𝐋 (b). (c) A cartoon view along the c-axis of the possible domain structure with the six easy-axis orientations of 𝐋. The domains with even labels (red) contribute Δ⁢F⁢(ω) with positive prefactor, the odd ones (blue) with negative prefactor. The red and blue dots indicate the positive and negative orientation of the out-of-plane magnetization 𝐦. (d) The out-of-plane canting of the moments in 6 T applied field does not strongly depend on the domain’s 𝐋 (The domain sizes and shapes were chosen randomly and are not intended to have physical meaning.)

If you cool down altermagnetic a-MnTe, the magnetic moments align in one of six different directions, and regions in which the moments align in one particular direction are called domains. Using this new XMCD technique, researchers are now able to detect the domains and visualise their location in the sample, things that previously have been extremely difficult.

 Prof Kuneš added:

To find out more about the I06 beamline or discuss potential applications, please contact Principal Beamline Scientist Larissa Ishibe-Veiga: [email protected].

Image credits: arXiv:2305.03588v2 [cond-mat.mtrl-sci] under license CC BY 4.0