As the word suggests, it is an effect which connects the magnetic ordering of a material (the north/south orientation of a magnet) to its elasticity, the ability of a material to change its volume when subjected to pressure or stress.
This ability depends on the electric properties of the material, since it is due to the electrons that glue atoms together. Hence, the magnetoelastic effect connects the magnetic and the electric  forces in matter. Its detailed microscopic description is very complex because one needs to go beyond many of the approximations typically used in condensed matter physics.

Two unexpected consequences of magnetoelasticity in two rather different materials
In two back-to-back papers appeared in Physical Review Letters, we have been able to measure two unexpected consequences of magnetoelasticity in two rather different materials, which advance our understanding of the phenomenon. In both cases, the key experiments were done in the laboratory at Fysikum at Stockholm University, in collaboration with researchers from different institutions: University of Konstanz, TU Dortmund, Tokyo Institute of Technology, Mainz University, CNRS Thales, Ca’ Foscari University of Venice.

In the first paper, we investigated the archetypal antiferromagnet nickel oxide, NiO. With a carefully tuned infrared laser pulse lasting a few tens of femtoseconds, resonant with an electronic transition in the material, we aimed at triggering the excitation of a coherent oscillation of the magnetic order. We succeeded, but to our great surprise we measured not one, but two oscillations, with two very different frequencies, and we noted that the two oscillations interacted, “talked” to each other. Textbook physics would suggest that this is theoretically impossible. (For those who have studied quantum mechanics, it is because they are two orthogonal states of the relevant Hamiltonian.)

Photon energy image. Credit Davide Bossini
Photon energy image. Credit Davide Bossini

We were able to understand that this is possible if one includes in the theoretical description both a nonlinear regime of the dynamics and the magnetoelastic effect. Magnetoelasticity allows the two oscillations to interact through a so-called domain wall, which separates areas of the sample with different magnetic ordering. The figure shows this idea schematically.

In he second paper, we looked at an ultrathin (a few nanometer thick) magnetic film of a ferrimagnetic material called yttrium iron garnet (YIG) which was doped with bismuth. This material was used decades ago in magnetooptical disks, but it is creating renewed interested since we are now able to realise it in ultrathin form. We used femtosecond laser pulses to trigger and observe the magnetization dynamics, again measured in the form of an oscillation of the magnetic order. No one had done it on this material before, and we wanted to try and study the details of the dynamics. To our great surprise, when we increased the intensity of the laser pulses, the frequency of the oscillations, instead of decreasing as expected by textbook physics, started to increase. We were able to explain this experimental evidence realising that the light pulses induce a mechanical pressure in the film and that elastic, electric energy is transformed in magnetic energy via the magnetoelastic effect.

Photon energy image. Credit Davide Bossini
Image credit: Davide Bossini Image credit: Davide Bossini

Why are these works relevant for physics and for society? One of the grand challenges of condensed matter physics in the past decades has been the one of controlling magnetism, which allows for information to be stored without energy cost, with electric fields, which offer the opportunity of minimising the cost of writing that magnetic information. Nowadays, in the data centers that contain all our digital information, the logical bits “1” and “0” are written by reversing the orientation (north/south poles) of tiny magnets by applying magnetic fields, not electric ones.

Why is that a problem?
Because magnetic fields cannot be focused properly (due to their zero divergence), and hence much energy is required to create them so that they can reverse the orientation of those tiny magnets. If, on the other hand, the magnetisation orientation could be controlled via modification of the electronic band structure, which determines the magnetic order, the energy dissipation can be greatly reduced. Furthermore, the electronic structure can respond to electric fields at a rate which is orders of magnitude faster than the magnetic response, under realistic magnetic fields. This would allow to realise digital storage which is both faster and more energy efficient. This requires of course that we better understand the fundamental physics that would allow for such devices.

Our studies in two rather different magnetic systems deepens our broad understanding of the complex problem of how electric forces, this time in the form of pressure that distorts the atomic orbitals, affect the magnetic order and its dynamics. They also are examples of the counterintuitive and fascinating phenomena that can appear when nature is observed in nonlinear and nonequilibrium regimes which can be attained with ultrafast laser pulses.

— Stefano Bonetti

[1] D. Bossini, M. Pancaldi, L. Soumah, M. Basini, F. Mertens, M. Cinchetti, O. Gomonay, and S.
Bonetti, Ultrafast   Amplification   and   Nonlinear   Magnetoelastic   Coupling   of   Coherent   
Magnon   Modes   in   an Antiferromagnet, Physical Review Letters 127, 077202 (2021)

[2]  L.  Soumah,  D.  Bossini,  A.  Anane,  S.  Bonetti,  Optical  Frequency  Up-Conversion  of  
the  Ferromagnetic Resonance  in  an  Ultrathin  Garnet  Mediated  by  Magnetoelastic  Coupling,  
Physical  Review  Letters  127, 077203 (2021)