Excitation of nanoscale coherent magnons by EUV transient gratings

Spin waves, or magnons, are propagating excitations in magnetic materials. Just as sound waves involve oscillations of the density of the medium, spin waves involve oscillations (more precisely, precession) of the magnetization. Spin waves are important both for understanding the fundamentals of magnetization dynamics and for practical applications, as they affect the operation of magnetic devices at high frequencies. The emergent field of magnonics aims to harness spin waves for storing, transporting and processing information. Magnons with nanoscale (< 100 nm) wavelengths are particularly promising for these emerging applications. Currently, such short-wavelength magnons are generated by microwaves using nanoscale antennas fabricated on the sample. It is also possible to generate and probe spin waves non-invasively by laser radiation. However, with conventional laser sources one can only handle relatively long-wavelength magnons. To deal with nanoscale magnons, one needs lasers with nanoscale wavelength. The free electron laser FERMI in Trieste operates in the extreme ultraviolet (EUV), just in the right wavelength range for manipulating nanoscale spin waves.

A team of researchers from the U.S.A., Italy and Germany took advantage of the EUV transient grating (TG) setup available at the TIMER beamline at FERMI. In the TG setup, two femtosecond EUV pulses (called pump or excitation pulses) are crossed at the sample to form an interference pattern with a period in the range of tens of nanometers  (see Figure 1A). The interaction of the EUV radiation with the sample results in a spatially periodic material excitation at the same period. The ensuing dynamics are probed via diffraction of a time-delayed probe EUV pulse.

Figure 1: A) The experimental setup. The sample is placed in a tilted magnetic field and irradiated by a pair of pump EUV pulses which generate coherent spin waves with wavelengths 50 – 90 nm. The spin waves cause diffraction of a time-delayed probe pulse, and the diffracted signal is measured by a CCD camera. B) Examples of CCD images taken at different pump-probe delays. C) Integrated signal as a function of the pump-probe delay revealing spin wave oscillations. The inset shows the initial dynamics of the signal indicating EUV-induced demagnetization on the femtosecond time scale.

The TIMER setup has been used previously to generate nanoscale acoustic waves, or phonons. In this study, the researchers used a magnetic sample consisting of ultrathin layers of gadolinium and iron and placed it in a tilted magnetic field to excite spin waves. To facilitate the detection of spin waves, the wavelength of the probe pulse was tuned to a specific atomic transition of gadolinium sensitive to its magnetic state. Figure 1B shows representative images of the diffracted probe beam on the CCD camera at different pump-probe delays. At negative delays, i.e. when the probe pulse arrives right before the excitation, the signal is zero. When the probe arrives after the excitation, a periodic modulation of the magnetization in the sample diffracts the probe pulse towards the CCD camera. The peculiar signature of spin waves is the oscillatory behavior of the signal. This can be clearly observed in the plot in Figure 1C, which shows the integrated signal from the CCD camera versus the pump-probe delay. By varying the wavelength of the excitation pulses, the researchers were able to vary the magnon wavelength in the range 50 – 90 nm and measure its dispersion, i.e., the dependence of the magnon frequency on the radiation wavelength. No existing spectroscopic technique would be able to make such a measurement in this wavelength range. Furthermore, the researchers found that a minor variation in Gd and Fe layer thicknesses in the sample results in a large change in the spin wave frequencies and propagation velocities.

In summary, the study unlocks the potential of EUV radiation for the study of spin waves. Magnon spectroscopy based on the EUV TMG approach overcomes the limitations of existing techniques and provides the means to broaden our understanding of ultrafast magnetic dynamics at the nanoscale and to facilitate research towards high-speed magnonic devices.

Peter R. Miedaner¹, Nadia Berndt¹, Jude Deschamps¹, Sergei Urazhdin², Nupur Khatu^3,4,5, Danny Fainozzi³, Marta Brioschi^6,7, Pietro Carrara^6,7, Riccardo Cucini⁷, Giorgio Rossi^6,7, Steffen Wittrock^8,9, Dmitriy Ksenzov¹⁰, Riccardo Mincigrucci³, Filippo Bencivenga³, Laura Foglia³, Ettore Paltanin^3,11, Stefano Bonetti^4,12, Dieter Engel⁹, Daniel Schick⁹, Christian Gutt¹⁰, Riccardo Comin¹³, Keith A. Nelson¹, and Alexei A. Maznev¹

¹ Department of Chemistry, Massachusetts Institute of Technology, Cambridge, MA, United States.
² Department of Physics, Emory University, Atlanta, Georgia, United States.
³ Elettra - Sincrotrone Trieste, Basovizza, Italy.
⁴ Department of Molecular Sciences and Nanosystems, Ca’Foscari University of Venice, Venice, Italy.
⁵ European XFEL, Schenefeld, Germany.
⁶ Dipartimento di Fisica, Università degli Studi di Milano, Milano, Italy.
⁷ CNR-Istituto Officina dei Materiali, Trieste, Italy.
⁸ Helmholtz-Zentrum Berlin Für Materialien und Energie GmbH, Berlin, Germany.
⁹ Max Born Institute for Nonlinear Optics & Short Pulse Spectroscopy, Berlin, Germany.
¹⁰ Department Physik, Universität Siegen, Siegen, Germany.
¹¹ Department of Physics, Università degli Studi di Trieste, Trieste, Italy.
¹² Department of Physics, Stockholm University, Stockholm, Sweden.
¹³ Department of Physics, Massachusetts Institute of Technology, Cambridge, MA, United States.

Contact persons email:
Local contact person email: