Spintronics at BESSY II: Real-time analysis of magnetic bilayer systems
The photo shows the laser bench with part of the laser beam path used to excite the sample. This is where the intensity, polarisation and size of the laser spot on the sample are adjusted before the laser beam is coupled into the vacuum chamber. © C. Schüßler-Langeheine / HZB
A view of the femtoslicing facility at BESSY II. This is where the measurements were carried out. © C. Schüßler-Langeheine / HZB
Schematic diagram of the sample (from left to right): A nine-atom-thick ferromagnetic iron (Fe) layer and an antiferromagnetic cobalt oxide (CoO) layer, also nine atoms thick, are situated on a silver single crystal. The sample is then excited by an ultrashort infrared laser pulse. In response to this, soft X-ray pulses are detected and analysed. This allows to determine the decay of the antiferromagnetic order. © PRL 2026
Spintronic devices enable data processing with significantly lower energy consumption. They are based on the interaction between ferromagnetic and antiferromagnetic layers. Now, a team from Freie Universität Berlin, HZB and Uppsala University has succeeded in tracking, for each layer separately, how the magnetic order changes after a short laser pulse has excited the system. They were also able to identify the main cause of the loss of antiferromagnetic order in the oxide layer: the excitation is transported from the hot electrons in the ferromagnetic metal to the spins in the antiferromagnet.
While conventional microelectronics involves the movement of electric charges, spintronics is based on electron spins. Manipulating spins requires less energy than transporting charged particles. Consequently, spintronic components offer the potential for significant energy savings and high processing speeds.
However, future applications will require clock speeds in the terahertz range, which are not yet achievable today. The clock speeds of current spin-based applications are up to a hundred times lower. In order to advance spintronics, a large team at the Transregio Collaborative Research Centre CRC/TRR 227 is investigating spin dynamics in solids at atomic resolution and on ultra-fast timescales.
Magnetic order is the key
The defining feature of spintronic devices is their combination of very thin ferromagnetic and antiferromagnetic films. While ferromagnetic thin films generate a magnetic field, antiferromagnetic thin films exhibit order but no magnetisation. Antiferromagnetic thin films have faster dynamics, generate no magnetic stray fields and allow for a wider choice of materials. However, they are more difficult to study due to the absence of a macroscopic magnetic moment. Although they are frequently used to influence the properties of adjacent ferromagnetic layers in devices, until now very little was known about how such bilayers respond to an ultrashort laser pulse on ultrashort timescales.
The team has now experimentally observed how the magnetic order in an antiferromagnetic-ferromagnetic bilayer system is lost at ultrahigh speeds. The sample consisted of an extremely thin antiferromagnetic layer of nine layers of cobalt oxide (CoO), which was deposited onto a silver crystal and covered with a nine-layer film of ferromagnetic iron (Fe).
The method: femtoslicing at BESSY II
At the BESSY II X-ray source, stroboscopic pump-probe experiments can be carried out using ultra-short soft X-ray and laser pulses. In this process, the sample is first excited and its response is then measured. This method, known as 'femtoslicing', enables true 'snapshots' of the magnetic state to be taken very shortly after excitation, within the femtosecond range. By using X-ray magnetic dichroism, in which the reflected intensity of circularly and linearly polarised soft X-rays is measured, the response of both layers can be precisely separated.
Both orders were found to collapse practically simultaneously within approximately 300 femtoseconds (1 femtosecond = 10−15 seconds) after the sample was struck by a laser pulse with a wavelength of 800 nanometres. 'This is surprising, as CoO is transparent at this wavelength and therefore does not directly absorb the laser pulse. Excitation transfer from the Fe layer to the CoO layer is thus the dominant mechanism for the ultrafast loss of antiferromagnetic order in CoO,’ says Wolfgang Kuch, who led the study.
Theoretical calculations indicate that only direct energy transfer from the excited electrons in iron to the spin system of cobalt oxide via the interface between the two layers can account for the experimental results on the ultrafast timescale. This is important for developing antiferromagnetic-ferromagnetic layered systems for use in the fastest spintronic applications. The model fits the experimental data very well.
‘This work demonstrates how we are continuously developing our experimental methods at BESSY II to gain ever greater clarity,’ says Christian Schüßler-Langeheine, X-ray spectroscopy expert at BESSY II. The measurement method used here is based on work carried out at BESSY II 23 years ago. At that time, it was demonstrated how certain types of antiferromagnetic order, such as that found in CoO, could be investigated with great sensitivity using X-ray reflection measurements (DOI: https://doi.org/10.1103/PhysRevB.67.052401).
‘Building on this, we are able to investigate ultrafast spin transport in a whole class of antiferromagnets. In recent years, the Slicing Facility has produced many groundbreaking studies on ultrafast antiferromagnetic dynamics. This new study has significantly deepened our understanding of ultrafast spin dynamics in bilayers,’ says Schüßler-Langeheine.
Transregio 227 ‘Ultrafast Spin Dynamics’
Transregio 227 ‘Ultrafast Spin Dynamics’ investigates electron spin on ultrashort timescales, i.e. within a few femtoseconds. This Transregional Collaborative Research Centre (TRR) was established in 2018 by the German Research Foundation (DFG) and has so far been funded with around 20 million euros. The DFG has now announced that it will provide a further ten million euros in funding for TRR 227 from 2026 to 2029.
Partners: Freie Universität Berlin, Martin Luther University Halle-Wittenberg, Technical University of Berlin, University of Potsdam, Helmholtz-Zentrum Berlin, Max Born Institute for Nonlinear Optics and Short-Time Spectroscopy, Berlin, Max Planck Institute for Microstructure Physics, Halle.