Single femtosecond laser pulse excitation of individual cobalt nanoparticles

Laser-induced manipulation of magnetism at the nanoscale is a rapidly growing research topic with potential for applications in spintronics. In this work, we address the role of the scattering cross section, thermal effects, and laser fluence on the magnetic, structural, and chemical stability of individual magnetic nanoparticles excited by single femtosecond laser pulses. We find that the energy transfer from the fs laser pulse to the nanoparticles is limited by the Rayleigh scattering cross section, which in combination with the light absorption of the supporting substrate and protective layers determines the increase in the nanoparticle temperature. We investigate individual Co nanoparticles (8 to 20 nm in size) as a prototypical model system, using x-ray photoemission electron microscopy and scanning electron microscopy upon excitation with single femtosecond laser pulses of varying intensity and polarization. In agreement with calculations, we find no deterministic or stochastic reversal of the magnetization in the nanoparticles up to intensities where ultrafast demagnetization or all-optical switching is typically reported in thin films. Instead, at higher fluences, the laser pulse excitation leads to photochemical reactions of the nanoparticles with the protective layer, which results in an irreversible change in the magnetic properties. Based on our findings, we discuss the conditions required for achieving laser-induced switching in isolated nanomagnets.

Figure 1

(a) Calculated time-dependent electron temperature $T_e$, lattice temperature $T_l$, and (b) relative demagnetization Δm/m in a thermally isolated cobalt nanoparticle upon excitation with a single femtosecond laser pulse with $\lambda =800\mathrm{nm}$ and a peak fluence ${\mathrm{\Phi }}_0=21\mathrm{mJ}\mathrm{c}{\mathrm{m}}^{-2}$ for $D\ll \lambda$. (c) Maximum electron and lattice temperature as a function of laser fluence.

Figure 2

(a) Calculated instantaneous temperature profile upon absorption of a single pulse with ${\mathrm{\Phi }}_0=9\mathrm{mJ}\mathrm{c}{\mathrm{m}}^{-2}$. (b) Simulated sample geometry and temperature distribution in the sample, 1 ps after the laser pulse. (c) Simulated time-dependent temperature evolution in the nanoparticle (red) and in the carbon film (black) due to heat diffusion. (d) Calculated temperature profile in a carbon-capped, 12-nm-thick Co film on a Si wafer with a native $\mathrm{Si}{\mathrm{O}}_x$ layer, using the same laser parameters as in (a). The inset shows the corresponding relative demagnetization calculated using the phenomenological 3TM.

Figure 3

(a) Schematic diagram of the sample and the experimental setup showing the direction of the incoming x ray and laser light and the setup for generation of single fs laser pulses. PP: pulse picker, FS: fast shutter, λ/2, λ/4: wave plates, P: polarizer, M: leaking mirror, L: lens, and PD: fast photodiode. (b) Laser-beam profile (dark area) determined using a Cs-covered reference sample in the XPEEM instrument.

Figure 4

(a) XPEEM image recorded with the photon energy set to the $\mathrm{Co}{L}_3$ edge. Bright spots correspond to cobalt nanoparticles. Marker structures appear as saturated bright features. (b) Magnetic contrast map of the same sample area. Solid circles highlight magnetically blocked nanoparticles. Dashed circles highlight nanoparticles without magnetic contrast. The inset shows a HAADF STEM image of a cobalt nanoparticle capped with an amorphous carbon layer. (c) XA spectra of the nanoparticles recorded at the $\mathrm{Co}{L}_3$ edge prior to the laser experiment. The gray tone intensities in (a) and (b) are set to visualize the nanoparticle signals.

Figure 5

Row A: XPEEM images of region ``ii'' in Fig. 4. Row B: Control sequence of 10 consecutively recorded magnetic contrast maps without laser exposure. Row C: Similar sequence, but each magnetic contrast map is recorded upon exposure to a single laser pulse with a peak fluence of ${\mathrm{\Phi }}_0=4\mathrm{mJ}\mathrm{c}{\mathrm{m}}^{-2}$ and linear polarization. The scale bar in panel (a) in row A corresponds to 1 μm. The solid circle denotes a magnetically blocked nanoparticle, while the two dashed circles highlight nanoparticles which exhibit no or varying contrast over the time of the experiment. The gray tones range from 0.97 (black) to 1.06 (white) in row A and from −0.07 (black) to +0.07 (white) in rows B and C.

Figure 6

Row A: XPEEM images of region ``i'' in Fig. 4. Row B: Control sequence of 10 consecutively recorded magnetic contrast maps without laser exposure. Row C: Similar sequence, but each magnetic contrast map is recorded upon exposure to a single laser pulse with ${\mathrm{\Phi }}_0=9\mathrm{mJ}\mathrm{c}{\mathrm{m}}^{-2}$ and linear polarization. The scale bar in panel a in row A corresponds to 1 μm. The solid circle denotes a magnetically blocked nanoparticle, while the two dashed circles highlight nanoparticles, which exhibit no, or varying contrast over the time of the experiment. The gray tones range from 0.97 (black) to 1.06 (white) in row A and from −0.07 (black) to +0.07 (white) in rows B and C.

Figure 7

(a) XA spectra of the nanoparticles before (dashed line) and after laser excitation with ${\mathrm{\Phi }}_0=9\mathrm{mJ}\mathrm{c}{\mathrm{m}}^{-2}$ (solid line). (b) XA reference spectra for $\mathrm{C}{\mathrm{o}}_3{\mathrm{O}}_4$ and fcc CoO from Refs. [53, 54]. (c) XA reference spectra of cobalt nanoparticles with a size of about 3 nm being embedded in a carbon matrix from Ref. [55]. The dashed lines denote the peak positions in the spectra recorded after the laser excitation experiments. (d) SEM image of the laser-exposed area of the sample recorded after the laser experiments. The dashed line denotes the position of the laser spot on the sample. (e), (f) SEM images of two laser-exposed cobalt nanoparticles. (g), (h) SEM images of two cobalt nanoparticles from the same sample, but from a region that was not exposed to the laser. The scale bar corresponds to 10 nm.