betaFeedback

WorksPhysics and Astronomy

Light-driven modulation of proximity-enhanced functionalities in hybrid nano-scale systems

Abstract

Advancing quantum information and communication technology (qICT) requires smaller and faster components with actively controllable functionalities. This work presents a novel strategy for dynamically modulating magnetic properties via proximity effects controlled by light. We demonstrate this concept using hybrid nanoscale systems composed of C60 molecules proximitized to a cobalt metallic ferromagnetic surface, where proximity interactions are particularly strong. Our findings show that by inducing excitons in the C60 molecules with resonant ultrashort light pulses, we can significantly modify the interaction at the cobalt/C60 interface, leading to a striking 60% transient shift in the frequency of the dipolar ferromagnetic resonance mode of the Cobalt. This effect, detected via a specifically designed time-resolved magneto-optical Kerr effect (tr-MOKE) experiment, persists on a timescale of hundreds of picoseconds. Since this frequency shift directly correlates with a transient change in the anisotropy field (an essential parameter for technological applications) our findings establish a new paradigm for ultrafast optical control of magnetism at the nanoscale.

Introduction

The quest to manipulate magnetism without external magnetic fields is driving innovations in information and memory storage devices. A particularly promising approach involves using femtosecond laser pulses to either quench[1, 2] or switch[3] magnetization. This interaction between ultrashort laser pulses and magnetically ordered materials has spurred the dynamic field of femto-magnetism[4]. Achieving optical control of magnetism with sub-wavelength spatial resolution could introduce new functionalities, such as optical spin-switching for information recording at femtosecond speeds[5, 6]. While ultrafast optical control has been demonstrated in dielectric materials[7], coherent control on the femtosecond time scale remains challenging in metallic systems due to rapid electron-hole pair dephasing and coherence loss caused by Coulomb screening[8].

To address this challenge, we propose exploiting the functionality inherent in nano-scale hybrid units formed by molecules in proximity to metallic ferromagnetic surfaces. Inspired by advancements in molecular spintronics, which reveal molecules as a revolutionary platform for exploring spin-dependent phenomena at the nano-scale[9, 10], we focus on harnessing proximity effects to modulate magnetic properties. Proximity effects at molecular-metal interfaces have been shown to induce significant modifications in magnetic properties, such as enhanced anisotropy[11, 12, 13, 14], increased coercivity[15, 16], and improved dynamic responses to external radiofrequency stimuli[17]. Despite these established benefits of proximity effects, their full potential remains untapped due to a focus on creating rather than controlling new functionalities. Recent suggestions[18] point towards developing devices where molecular components translate optical excitations into modified proximity effects, thereby generating new functionalities within the hybrid units.

This study demonstrates the optical functionality of Co films in proximity to C60C_{60} molecules, realizing a novel approach to modulating GHz spin dynamics. By inducing excitons in C60C_{60} with resonant ultrashort light pulses, we significantly alter hybridization at the Co/C60C_{60} interface, achieving up to a 60% modification in the frequency of the dipolar ferromagnetic resonance (FMR) mode. This shift directly correlates with changes in the anisotropy field, introducing a groundbreaking method for optical control of a key parameter in nanoscale hybrid systems.

Anisotropy plays a crucial role in data storage[19], field-free magnonics[20], and neuromorphic computing[21]. Our results underscore the potential for leveraging molecular optical functionality in next-generation qICT technologies, paving the way for new paradigms in spin-based information processing.

Results

Proximity-induced enhancement of the anisotropy field in Co/C60C_{60} observed by tr-MOKE

In our investigation, we delve into the optically-induced spin dynamics of Co/C60C_{60} and reference Co/Al bilayers, examining their dynamical behavior across varying temperatures and applied magnetic fields. The bilayers consist of thin cobalt films with 5 nm nominal thickness deposited on an Al2_2O3_3 (1000) substrate and covered respectively with 25 nm C60C_{60} and 3 nm Al. The latter was chosen as capping layer for its effectiveness in preventing oxidation of the Co layer[22]. The experimental setup, as depicted in Figure 1a, employs a variable out-of-plane magnetic field (H\mathbf{H}) to align the magnetization (M\mathbf{M}) along an effective field (Heff\mathbf{H}_{\mathrm{eff}}), which constitutes a vector sum of the external field (H\mathbf{H}) and an internal field (Hint\mathbf{H}_{\mathrm{int}}). This internal field is itself the sum of the shape anisotropy field (Hsh\mathbf{H}_{\mathrm{sh}}) and the intrinsic sample-dependent anisotropy field (Hanis\mathbf{H}_{\mathrm{anis}}). In the tr-MOKE experiments, an ultrafast laser pump pulse transiently perturbs the sample, leading to

Schematic experimental setup, hysteresis loops, transient Kerr rotation signals, and precession frequency

Figure 1. a, schematic representation of the experimental setup utilized for both static and time-resolved magneto-optical Kerr effect (TR-MOKE) measurements. We utilize the polar MOKE geometry, in which the external magnetic field is applied perpendicular to the sample surface (out-of-plane direction). b, Hysteresis loops recorded at 10 K for Co/Al and Co/C60C_{60}, demonstrating that the magnetic hard axis matches the normal axis of the film. c, tr-MOKE signals measured for HextH_{\mathrm{ext}}=0.5 T as a function of the temperature for the reference Co/Al sample and the Co/C60C_{60} sample, respectively. d, values of the precession frequency (ν\nu) extracted from the data in c) as described in the main text.

ultrafast demagnetization within hundreds of femtoseconds and causing Heff\mathbf{H}_{\mathrm{eff}} to deviate from its equilibrium position[23]. Following this initial perturbation, the magnetization experiences damped precession around Heff\mathbf{H}_{\mathrm{eff}}, concurrently undergoing gradual remagnetization (for more details see SI). The observed magnetization precession corresponds to the dipolar ferromagnetic resonance (FMR) mode, characterized by collective oscillations of ferromagnetic (FM) spins with wave vector k∼0\mathbf{k}\sim0. In this very well-established experimental framework[23], the frequency of the FMR mode (ν\nu) emerges as a crucial parameter, since it is directly proportional to ∣Heff∣\lvert H_{\mathrm{eff}}\rvert and, consequently, to ∣Hanis∣\lvert H_{\mathrm{anis}}\rvert, though the relationship is intricately modulated by the applied external field direction.

Figure 1b shows the static MOKE measurements performed at 10 K for the the Co/C60_{60} and Co/Al bilayers. The measurements indicate that for both samples the film normal axis behaves as a hard axis. For the Co/Al bilayer, the shape anisotropy is indeed expected to dominate over other potential anisotropies that might favor an out-of-plane easy axis[24]. For the Co/C60_{60} bilayer our observation agrees with previous findings[14, 25], that have additionally shown that although the hybridization between Co and the C60_{60} molecules strongly modifies the magnetic anisotropy, the out-of-plane direction remains a hard axis.

Figure 1c compares the tr-MOKE signal measured from the Co/C60_{60} sample at various temperatures (between 50 K and 290 K) to the Co/Al signal at 50 K. The external magnetic field was Hext=0.5H_{\mathrm{ext}}=0.5 T. The measured curves have two components: a non-oscillatory part corresponding to the initial suppression of the out-of-plane magnetization component followed by its thermal recovery, and the oscillatory component related to the magnetization precession around HeffH_{\mathrm{eff}}. First of all, we observe that the oscillations observed for the Co/Al sample are temperature independent. Second, the Co/C60_{60} interface exhibits higher precession frequencies compared to the Co/Al reference, albeit with a rapid damping after just a few oscillation periods.

For a quantitative analysis, we extracted the oscillation frequency (ν\nu) by subtracting the non-oscillating thermal recovery background and then performing a fit with a damped sine function: Asin⁡[2πνt+ϕ0]exp⁡(−t/τD)A\sin[2\pi\nu t+\phi_0]\exp(-t/\tau_D), where τD\tau_D represents the damping time. The frequencies extracted using this fit procedure are depicted in Figure 1d as a function of the temperature. The difference between the Co/C60_{60} and Co/Al precession frequency becomes larger for decreasing temperature. At low temperatures, the data reveal an approximately fivefold increase of ν\nu, aligning with the literature's observations of an enhanced HanisH_{\mathrm{anis}} value[17, 25]. This enhancement is attributed to the hybridization of electronic states at the Co/C60_{60} interface, which significantly alters the magnetic anisotropy[11, 13]. This set of measurements demonstrates that we can use the frequency of the FMR mode extracted from the tr-MOKE experiments as a measure of the anisotropy field, a technologically relevant parameter that is strongly modified by the adsorption of C60_{60} on Co.

Before demonstrating the optical modulation of the FMR frequency and anisotropy field, it is important to note that a microscopic description of the hybridized Co/C60_{60} interface extends beyond a simple quadratic anisotropy model. Recent work[26] shows that the FMR frequency's dependence on the out-of-plane magnetic field deviates from standard FMR formulas and seems inconsistent with static out-of-plane magnetization measurements, suggesting a more complex anisotropy landscape. Upon hybridization with molecules, the interfacial Co layer develops a strongly anisotropic in-plane magnetic state. This state predominantly influences the FMR response, which can be effectively described using an in-plane exchange-bias-like free energy term combined with a reduction of the out-of-plane hard-axis anisotropy. A detailed theoretical treatment is discussed elswhere[26]. As this manuscript primarly focuses on the laser-induced modulation of the effective anisotropy, we describe our results in terms of the effective anisotropy field extracted directly from our measurements. This approach aligns with well-established literature[27, 28] and avoids additional assumptions.

Pump wavelength dependence of the oscillation frequency

Our main goal is to actively manipulate the interface-driven modifications in magnetic anisotropy by optically exciting the Co/C60_{60} hybrid units, utilizing the same pump pulse responsible for initiating the ultrafast magnetization dynamics within the Co layer. Optical modification of the magnetic anisotropy will then lead to a modification of the FMR precession frequency that we detect in our tr-MOKE experiments. To accomplish this, we conducted an experimental sequence where we captured several tr-MOKE signals at the same out-of-plane magnetic field utilized in Figure 1c (μ0H=0.5\mu_0 H=0.5 T), varying the pump fluence and the photon energy of the pump laser. From these signals, we extracted the oscillation frequency and the precession decay time for each set of conditions, facilitating a comparative analysis of the results. To ascertain that the variations in the magnetization dynamics were attributable to the excitation of the molecular overlayer and not solely to the Co layer, we replicated the experiments on the reference Co/Al sample.

To select the most effective excitation wavelengths, we first conducted temperature-dependent absorbance measurements on the Co/C60_{60} system in the wavelength range 350 - 800 nm (3.5 eV - 1.5 eV). Additionally, we analyzed the Co/Al system to identify any absorption peaks characteristic of Co. We anticipated that most of the Al would be oxidized after exposure to ambient conditions[22], thus contributing minimally to the measurements. The absorbance spectra measured at 80 K are shown in Figure 2a, with the full dataset available in the supplementary material (SI). Compared to Co/Al, the absorbance spectrum of Co/C60_{60} shows a distinct peak emerging below 400 nm. This feature corresponds to a peak at 355 nm, resulting from the first allowed optical transition of isolated C60_{60} molecules[29, 30, 31, 32, 33, 34]. The intramolecular S0_0→S1_1 absorption, being symmetry forbidden, appears as a very weak feature in the spectrum at 650nm. In the spectral range between 390 nm and 550 nm, the spectrum is dominated by a broad feature attributed to intermolecular excitations, which include both charge transfer and localized Frenkel excitations. For longer wavelengths, only localized excitations contribute to the spectrum[32]. The inset of Figure 2a shows the fit of the broad feature at 390 nm-550 nm with three Gaussian functions. The SI includes the temperature dependence of the extracted peak position and FWHM. We observe a non-monotonic behavior, with a change in slope occurring between 80K and 120K. This behavior is explained by a structural transition in the

Absorbance spectra and oscillation frequency plotted against absorbed energy density

Figure 2. a, absorbance spectra of the Co/C60 and reference Co/Al samples at 80 K. Vertical lines mark the selected pump photon energies. Inset: Fit of the spectral feature at 390nm-550nm with three Gaussian functions. b, oscillation frequency plotted against the absorbed energy density in the Co layer for all selected pump wavelengths, with linear fits depicted by the lines.

C60_{60} layer around this temperature[35], which in turn influences the properties of the charge transfer excitons in this spectral range.

Given these findings, we have chosen two wavelengths in the spectral range of the broad absorption feature related to intermolecular excitations for resonantly exciting the Co/C60_{60} system and tuned the pump wavelength to these values: 450 nm (2.75 eV) and 400 nm (3.1 eV). In addition, we have chosen 650 nm (1.9 eV) pump to assess the response of the system in the presence of a very weak intramolecular excitation. For the probe beam, we selected the wavelength of 800 nm (1.5 eV), which is in the transparency window of C60_{60}, ensuring that the observed response is specifically sensitive to the Co layer dynamics. Moreover, systematic studies of Co/molecules heterostructures[26], indicate that the Co/molecules interfacial layer primarily modulates the effective anisotropy field of the Co layer rather than exhibiting independent spin dynamics.

We conducted tr-MOKE experiments on both Co/C60_{60} and Co/Al samples using the three selected pump photon energies, analyzing the results as a function of the absorbed energy density in the Co layer (wCow_{Co}). This parameter, which accounts for wavelength-dependent absorption differences, is determined in the SI. Additionally, we used the maximal demagnetization peak as a verification parameter to ensure the effective laser fluence was correctly chosen. The recorded data are presented in the SI. In Figure 2b, we report the values of ν\nu extracted from the data in the low pump-fluence region (up to 0.15 mJ/cm2^2). In this fluence range, the Co/C60_{60} the precession frequency shows a linear dependence on wCow_{Co}, whereas in the reference Co/Al sample, the frequency remains constant, and agrees well with the pristine Co resonance frequencies. (Exemplarily data for the Co/C60_{60} system at higher fluences are reported in the SI.)

By performing linear fits, we extracted the extrapolated values of ν\nu at wCo=0w_{Co} = 0 (ν0\nu_0) and the slopes s=dνdwCos = \frac{d\nu}{dw_{Co}}. For ν0\nu_0 we obtained virtually identical values (within the error bars) for the three pump wavelengths: (26.0 ± 1.0) GHz, (26.0 ± 0.9) GHz, and (25.8 ± 0.8) GHz for 650 nm, 450 nm, and 400 nm, respectively. We therefore interpret ν0\nu_0 as the FMR frequency of the pristine Co/C60_{60} system, i.e. for the case where the magnetization is tilted adiabatically out of its equilibrium position, without altering the electronic properties of the system itself. Moving to the extracted slopes (ss), we observe a significant dependence on the pump wavelength: (−42±15) GHz/(mJ/cm2)(-42 \pm15)\ \mathrm{GHz}/(\mathrm{mJ}/\mathrm{cm}^{2}), (−216±33) GHz/(mJ/cm2)(-216 \pm33)\ \mathrm{GHz}/(\mathrm{mJ}/\mathrm{cm}^{2}), and (−93±18) GHz/(mJ/cm2)(-93 \pm18)\ \mathrm{GHz}/(\mathrm{mJ}/\mathrm{cm}^{2}) for 650 nm, 450 nm, and 400 nm respectively, with a much larger slope observed for pump photon energy in the intermolecular excitations peak. Importantly, in the Co/Al sample, no difference emerges between the two tested pump wavelengths, indicating that the pronounced wavelength dependence of ss in the Co/C60_{60} sample is attributable to the excited state of the molecular layer.

Exciton influence on the spin dynamics

We now move to the discussion of the experimental results. As already mentioned, we interpret ν0\nu_0 value as the FMR frequency of the electronically unperturbed Co/C60_{60} layer. Regarding the observed dependence of ν\nu from wCow_{\mathrm{Co}}, we observe that when the pump photon energy is in the region of the intermolecular absorption (resonant pump) the decrease of ν\nu is much steeper than for the pump at 650nm (off-resonant pump). In general, increasing wCow_{\mathrm{Co}} leads to a progressing decrease of the precession frequency of Co/C60_{60} toward the precession frequency of the pristine Co sample. If the pump is in resonance, much lower values of ν\nu can be achieved than for the off-resonant case. Since the precession frequency is proportional to the anisotropy field, the observed behavior indicates that resonant pumping effectively reduces the anisotropy field of the Co/C60_{60} system towards the value of the pristine Co sample. The higher the pump fluence, the stronger the quenching effect. Therefore, we conclude that the effect of resonant pump photons is to quench the hybridization at the Co/C60_{60} interface, causing it to behave magnetically similar to the pristine Co layer. This quenching effect underscores the impact of resonant excitations on the magnetic properties of the hybrid

Femtosecond-resolved transient absorption spectra, line profiles, and exciton scheme of C60

Figure 3. a, Femtosecond-resolved transient absorption spectra of Co/C60_{60} following on-resonance excitation with 400 nm pump pulses. b, Line profiles extracted from the data in a. c, Exciton scheme of C60_{60}.

system, providing insights into the tunability of magnetic anisotropy through optical excitation.

To quantify the effect of this novel exciton-mediated optical control of GHz spin dynamics, we quantify the modification of the precession frequency at ωCo=0.07\omega_{\mathrm{Co}} = 0.07 mJ/cm2^{2}. At this energy density, we note an extremely significant modulation of more than 60% between the resonant and off-resonant pumping. This highlights the profound impact of resonant optical excitation on the spin dynamics and magnetic anisotropy in molecule-interfaced Co thin films.

Before concluding, we turn to the mechanism leading to the optically induced quenching of hybridization at the Co/C60_{60} interface and the involved timescales. To shed light on this point, we performed fs-resolved transient reflectivity experiments following excitation with (351±\pm10) nm pump pulses, recorded between 0-12 ps in the wavelength range of 450-700 nm. The measurements, together with the line profiles extracted from those measurements at different time delays of 0.5ps, 2ps, 5ps, are shown in Fig. 3a and Fig. 3b, respectively. In these spectra, we observe a negative peak around 500 nm, followed by a positive peak at 540 nm. This feature has been attributed in the literature to the presence of local electric fields generated by the direct population of intermolecular charge transfer states and their associated strong electric dipoles[32].

Excitation at 351 nm leads to the formation of charge transfer excitons[33, 34]. In the Supplementary Information we extract the timescales related to exciton dynamics, that we summarize in Fig. 3c. The charge-transfer excitons decay on a timescale τ1=0.18\tau_{1} = 0.18 ps to lower-lying charge transfer (CT) states or localized Frenkel excitons. The formed Frenkel excitons, as well as those directly excited by the pump laser, decay with a time constant τ2=4.5\tau_{2} = 4.5 ps. Finally, the relaxed Frenkel excitons decay with a much longer time constant, in agreement with the 150ps reported in literature[32]. This latter relaxation time scale is well within the damping time of the oscillations observed in the Co/C60_{60} system (see SI for details on the extraction of the time constants from the experimental data as well as on the damping constant of the spin dynamics in Co/C60_{60}). It is also observed when pumping off-resonance with 650 nm[32]. However, in this case, Frenkel excitons can only form via direct laser excitation, and since intramolecular S0→S1S_{0}\rightarrow S_{1} absorption is symmetry forbidden, significantly fewer Frenkel excitons are formed during off-resonant excitation.

The observed exciton dynamics lead us to the following explanation for the experimental data in fig:2b: using resonant pump excitation, we create charge transfer excitons in C60_{60}. When the pump pulse is absorbed by the C60_{60} molecules, charge transfer (CT) excitons are formed within the first few hundred femtoseconds, rapidly decaying (in a few picoseconds) into Frenkel-like excitons (S1) that remain stable for several tens of picoseconds, the same time-scale on which we observe the coherent spin dynamics in Co/C60_{60}. These excitons consist of a hole in the Highest Occupied Molecular Orbital (HOMO) of the molecule and an electron in the Lowest Unoccupied Molecular Orbitals (LUMO), specifically the LUMO+1 for CT excitons or LUMO for S1 excitons[36].

Although the complex magnetic structure of the Co/C60_{60} interface has only recently been elucidated[37], on a local scale, the chemisorption of C60_{60} molecules on the Co surface can be understood through the formation of hybrid molecule-metal dz2_{z^2} bonds[13, 14], which strongly affect the surface properties[12, 14]. Crucially, only the few C atoms in direct contact with the surface contribute to such hybridization, as indicated by the bonds depicted in Figure 4a. In contrast, the other C atoms interact weakly with the surface owing to the “soccer ball” molecular structure. This is seen in the density of states in Figure 4b (see SI for more information).

C$_{60}$ adsorbed on the Co surface and Co- and C$_{60}$ HOMO-PDOS calculated by DFT

Figure 4. a, C60_{60} adsorbed on the Co surface. b, Co- and C60_{60} HOMO-PDOS, calculated by DFT. Positive (negative values) are for spin-up (spin-down) states. The Co surface DOS is obtained by summing the data for all eighteen surface cobalt atoms in the supercell. Both the HOMO- and CO-PDOS are normalized so that their integral over the energy is equal to one.

Consequently, C60_{60} exhibits dual properties: it retains its molecular character, allowing it to display excitonic features, while, through the few hybridized C atoms, it can significantly alter the surface's magnetic properties. When the Co/C60_{60} sample is pumped with resonant photons, the long-living Frenkel-like excitons formed in C60_{60} effectively result in an excited molecular state which in turn modifies the molecular properties and, therefore, of the molecule-surface interaction. Specifically, our observations indicate that this modification partially quenches the interfacial hybridization strength of the entire Co/C60_{60} system. The higher the absorbed fluence, the stronger the quenching effect becomes, causing the Co/C60_{60} hybrid units to behave more like pristine cobalt. Consequently, as the quenching effect intensifies, the anisotropy field of the Co/C60_{60} system approaches that of pristine Co films (which is lower). This explains the observed reduction in oscillation frequency with increasing absorbed fluence.

Given that the exciton lifetime is approximately 150 ps, comparable to the timescale during which k∼0k \sim0 spin waves can be detected, the exciton density undergoes an exponential decay, which suggests a potential time dependence of the oscillation frequency. To investigate this effect, we conducted an extended analysis of our experimental data, revealing that the oscillation frequency is indeed time-dependent. The frequency values reported in the main manuscript correspond to the early-time oscillations, where the exciton density is still high and before significant decay occurs. For times approaching the exciton lifetime, deviations from a purely sinusoidal behavior emerge, indicating a gradual frequency shift (see Figure S5 in the SI). This time-dependent frequency evolution does not contradict our primary claim—that the FMR frequency can be optically tuned by selectively exciting the C60_{60} layer. Rather, it reinforces the transient nature of the exciton-driven modulation, further supporting the role of molecular excitations in controlling interfacial magnetic properties. Our results demonstrates that not only the modulation is strong but also localized at the nanoscale, as it originates from complex quantum behaviors. Moreover, we propose that the underlying physics is universal, potentially allowing for the optical tuning of any proximity-induced physical property, beyond just magnetic ones.

Methods

Sample fabrication

Thin Co (5 nm thickness) films were deposited by electron beam evaporation on Al2_2O3_3(0001) substrates at the room temperature and base pressure of 1.1×10−101.1 \times10^{-10} mbar. The organic layer or the Al layer were subsequently deposited on top of the Co layer by thermal evaporation (base pressure 1.1×10−91.1 \times10^{-9} mbar) at room temperature without breaking the vacuum.

Setup for absorbance measurements

The static absorbance measurements were performed using a commercial spectrophotometer (CARY 6000i, Agilent Technologies) in transmission. The sample was mounted in a He-flow cryostat (Oxford Instruments) for temperature dependent measurements with a precision of ±1\pm1 K. For the baseline correction, an identical sample holder was placed in the reference beam.

Setup for time-resolved magneto-optical Kerr effect measurements

The tr-MOKE measurements were done by means of two different table-top setups for optical pump–probe measurements. The fixed-fluence characterization was performed with a two-color pump-probe setup based on a high repetition-rate (250 kHz) 50-femtosecond Ti:sapphire laser amplifier and a split-coil superconducting 7 T optical magnet with variable temperature He exchange gas sample insert. A part of the output pulse train was frequency doubled (λ=400\lambda=400 nm, 3.1 eV photon energy) to derive the pump pulses while the probe pulses were derived from the remaining fundamental pulse train (λ=800\lambda=800 nm, 1.55 eV). The reflected-probe-beam transient polarization rotation was detected by means of a balanced detection using a Wollaston prism and a pir of silicon PIN photo diodes. The pump beam was modulated with an optical chopper at ∼2.5\sim2.5 kHz and a standard lock-in detection scheme was used to acquire the photodiodes differential signal.

The TR-MOKE measurements with variable pump wavelengths were done by means of another setup with the pump and probe light pulses independently tunable in the range 0.5 - 3.5 eV. The setup described in Refs. [38, 39]. The probe beam was kept fixed at λPR=800\lambda_{\mathrm{PR}}=800nm with 0.2 mW fluence, with a 30 μ\mum spot diameter and repetition rate of 100 kHz. The pump spot diameters were: 36μ\mum (λPU=650\lambda_{\mathrm{PU}}=650 nm), 50 μ\mum (λPU=450\lambda_{\mathrm{PU}}=450 nm) and 50 μ\mum for (λPU=400\lambda_{\mathrm{PU}}=400 nm); and the repetition rate was 50 kHz. The duration of both pump and probe pulses was ≈50\approx50 fs calculated as the FWHM (corresponding to a spectral energy spread, ΔE=0.04\Delta E=0.04 eV).

Setup for ultrafast transient reflection spectroscopy

A femtosecond laser beam was focused on a 1 mm sapphire crystal to generate a pulsed white-light probe with a repetition rate of 100 kHz. The probe beam was reflected from the sample and dispersed by a diffraction grating before being focused on a 1D CMOS array (Stresing GmbH). The pump beam, synchronized with the probe beam, was modulated with an optical chopper, which was triggered at one-256th of the laser repetition. In the pump-probe scheme, the probe beam was detected at a frequency of one-128th of the laser repetition, allowing the isolation of the signal at the photoexcited state. With this technique, we acquire the differential reflectivity ΔR/R\Delta R/R data as a function of probe wavelength and time delay.

Computational details

The calculations are performed by using an implementation of DFT based on the Green's function technique. Specifically, we employ the SMEAGOL code [40] that obtains the Kohn-Sham Hamiltonian from the SIESTA package[41]. We assume Co to have fcc structure. The C60C_{60} molecule is adsorbed on Co in the so-called “hexagon-pentagon” configuration, as described in Ref. 13. The surface region is contained in a 4×44 \times4 supercell, comprising the C60C_{60} molecule and six Co atomic layers, which are coupled to a semi-infinite Co bulk region through an embedding self-energy[42]. Periodic boundary conditions are assumed along the in-plane directions. Core electrons are treated with norm-conserving Troullier-Martins pseudopotentials[43]. The valence states are expanded through a numerical atomic orbital basis set including multiple-z and polarized functions. The local spin density approximation (LSDA) [44] is assumed for the exchange correlation functional. The electronic temperature is 300 K. The real space mesh is set by an equivalent energy cutoff of 300 Ry. We use a 6×66 \times6 k-point mesh in the two-dimensional surface Brillouin zone. The DOS projected on the HOMO shown in Figure 4b is obtained using the algorithm in Ref. 45.

Acknowledgments

We acknowledge the help of Cristian Manzoni for the improvement of the ultrafast transient reflectivity setup. This work was supported by the EC H2020 programme under grant agreement No. 965046, FET-Open project INTERFAST (Gated interfaces for fast information processing).

Data Availability statement

The raw data for datasets used in the article figures will be posted on the Zenodo database.

References

  1. [1]Koopmans, B., Malinowski, G., Longa, F. D., Steiauf, D., Fähnle, M., Roth, T., Cinchetti, M. & Aeschlimann, M. Explaining the paradoxical diversity of ultrafast laser-induced demagnetization. Nature Materials 9, 259–265 (2009).DOI
  2. [2]Beaurepaire, E., Merle, J. & Daunois, A. Ultrafast spin dynamics in ferromagnetic nickel. Physical Review Letters 76, 4250 (1996).
  3. [3]Lambert, C. H., Mangin, S., Varaprasad, B. S. D. C. S., Takahashi, Y. K., Hehn, M., Cinchetti, M., Malinowski, G., Hono, K., Fainman, Y., Aeschlimann, M. & Fullerton, E. E. All-optical control of ferromagnetic thin films and nanostructures. Science 345, 1337–1340 (2014).
  4. [4]Bigot, J.-Y., Vomir, M. & Beaurepaire, E. Coherent ultrafast magnetism induced by femtosecond laser pulses. Nature Physics 5, 515–520 (2009).DOI
  5. [5]Stanciu, C. D., Hansteen, F., Kimel, A. V., Kirilyuk, A., Tsukamoto, A., Itoh, A. & Rasing, Th. All-Optical Magnetic Recording with Circularly Polarized Light. Phys Rev Lett 99, 047601 (2007).pubmed.ncbi.nlm.nih.gov/17678404
  6. [6]Guyader, L. L., Savoini, M., Moussaoui, S. E., Buzzi, M., Tsukamoto, A., Itoh, A., Kirilyuk, A., Rasing, T., Kimel, A. V. & Nolting, F. Nanoscale sub-100 picosecond all-optical magnetization switching in GdFeCo microstructures. Nat Commun 6, 5839 (2015).
  7. [7]Kirilyuk, A., Kimel, A. & Rasing, T. Ultrafast optical manipulation of magnetic order. Rev.Mod.Phys. 82, 2731–2784 (2010).DOI
  8. [8]Huber, R., Tauser, F., Brodsch el m, A., Bichler, M., Abstreiter, G. & Leitenstorfer, A. How many-particle interactions develop after ultrafast excitation of an electron–hole plasma. Nature 414, 286–289 (2001).pubmed.ncbi.nlm.nih.gov/11713523
  9. [9]Bergenti, I. & Dediu, V. Spinterface: A new platform for spintronics. Nano Mater. Sci. 1, 149–155 (2019).DOI
  10. [10]Dediu, V. A. Organic spintronics: Inside the interface. Nature Physics 1–1 (2013). doi:10.1038/nphys2569DOI
  11. [11]Bairagi, K., Bellec, A., Repain, V., Chacon, C., Girard, Y., Garreau, Y., Lagoute, J., Rousset, S., Breitwieser, R., Hu, Y.-C., Chao, Y. C., Pai, W. W., Li, D., Smogunov, A. & Barreteau, C. Tuning the Magnetic Anisotropy at a Molecule-Metal Interface. Phys Rev Lett 114, 247203 (2015).pubmed.ncbi.nlm.nih.gov/26197003
  12. [12]Bairagi, K., Bellec, A., Repain, V., Fourmental, C., Chacon, C., Girard, Y., Lagoute, J., Rousset, S., Laurent, L. L., Smogunov, A. & Barreteau, C. Experimental and theoretical investigations of magnetic anisotropy and magnetic hardening at molecule/ferromagnet interfaces. Phys. Rev. B 98, 085432 (2018).DOI
  13. [13]Halder, A., Bhandary, S., O’Regan, D. D., Sanvito, S. & Droghetti, A. Theoretical perspective on the modification of the magnetocrystalline anisotropy at molecule-cobalt interfaces. Phys. Rev. Mater. 7, 064409 (2023).DOI
  14. [14]Moorsom, T., Wheeler, M., Khan, T. M., Ma’Mari, F. A., Kinane, C., Langridge, S., Ciudad, D., Bedoya-Pinto, A., Hueso, L., Teobaldi, G., Lazarov, V. K., Gilks, D., Burnell, G., Hickey, B. J. & Cespedes, O. Spin-polarized electron transfer in ferromagnet/C60interfaces. Physical Review B 90, 125311–6 (2014).DOI
  15. [15]Raman, K. V., Kamerbeek, A. M., Mukherjee, A., Atodiresei, N., Sen, T. K., Lazić, P., Caciuc, V., Michel, R., Stalke, D., Mandal, S. K., Blügel, S., Münzenberg, M. & Moodera, J. S. Interface-engineered templates for molecular spin memory devices. Nature 493, 509–513 (2013).DOI
  16. [16]Boukari, S., Jabbar, H., Schleicher, F., Gruber, M., Avedissian, G., Arabski, J., Costa, V. D., Schmerber, G., Rengasamy, P., Vileno, B., Weber, W., Bowen, M. & Beaurepaire, E. Disentangling Magnetic Hardening and Molecular Spin Chain Contributions to Exchange Bias in Ferromagnet/Molecule Bilayers. Nano Lett. 18, 4659–4663 (2018).DOI
  17. [17]Benini, M., Allodi, G., Surpi, A., Rimucci, A., Lin, K., Sanna, S., Dediu, V. A. & Bergenti, I. In-Depth NMR Investigation of the Magnetic Hardening in Co Thin Films Induced by the Interface with Molecular Layers. Adv. Mater. Interfaces 9, (2022).DOI
  18. [18]Cinchetti, M., Dediu, V. A. & Hueso, L. E. Activating the molecular spinterface. Nat Mater 16, 507 515 (2017).DOI
  19. [19]Shao, Y. & Amiri, P. K. Progress and Application Perspectives of Voltage-Controlled Magnetic Tunnel Junctions. Adv. Mater. Technol. 8, (2023).DOI
  20. [20]Taghinejad, H., Yamakawa, K., Huang, X., Lyu, Y., Cairns, L. P., Ramesh, R. & Analytis, J. G. Low-Field Regime of Magnon Transport in Yttrium Iron Garnet. arXiv (2024). doi:10.48550/arxiv.2411.14428DOI
  21. [21]Sethi, P., Sanz-Hernández, D., Godel, F., Krishnia, S., Ajejas, F., Mizrahi, A., Cros, V., Marković, D. & Grollier, J. Compensation of Anisotropy in Spin Hall Devices for Neuromorphic Applications. Phys. Rev. Appl. 19, 064018 (2023).arxiv.org/abs/2301.03794
  22. [22]Gan, L., Gomez, R. D., Powell, C. J., McMichael, R. D., Chen, P. J. & Egelhoff, W. F. Thin Al, Au, Cu, Ni, Fe, and Ta films as oxidation barriers for Co in air. J. Appl. Phys. 93, 8731–8733 (2003).DOI
  23. [23]Kampen, M. van, Jozsa, C., Kohlhepp, J. T., LeClair, P., Lagae, L., Jonge, W. J. M. de & Koopmans, B. All-Optical Probe of Coherent Spin Waves. Phys. Rev. Lett. 88, 227201 (2002).pubmed.ncbi.nlm.nih.gov/12059451
  24. [24]Chappert, C., Dang, K. L., Beauvillain, P., Hurdequint, H. & Renard, D. Ferromagnetic resonance studies of very thin cobalt films on a gold substrate. Phys. Rev. B 34, 3192–3197 (1986).
  25. [25]Moorsom, T., Alghamdi, S., Stansill, S., Poli, E., Teobaldi, G., Beg, M., Fanghor, H., Rogers, M., Aslam, Z., Ali, M., Hickey, B. J. & Cespedes, O. π-anisotropy: A nanocarbon route to hard magnetism. Phys. Rev. B 101, 060408 (2020).DOI
  26. [26]Strohsack, J., Shumilin, A., Zhao, H., Jecl, G., Kabanov, V. V., Benini, M., Rakshit, R., Dediu, V. A., Rogers, M., Ozdemir, S., Cespedes, O., Parlak, U., Cinchetti, M. & Mertelj, T. Anomalous hardening of spin waves in cobalt/molecular-semiconductor heterostructures reveals strongly anisotropic spinterface magnetism. arXiv (2024). doi:10.48550/arxiv.2412.08677DOI
  27. [27]Ferromagnetic Resonance: The Phenomenon of Resonant Absorption of a High-Frequency Magnetic Field in Ferromagnetic Substances. (Pergamon).
  28. [28]Bigot, J.-Y., Vomir, M., Andrade, L. H. F. & Beaurepaire, E. Ultrafast magnetization dynamics in ferromagnetic cobalt: The role of the anisotropy. Chem. Phys. 318, 137–146 (2005).DOI
  29. [29]Krätschmer, W., Lamb, L. D., Fostiropoulos, K. & Huffman, D. R. Solid C60: a new form of carbon. Nature 347, 354–358 (1990).DOI
  30. [30]Lu, H.-C., Lin, M.-Y., Peng, Y.-C., Chou, S.-L., Lo, J.-I. & Cheng, B.-M. Absorption, emission and photolysis of C60 with far-UV excitation. Mon. Not. R. Astron. Soc. 452, 2788–2793 (2015).DOI
  31. [31]Ruani, G., Fontanini, C., Murgia, M. & Taliani, C. Weak intrinsic charge transfer complexes: A new route for developing wide spectrum organic photovoltaic cells. J. Chem. Phys. 116, 1713–1719 (2002).DOI
  32. [32]Causa’, M., Ramirez, I., Hardigree, J. F. M., Riede, M. & Banerji, N. Femtosecond Dynamics of Photoexcited C60 Films. J. Phys. Chem. Lett. 9, 1885–1892 (2018).DOI
  33. [33]Stadtmüller, B., Emmerich, S., Jungkenn, D., Haag, N., Rollinger, M., Eich, S., Maniraj, M., Aeschlimann, M., Cinchetti, M. & Mathias, S. Strong modification of the transport level alignment in organic materials after optical excitation. Nat Commun 10, 1 9 (2019).
  34. [34]Emmerich, S., Hedwig, S., Arnoldi, B., Stöckl, J., Haag, F., Hemm, R., Cinchetti, M., Mathias, S., Stadtmüller, B. & Aeschlimann, M. Ultrafast Charge-Transfer Exciton Dynamics in C 60Thin Films. The journal of Physical Chemistry C 124, 23579–23587 (2020).DOI
  35. [35]Alers, G. B., Golding, B., Kortan, A. R., Haddon, R. C. & Theil, F. A. Existence of an Orientational Electric Dipolar Response in C60 Single Crystals. Science 257, 511–514 (1992).pubmed.ncbi.nlm.nih.gov/17778682
  36. [36]Long, J. P., Chase, S. J. & Kabler, M. N. Photoelectron spectroscopy and dynamics of excitons in C60 and photopolymerized C60 films. Chem. Phys. Lett. 347, 29–35 (2001).DOI
  37. [37]Benini, M., Shumilin, A., Rakshit, R., Sahoo, A., Halder, A., Droghetti, A., Cugini, F., Solzi, M., Bisero, D., Graziosi, P., Riminucci, A., Bergenti, I., Singh, M., Gnoli, L., Sanna, S., Mertelj, T., Kabanov, V., Sanvito, S. & Dediu, V. The Ferromagnetic Glass State: collapse of the standard ferromagnetic domain structure. (2024). doi:10.21203/rs.3.rs-4540787/v1
  38. [38]Mertens, F., Mönkebüscher, D., Parlak, U., Boix-Constant, C., Mañas-Valero, S., Matzer, M., Adhikari, R., Bonanni, A., Coronado, E., Kalashnikova, A. M., Bossini, D. & Cinchetti, M. Ultrafast Coherent THz Lattice Dynamics Coupled to Spins in the van der Waals Antiferromagnet FePS3. Adv. Mater. 35, e2208355 (2023).
  39. [39]Mertens, F., Terschanski, M., Mönkebüscher, D., Ponzoni, S., Bossini, D. & Cinchetti, M. Wide spectral range ultrafast pump–probe magneto-optical spectrometer at low temperature, high-magnetic and electric fields. Review of Scientific Instruments 1–9 (2020). doi:10.1063/5.0024449DOI
  40. [40]Rocha, A. R., García-Suárez, V. M., Bailey, S., Lambert, C., Ferrer, J. & Sanvito, S. Spin and molecular electronics in atomically generated orbital landscapes. Phys. Rev. B 73, 085414 (2005).DOI
  41. [41]Soler, J. M., Artacho, E., Gale, J. D., García, A., Junquera, J., Ordejón, P. & Daniel Sánchez-Portal. The SIESTA method for ab initio order-N materials simulation. J. Phys.: Condens. Matter 14, 2745 (2002).DOI
  42. [42]Rungger, I. & Sanvito, S. Algorithm for the construction of self-energies for electronic transport calculations based on singularity elimination and singular value decomposition. Phys. Rev. B 78, 035407 (2008).DOI
  43. [43]Troullier, N. & Martins, J. L. Efficient pseudopotentials for plane-wave calculations. Phys. Rev. B 43, 1993–2006 (1991).pubmed.ncbi.nlm.nih.gov/9997467
  44. [44]Barth, U. von & Hedin, L. A local exchange-correlation potential for the spin polarized case. J. Phys. C: Solid State Phys. 5, 1629 (1972).DOI

Paper details

vlnc:W_uednzps6
Version
v1 · latest
Published
Fresh — added to the Hub within the last 30 days
Added
License
CC BY 4.0
Usage

views · 0

downloads · 0

A view counts once per browser per day, a download each time the PDF is requested.

  • Physics and Astronomy

Not recorded for this paper

View statement
Looking for related papers…
Contents