Surface Aligned Magnetic Moments and Hysteresis of an Endohedral Single-Molecule Magnet on a Metal

The interaction between the endohedral unit in the single-molecule magnet ${\mathrm{Dy}}_2\mathrm{ScN}@{\mathrm{C}}_{80}$ and a rhodium (111) substrate leads to alignment of the Dy $4f$ orbitals. The resulting orientation of the ${\mathrm{Dy}}_2\mathrm{ScN}$ plane parallel to the surface is inferred from comparison of the angular anisotropy of x-ray absorption spectra and multiplet calculations in the corresponding ligand field. The x-ray magnetic circular dichroism is also angle dependent and signals strong magnetocrystalline anisotropy. This directly relates geometric and magnetic structure. Element specific magnetization curves from different coverages exhibit hysteresis at a sample temperature of $\sim 4\mathrm{K}$. From the measured hysteresis curves, we estimate the zero field remanence lifetime during x-ray exposure of a submonolayer to be about 30 seconds.

Figure 1

(a) Ball-and-stick model of ${\mathrm{Dy}}_2\mathrm{ScN}@{\mathrm{C}}_{80}$. (b) Measurement geometry with the angular momentum of the x rays $L_{ph}$, parallel or antiparallel to the magnetic field and at an angle of $\theta$ with respect to the normal of the Rh(111) surface.

Figure 2

(a) XAS measured at the Dy $M_5$ edge from a sub-ML of ${\mathrm{Dy}}_2\mathrm{ScN}@{\mathrm{C}}_{80}/\mathrm{Rh}(111)$ $T=4\mathrm{K}$, ${\mu }_{0}H=6.5\mathrm{T}$, measurement geometry in Fig. 1. Each data set is normalized to the integrated intensity. (b) Calculated absorption with the x-ray beam and external field oriented parallel $I_z$, and perpendicular $I_x$, to the magnetic easy axis (Dy-N bond).

Figure 3

Sub-ML of ${\mathrm{Dy}}_2\mathrm{ScN}@{\mathrm{C}}_{80}/\mathrm{Rh}(111)$, $T=4\mathrm{K}$, ${\mu }_{0}H=6.5\mathrm{T}$, measurement geometry of Fig. 1. The polarization dependent XAS spectra (left panel), and the corresponding XMCD spectra (right panel), were measured at an incidence angle of $\theta =0°$ (a) and $\theta =60°$ (b). (c) Angle dependence of the integrated XMCD signal normalized to the integrated XAS over the Dy $M_5$ edge. The dashed line corresponds to the expected angle dependence for magnetic moments oriented parallel to the surface, whereas the blue line takes into account a Gaussian distribution, centered in the surface plane and with a standard deviation of 16°.

Figure 4

Hysteresis curves measured from a multilayer and a sub-ML of ${\mathrm{Dy}}_2\mathrm{ScN}@{\mathrm{C}}_{80}/\mathrm{Rh}(111)$ at a magnetic field sweep rate of $2\mathrm{T}/\min$ and a sample temperature of $\sim 4\mathrm{K}$. The x-ray flux was $5\times 10^{10}\text{photons}/{\mathrm{mm}}^2/\mathrm{s}$. The data were recorded with the x-ray beam and the magnetic field at an incidence angle of $\theta =60°$. The magnetization curves correspond to the average of several independent measurements, where the error bars are the standard deviation at each external magnetic field. The arrows indicate the ramping direction of the magnetic field, the lines are guides to the eye, and $m_{\text{sat}}$ is the saturated value at $\pm 6.5\mathrm{T}$. The drop in magnetization at zero field is a consequence of the time of 30 s it takes for the magnet to switch polarity.