Antiferromagnetic resonance in methylaminated potassium fulleride $\left(\mathrm{C}{\mathrm{H}}_3\mathrm{N}{\mathrm{H}}_2\right){\mathrm{K}}_3{\mathrm{C}}_{60}$

High-frequency magnetic resonance measurements $({\nu }_L=9.6–420\mathrm{GHz})$ were employed to investigate the low-temperature antiferromagnetic ground state of the $\left(\mathrm{C}{\mathrm{H}}_3\mathrm{N}{\mathrm{H}}_2\right){\mathrm{K}}_3{\mathrm{C}}_{60}$ fulleride. The frequency and temperature dependence of the intensity, linewidth, and center of the resonance signal detected below $T_N$ are characteristic of antiferromagnetic resonance (AFMR). The AFMR intensity is consistent with an ordered magnetic moment of ${\mu }_{\mathit{eff}}=0.7\left(1\right){\mu }_B∕{\mathrm{C}}_{60}$, while the narrowing of the AFMR signal with increasing resonance frequency can be modeled with a spin-flop field of $H_{\mathit{sf}}=840\left(80\right)\mathrm{G}$ and a $g$-factor anisotropy of $\delta \gamma =710\left(50\right)\mathrm{ppm}$. We stress that the spin-flop field is reduced compared to the ammoniated analog $\left(\mathrm{N}{\mathrm{H}}_3\right){\mathrm{K}}_3{\mathrm{C}}_{60}$ on the account of reduced $\mathrm{C}_{60}{}^{3-}\text{−}\mathrm{C}_{60}{}^{3-}$ exchange interactions. Differences in the level of the anisotropic expansion between $\mathrm{C}{\mathrm{H}}_3\mathrm{N}{\mathrm{H}}_2$ and $\mathrm{N}{\mathrm{H}}_3$ cointercalated fullerides are likely to be responsible for the differences in the electronic structure between the two systems and ultimately may account for the reduced Néel temperature in $\left(\mathrm{C}{\mathrm{H}}_3\mathrm{N}{\mathrm{H}}_2\right){\mathrm{K}}_3{\mathrm{C}}_{60}$.

Figure 1

Temperature dependence of the integrated magnetic resonance spectra of polycrystalline $\left(\mathrm{C}{\mathrm{H}}_3\mathrm{N}{\mathrm{H}}_2\right){\mathrm{K}}_3{\mathrm{C}}_{60}$ measured at $9.6\mathrm{GHz}$ (left panel) and $51.6\mathrm{GHz}$ (right panel).

Figure 2

Comparison of the temperature dependence of the linewidth measured at $9.6\mathrm{GHz}$ (solid circles) and $51.6\mathrm{GHz}$ (open circles) in $\left(\mathrm{C}{\mathrm{H}}_3\mathrm{N}{\mathrm{H}}_2\right){\mathrm{K}}_3{\mathrm{C}}_{60}$ powder.

Figure 3

EPR spectra measured at $T=4\mathrm{K}$ and at resonance frequencies of 24.9, 51.6, and $103\mathrm{GHz}$. The dotted vertical lines are guides to the eyes to show the linewidth reduction with increasing resonance frequency. The defect impurity line has been subtracted from the raw data.

Figure 4

Frequency dependence of the EPR linewidth measured at $4\mathrm{K}$ (solid circles), $8\mathrm{K}$ (open triangles), and $25\mathrm{K}$ (stars). The solid line is a fit of the $T=4\mathrm{K}$ data to Eq. (2) with $\delta \gamma =710\left(50\right)\mathrm{ppm}$ and $H_{sf}=840\left(80\right)\mathrm{G}$, while the dotted line is a linear fit of the $T=25\mathrm{K}$ data with $\delta \gamma =750\left(30\right)\mathrm{ppm}$.

Figure 5

Temperature dependence of the paramagnetic resonance signal intensity (open circles, determined from the $X$-band EPR data) is compared with the intensity of the $51.6\mathrm{GHz}$ data (solid circles), which is below $T_N=11\mathrm{K}$ related to the perpendicular susceptibility ${\chi }_{\perp }$ given by the antiferromagnetic resonance signal intensity (see text for details).

Figure 6

Temperature dependence of the normalized ${[M_2^{\mathit{AFMR}}\left(T\right)∕M_2^{\mathit{AFMR}}(T=0\mathrm{K})]}^{1∕4}$ parameter, which is a measure of the sublattice magnetization $M$ obtained from the high-frequency EPR experiments at ${\nu }_L=51.6\mathrm{GHz}$. The solid line is a fit of the data to a critical law ${\left[\frac{(T_N-T)}{T_N}\right]}^{\beta }$ with $T_N=11\mathrm{K}$ (fixed) and $\beta =0.35\left(1\right)$.