Measurements related to the magnetism of curium metal

We report a series of measurements on a $<1$ mg sample of ${}^{248}\mathrm{Cm}$ metal. Magnetization measurements are similar to those reported earlier. X-ray absorption and magnetic circular dichroism (XMCD) measurements at the ESRF synchrotron at the $M_{4,5}$ absorption edges show that the element has an orbital moment of $\sim 6\%$ of the spin moment, and the two are parallel. This confirms the predictions of intermediate coupling, which has been suggested for Cm by other spectroscopies. In an XMCD measurement 2.5 years later, a field of 17 T was found to induce a larger induced signal (by a factor of $\sim 2$) than that found earlier, indicating a change in the sample. We attribute this to the gradual loss of crystallinity with time since preparation, as a result of the damaged caused by fragments from the spontaneous fission within the sample itself. The final measurements aimed to determine the antiferromagnetic structure with resonant x rays were unable to observe any diffraction peaks from the sample. The ground-state antiferromagnetic structure remains unresolved.

Figure 1

(a) Magnetization performed on a thin foil (1.28 mg) of Cm in 2007/02 using a QD-MPMS-7 device. The figure shows magnetization at $H=7$ T. By plotting $H/M$, the data above 100 K can be fit to $M/H=\chi ={\chi }_0+\frac{C}{T+{\theta }_p}$, where we deduce ${\chi }_0$= $4.3\times {10}^{-3}$ emu/mol, ${\theta }_p$= +286 K, and ${\mu }_{\mathrm{eff}}$= 7.58 ${\mu }_B$. (b) The original susceptibility deduced at the beginning of the project (2007/02) on the total sample of 1.28 mg is shown as the black line. The red line shows the magnetization as deduced from the 0.55-mg sample, which was used for the XAS and XMCD experiments, measured in 2016/10 after the x-ray measurements. The piece was extracted from the original 1.28-mg sample, and still shows the local maximum in the susceptibility at 65 K.

Figure 2

Absorption (black, left-hand side) and XMCD (red, right-hand side) from ${}^{248}\mathrm{Cm}$ taken at 30 K with an applied magnetic field of 17 T.

Figure 3

Overview of XMCD spectra at the $M_{4,5}$ edges for all actinides up to Cm with the $M_5$ energy chosen as $E=0$. Note that these data have been normalized at the $M_4$ edge with an amplitude of unity. This is somewhat misleading in comparing the two edges, as $M_4$ spectral intensity itself decreases in amplitude from U to Am. This effect arises from the decrease in the number of holes in the $j=5/2$ subshell. The sudden change of sign in the $M_4$ line in Cm, as compared to the earlier elements, shows that the orbital and spin moments are now parallel, whereas they are antiparallel for U to Am. ${\mathrm{UFe}}_2$–unpublished data: ${\mathrm{NpFe}}_2$ and ${\mathrm{PuFe}}_2$–Ref. [20]; ${\mathrm{AmFe}}_2$–Ref. [2]; Cm–this work.

Figure 4

(a) $M_4$ XMCD signal as a function of temperature after cooling the sample to 2 K, applying 17 T, and then reducing the magnetic field to $H=7$ T. (b) Plot of the peak $M_4$ dichroic signal, which should be proportional to the susceptibility $M/H$ plotted as $H/M$ vs $T$. Notice the straight-line fit, with no sign of $T_N\sim 65$ K, as in Fig. 1, and the intercept of −33 K.