Studies of the non-Fermi-liquid behavior in $\text{U}{\left({\text{Pt}}_{0.94}{\text{Pd}}_{0.06}\right)}_3$

Coexistent magnetism and non-Fermi-liquid (nFl) behavior occur in only a few of the more than 100 known nFl systems. We have investigated the resistivity, dc magnetic susceptibility, and specific heat as a function of field (0–13 T) down to 0.4 K (0.09 K for resistivity) in high-purity polycrystalline $\text{U}{\left({\text{Pt}}_{0.94}{\text{Pd}}_{0.06}\right)}_3$, where antiferromagnetism at 6 K is known to precede non-Fermi-liquid behavior in the specific heat at lower temperatures in zero field. Unusually for a system that shows nFl behavior in the bulk specific heat, both the dc magnetic susceptibility $\chi$ and the electrical resistivity $\rho$ of $\text{U}{\left({\text{Pt}}_{0.94}{\text{Pd}}_{0.06}\right)}_3$ show Fermi-liquid behavior down to our lowest temperature of measurement. Possible inferences about the $q$ dependence of the non-Fermi-liquid fluctuation spectrum are discussed. As expected, magnetic field succeeds in driving the specific heat of the system back into a Fermi-liquid ground state with $C/T$ saturating to a constant value below about 0.9 K.

Figure 1

(Color online) Electrical resistivity $\rho$ vs temperature $T$ for $\text{U}{\left({\text{Pt}}_{0.94}{\text{Pd}}_{0.06}\right)}_3$. Note the anomaly at $T_N$ around 6 K in the inset showing the higher temperature data. The zero-field low-temperature data, together with a fit of the temperature dependence, are shown here on an expanded scale. The resultant temperature dependence is consistent with Fermi-liquid behavior. Field data $H=13\text{T}$ (not shown) showed considerably more noise for $T<0.25\text{K}$, but again gave a temperature dependence consistent with $\rho \sim T^2$.

Figure 2

(Color online) Magnetic susceptibility in 0.5 T from 0.4 to 5 K of $\text{U}{\left({\text{Pt}}_{0.94}{\text{Pd}}_{0.06}\right)}_3$ measured in a Faraday magnetometer, together with data from 2–10 K taken on an adjacent piece of sample in a Quantum Design MPMS system. Note the anomaly at $T_N\sim 6\text{K}$. In the expanded view in the insert, data (Ref. 18) for the heavy Fermion antiferromagnet ${\text{U}}_2{\text{Zn}}_{17}$, $T_N=9.7\text{K}$, are also shown for comparison down to 1.4 K.

Figure 3

(Color online) Specific heat divided by temperature $C/T$ vs $\text{log}T$ in 0 and 13 T applied magnetic field. The red solid line is a fit of the low-temperature zero-field data to $\text{log}T$. Note that with the suppression of the magnon contribution to $C$ caused by the 13 T applied field that $C/T\left(13\text{T}\right)$ varies as $\text{log}T$ up to about 4 K. Note further the saturation behavior in $C/T\left(13\text{T}\right)$ below about 0.9 K, implying that field suppresses the nFl behavior in $\text{U}{\left({\text{Pt}}_{0.94}{\text{Pd}}_{0.06}\right)}_3$ just as it does[2, 3] in other nonmagnetic nFl systems.

Figure 4

(Color online) The first derivative of $C/T$ vs $T$ to delineate more accurately the suppression of $T_N$ with 13 T magnetic field. Two temperatures in each field are marked with an arrow, the midpoint and the low-temperature-end finish of the jump in $d\left(C/T\right)/dT$. Thus, the suppression of the antiferromagnetic ordering anomaly in 13 T is approximately 5.70–4.91 K (0.79 K) or 5.41–4.50 K (0.91 K).