Low temperature specific heat, resistivity, and magnetization for ${\mathrm{Ce}}_{1-x}{\mathrm{Th}}_x\mathrm{RhSb}$ $(x=0.2,$ 0.3 and 0.4), have been investigated. ${\mathrm{Ce}}_{0.6}{\mathrm{Th}}_{0.4}\mathrm{RhSb}$ and ${\mathrm{Ce}}_{0.7}{\mathrm{Th}}_{0.3}\mathrm{RhSb}$ show a clear magnetic behavior at 0.35 and 0.15 K, respectively, while ${\mathrm{Ce}}_{0.8}{\mathrm{Th}}_{0.2}\mathrm{RhSb},$ with a slight remanent magnetic behavior at $\sim 0.08\mathrm{K},$ shows a non-Fermi-liquid behavior in the specific heat over almost two decades of temperature in the vicinity of $x_{\mathrm{crit}}$ (where $T_{\mathrm{mag}}\to 0$ at $x_{\mathrm{crit}}),$ consistent with a quantum critical point scenario. The low temperature specific heat divided by temperature, measured between 0.05 and 8 K, of ${\mathrm{Ce}}_{0.8}{\mathrm{Th}}_{0.2}\mathrm{RhSb}$ can be fit to either a logarithmic temperature dependence (consistent with various theories for behavior near a quantum critical point) between $\sim 0.15$ and 8 K or to a $T^{-1+\lambda }$ temperature dependence (consistent with the Griffiths phase disorder theory) with $\lambda =0.74,$ but only between 0.05 and 1.3 K. The low temperature magnetic susceptibility, measured down to 1.8 K, of ${\mathrm{Ce}}_{1-x}{\mathrm{Th}}_x\mathrm{RhSb}$ $(x=0.2,$ 0.3, and 0.4) also exhibits a power-law temperature dependence (Griffiths phase model), with an exponent λ $(\sim 0.6)$ comparable to that found from the specific heat data. Electrical resistivity of ${\mathrm{Ce}}_{0.8}{\mathrm{Th}}_{0.2}\mathrm{RhSb}$ follows approximately $\rho ={\rho }_0-AT$ between 0.2 and 2 K with both a very large ${\rho }_0$ (1210 μΩ cm) and a gigantic coefficient A (40.7 μΩ cm/K) which is a factor of six larger than the previous record value for a non-Fermi-liquid system found in ${\mathrm{UCu}}_4\mathrm{Pd}.$ The possibility that these unusually large values are related to the gap formation seen in pure CeRhSb at low temperatures is discussed. As a further method to resolve whether a quantum critical point or a disorder model best describe this system, the field dependences of the magnetization at 1.8 K and the specific heat down to 0.06 K are compared to predictions for Griffiths phase behavior. There is good agreement between the theory and the magnetization behavior with field, while the specific heat data in field deviate from the theory’s predictions at low temperatures, again displaying a Fermi-liquid behavior below 0.3 K. This reentrance in magnetic field into the Fermi-liquid state below a temperature $T^{*}$ could be explained by invoking freezing of the spin cluster tunneling due to dissipation effects below a crossover temperature $T^{*}.$ The physical properties of the ${\mathrm{Ce}}_{1-x}{\mathrm{Th}}_x\mathrm{RhSb}$ system are compared to those found for two non-Fermi-liquid systems: the ${\mathrm{UCu}}_{5-x}{\mathrm{Pd}}_x$ system, where $C/T$ is best fit by $\log T$ and some sort of quantum critical scenario (perhaps including spin glass effects) appears to obtain; and the ${\mathrm{Ce}}_{1-x}{\mathrm{La}}_x{\mathrm{RhIn}}_5$ system, where $C/T$ is best fit by $T^{-1+\lambda }$ and a Griffiths phase model has been applied.

• Received 23 December 2002

DOI:https://doi.org/10.1103/PhysRevB.67.184401