Competing magnetic phases and field-induced dynamics in DyRuAsO

Analysis of neutron diffraction, dc magnetization, ac magnetic susceptibility, heat capacity, and electrical resistivity for DyRuAsO in an applied magnetic field are presented at temperatures near and below those at which the structural distortion ($T_S=25$ K) and subsequent magnetic ordering ($T_N=10.5$ K) take place. Powder neutron diffraction is used to determine the antiferromagnetic order of Dy moments of magnitude 7.6(1)${\mu }_B$ in the absence of a magnetic field, and demonstrate the reorientation of the moments into a ferromagnetic configuration upon application of a magnetic field. Dy magnetism is identified as the driving force for the structural distortion. The magnetic structure of analogous TbRuAsO is also reported. Competition between the two magnetically ordered states in DyRuAsO is found to produce unusual physical properties in applied magnetic fields at low temperature. An additional phase transition near $T^{*}=3$ K is observed in heat capacity and other properties in fields $\gtrsim 3$ T. Magnetic fields of this magnitude also induce spin-glass-like behavior including thermal and magnetic hysteresis, divergence of zero-field-cooled and field-cooled magnetization, frequency dependent anomalies in ac magnetic susceptibility, and slow relaxation of the magnetization. This is remarkable since DyRuAsO is a stoichiometric material with no disorder detected by neutron diffraction, and suggests analogies with spin-ice compounds and related materials with strong geometric frustration.

Figure 1

(a) Neutron powder-diffraction patterns from DyRuAsO collected in the tetragonal state at 40 K, the orthorhombic state at 15 K, and the magnetically ordered state at 1.5 K. The insets in (a) show the splitting of the tetragonal 400 reflection resulting from the structural transition and the intensity of the 001 reflection (labeled in the main panel) as a function of temperature, illustrating the onset of magnetic order below about 10 K. (b) Rietveld fits of the nuclear and magnetic structures of DyRuAsO at 1.5 K. The lower ticks locate reflections from the Dy${}_2$O${}_3$ impurity. The AFM arrangement of the Dy moments determined from the diffraction analysis is shown in (c).

Figure 2

(a) Neutron powder-diffraction patterns collected at 3 K with applied magnetic field of ${\mu }_{0}H$ = 0, 2, and 5 T. The difference between the data collected at 2 T and zero field, and between the data collected at 5 and 2 T are also shown. The lowest two patterns are simulations including only magnetic scattering for AFM and FM moments on Dy. Patterns are offset vertically for clarity. (b) The field dependence of the relative scattered intensity at the peaks marked by the square and circle in (a), showing a divergence for fields above about 2 T. The data are labeled by Miller indices of the overlapping reflections which contribute magnetic scattering intensity to the measured peaks. (c) Rietveld refinement results for ${\mu }_{0}H=5$ T and $T=3$ K using a predominately FM model (see text for details) with all Dy moments along the $b$ axis.

Figure 3

(a) The field dependence of the magnetic moment per formula unit determined from dc magnetization measurements. (b) Low-temperature behavior of $M/H$ measured on warming after zero field cooling (ZFC) and field cooling (FC), showing a divergence which is strongest at intermediate fields.

Figure 4

(a) Heat capacity of DyRuAsO at the indicated magnetic fields. The insets shows the behaviors near $T_S=25\text{K}$ and $T*=3\text{K}$. (b) The entropy change determined by integration of $c_P/T$, after subtracting a estimated background contribution determined by scaling data from LaRuAsO to match the data measured at zero field and 50 K.

Figure 5

(a) Resistivity ($\rho$) below 50 K in the main panel, with the behavior near and below $T_N$ shown in the inset. Data were collected on cooling in the indicated applied field (open markers), followed immediately by warming in the same field (solid markers). (b) Temperature derivative of $\rho$ with phase-transition temperatures marked on the plot. The curves in (b) are offset vertically for clarity.

Figure 6

(a) Time dependence of the magnetic moment of DyRuAsO after increasing the field as indicated in the legend at a temperature of 2 K. (b) Time dependence of the magnetic moment after increasing the field from 2 to 3 T at $T=2$, 4, and 6 K. (c) Field dependence of the magnetic moment measured upon increasing then decreasing the field at $T=2$ and 4 K. (d) Field dependence of the magnetoresistance relative to the zero-field resistivity values measured upon increasing then decreasing the field at $T=2$ and 4 K.

Figure 7

(a) Heat capacity, (b) electrical resistivity, (c) dc “susceptibility” ($M$/$H$), and (d) ac magnetization of DyRuAsO near $T^{*}$ in an applied magnetic field of 3 T. In (c) the temperature derivative of $M$/$H$ is also shown, in arbitrary units.

Figure 8

(a)–(d) Temperature dependence of the real ($m^{\prime }$) and imaginary ($m^{\prime \prime }$) parts of the ac magnetization measured at ${\mu }_{0}H=0$ and 3 T using frequencies indicated in panel (a). (e),(f) Contour plots of the real part ($m^{\prime }$) and imaginary part ($m^{\prime \prime }$) of the ac magnetization measured at 100 Hz as functions of temperature and applied dc magnetic field. (g) Arrhenius fit using the temperatures ($T_P$) at which $m^{\prime }$ peaks at different frequencies for $H=3$ T. (h) Frequency dependence of $m^{\prime \prime }$ at $H=3$ T and the indicated temperatures. All measurements were conducted using a sample of mass 27.1 mg and an ac excitation field of 10 Oe.