Unconventional enhancement of ferromagnetic interactions in Cd-doped $\mathrm{Gd}{\mathrm{Fe}}_2{\mathrm{Zn}}_{20}$ single crystals studied by ESR and ${}^{57}\mathrm{Fe}$ Mössbauer spectroscopies

Single crystals of $\mathrm{Gd}{\mathrm{Fe}}_2{\mathrm{Zn}}_{20-x}{\mathrm{Cd}}_x$ ($0.0<x<1.4$) were grown and characterized through structural, magnetic, and electronic properties using x-ray diffraction, field- and temperature-dependent magnetization, specific heat, ${}^{57}\mathrm{Fe}$ Mössbauer spectroscopy, and electron spin resonance (ESR). A negative chemical pressure effect is accompanied by an unexpected increase of $T_C$ from 86 to 96 K, together with a reduction of the magnetic effective moment and saturation magnetic moment, as evidenced by all of the experimental techniques. From the microscopic point of view, probing at the $4f$ electron level and the Fe nucleus has allowed the extraction of important information about the configuration and the effective role of the partial Cd substitution for Zn in this ferromagnetic system. Our ${}^{57}\mathrm{Fe}$ Mössbauer spectroscopy experiments show a negligible variation of the hyperfine field at the Fe site, and ESR experiments reveal an enhancement of the Korringa-type relaxation and a molecular field effect as Cd is incorporated. This complex behavior is assigned to a possible reconstruction of the Fermi surface and/or a new distribution of the $d$ type of conduction electrons in response to the negative chemical pressure, leading to an enhancement of the ferromagnetic transition temperature in a generalized Ruderman-Kittel-Kasuya-Yosida interaction scenario.

Figure 1

Observed (black line with plus symbol), calculated (red line), background (green line), and difference (blue line) XRD profiles of $\mathrm{Gd}{\mathrm{Fe}}_2{\mathrm{Zn}}_{20-x}{\mathrm{Cd}}_x$ ($0\le x\le 1.4$) recorded at room temperature. The Rietveld refinement was performed using GSAS. Asterisk denotes Zn flux.

Figure 2

(a) $C_P\times T$ for $\mathrm{Gd}{\mathrm{Fe}}_2{\mathrm{Zn}}_{20-x}{\mathrm{Cd}}_x$ (0 $\le x\le$ 1.4) and (b) $C_P/T\times T^2$ for the reference system $\mathrm{Y}{\mathrm{Fe}}_2{\mathrm{Zn}}_{20}$ and $\mathrm{Y}{\mathrm{Fe}}_2{\mathrm{Zn}}_{18.6}{\mathrm{Cd}}_{1.4}$.

Figure 3

dc magnetic susceptibility of $\mathrm{Gd}{\mathrm{Fe}}_2{\mathrm{Zn}}_{20-x}{\mathrm{Cd}}_x$ for 0.0 $\le x\le$ 1.4.

Figure 4

Magnetization ($M$) vs magnetic field ($H$) at 2 K for $\mathrm{Gd}{\mathrm{Fe}}_2{\mathrm{Zn}}_{20-x}{\mathrm{Cd}}_x$ for 0.0 $\le x\le$ 1.4 and a zoom for low magnetic fields (inset).

Figure 5

Temperature-dependent ${}^{57}\mathrm{Fe}$ Mössbauer spectra for $\mathrm{Gd}{\mathrm{Fe}}_2{\mathrm{Zn}}_{20-x}{\mathrm{Cd}}_x$ with (a) $x=0$ and (b) $x=1.4$. In addition, the Fe sites employed for the fits are shown above and below $T_C$.

Figure 6

${}^{57}\mathrm{Fe}$ Mössbauer hyperfine parameters for $\mathrm{Gd}{\mathrm{Fe}}_2{\mathrm{Zn}}_{20-x}{\mathrm{Cd}}_x$: (a) Chemical isomer shift $\delta$ for the main and the so-called impurity phases, (b) electric quadrupole splitting $\mathrm{\Delta }{E}_Q$ (dotted lines represent the values at low temperatures above $T_C$), and (c) magnetic hyperfine field at the Fe site $B_{hf}$ and their respective fitting curves, and (d) the resonance linewidth $\mathrm{\Gamma }$ (dotted lines represent the values found for the nonmagnetic specimens $\mathrm{Lu}{\mathrm{Fe}}_2{\mathrm{Zn}}_{20}$ and $\mathrm{Yb}{\mathrm{Fe}}_2{\mathrm{Zn}}_{20}$ [10]).

Figure 7

ESR of ${\mathrm{Gd}}^{3+}$ at 9.5 GHz, $2\mu \mathrm{W}$ at $T=110$ K for $\mathrm{Gd}{\mathrm{Fe}}_2{\mathrm{Zn}}_{20-x}{\mathrm{Cd}}_x$ ($0.0\le x\le 1.4$).

Figure 8

$T$-dependent ${\mathrm{Gd}}^{3+}$ ESR linewidth $\mathrm{\Delta }H$ for $\mathrm{Gd}{\mathrm{Fe}}_2{\mathrm{Zn}}_{20-x}{\mathrm{Cd}}_x$ ($0.0\le x\le 1.4$).

Figure 9

$g$ shift ($\mathrm{\Delta }g$) for $\mathrm{Gd}{\mathrm{Fe}}_2{\mathrm{Zn}}_{20-x}{\mathrm{Cd}}_x$ for $0.0\le x\le 1.4$.

Figure 10

(a) Comparison of the temperature dependence of the dc magnetic susceptibility and effective $g$ shift for $\mathrm{Gd}{\mathrm{Fe}}_2{\mathrm{Zn}}_{20-x}{\mathrm{Cd}}_x$ for 0.0 $\le x\le$ 1.4 near to their respective FM transition and (b) internal magnetic field [$H_{\mathrm{int}}$ = $H_{\mathrm{res}}$(low $T$)-$H_{\mathrm{res}}$(high $T$) vs magnetization ($M$) evaluated at the same temperature.