cond-mat0304163/afm.tex
1: % ****** Start of file apssamp.tex ******
2: %
3: %   This file is part of the APS files in the REVTeX 4 distribution.
4: %   Version 4.0 of REVTeX, August 2001
5: %
6: %   Copyright (c) 2001 The American Physical Society.
7: %
8: %   See the REVTeX 4 README file for restrictions and more information.
9: %
10: % TeX'ing this file requires that you have AMS-LaTeX 2.0 installed
11: % as well as the rest of the prerequisites for REVTeX 4.0
12: %
13: % See the REVTeX 4 README file
14: % It also requires running BibTeX. The commands are as follows:
15: %
16: %  1)  latex apssamp.tex
17: %  2)  bibtex apssamp
18: %  3)  latex apssamp.tex
19: %  4)  latex apssamp.tex
20: %
21: %\documentclass[twocolumn,showpacs,preprintnumbers,amsmath,amssymb]{revtex4}
22: %\documentclass[preprint,showpacs,preprintnumbers,amsmath,amssymb]{revtex4}
23: 
24: % Some other (several out of many) possibilities
25: %\documentclass[preprint,aps]{revtex4}
26: %\documentclass[preprint,aps,draft]{revtex4}
27: %\documentclass[twocolumn,aps,prb]{revtex4}% Physical Review B
28: %\documentclass[preprint,showpacs,aps,prb,draft,amsmath,amssymb]{revtex4}% Physical Review B
29: %\documentclass[preprint,showpacs,aps,prb,amsmath,amssymb]{revtex4}% Physical Review B
30: 
31: \documentclass[showpacs,prb,amsmath,amssymb]{revtex4}% Physical Review B
32: 
33: \usepackage{graphicx}% Include figure files
34: %\usepackage{dcolumn}% Align table columns on decimal point
35: %\usepackage{bm}% bold math
36: 
37: %\nofiles
38: 
39: \begin{document}
40: 
41: %\preprint{APS/123-QED}
42: 
43: \title{Separation Between Antiferromagnetic and Ferromagnetic Transitions in
44: Ru$_{1-x}$Cu$_{x}$Sr$_{2}$EuCu$_{2}$O$_{8+\delta}$}
45: 
46: \author{Y.~Y.~Xue$^{1}$}
47: \author{F.~Chen$^{1}$}
48: \author{J.~Cmaidalka$^{1}$}
49: \author{R.~L.~Meng$^{1}$}
50: \author{C.~W.~Chu$^{1,2,3}$}
51: \affiliation{$^{1}$Department of Physics and Texas Center for Superconductivity, 
52: University 
53: of Houston, 202 Houston Science Center, Houston, Texas 77204-5002}
54: \affiliation{$^{2}$Lawrence Berkeley National Laboratory, 1 Cyclotron Road, 
55: Berkeley, California 94720}
56: \affiliation{$^{3}$Hong Kong University of Science and Technology, 
57: Hong Kong}
58: 
59: \date{\today}% It is always \today, today,
60:              %  but any date may be explicitly specified
61: 
62: \begin{abstract}
63: The macroscopic magnetizations of Ru$_{1-x}$Cu$_{x}$Sr$_{2}$EuCu$_{2}$O$_{8+\delta}$ with 
64: $x$ between 0 and 0.15 were investigated. A ferromagnet-like transition as well as an 
65: antiferromagnet-like transition appear around $T_{M}$ in the low-field magnetization and 
66: around $T_{AM}$ in the high-field differential susceptibility, respectively. The separation 
67: between them, which is accompanied by a flat plateau in the magnetic $C_{p}$, increases with 
68: $x$. Superparamagnetic $M(H)$ and slow spin dynamics, i.e. characteristics of nanomagnetic 
69: clusters, were observed far above $T_{M}$. A comparison with 
70: RuSr$_{2}$(Eu$_{1-y}$Ce$_{y}$)Cu$_{2}$O$_{10+\delta}$ and some manganites further suggests 
71: that a phase separation occurs, which can describe well the conflicting 
72: magnetic-superconductivity data previously reported.
73: \end{abstract}
74: 
75: \pacs{74.81.-g, 74.72.-h, 75.40.Cx}% PACS, the Physics and Astronomy
76:                              % Classification Scheme.
77: %\keywords{Suggested keywords}%Use showkeys class option if keyword
78:                               %display desired
79: \maketitle
80: 
81: The puzzling bulk, yet granular, superconductivity (SC) in rutheno-cuprates 
82: RuSr$_{2}$RCu$_{2}$O$_{8+\delta}$ (Ru1212R) and RuSr$_{2}$(R,Ce)$_{2}$Cu$_{2}$O$_{10+\delta}$ 
83: (Ru1222R) with R = Gd, Eu, or Y,\cite{fel,ber,tal,xue} which coexists with a weak 
84: ferromagnetism (FM), is closely related to their magnetic structure. While a homogeneous 
85: canted antiferromagnetic (CAFM) spin-order may coexist with more-or-less ordinary 
86: superconductivity, such as the proposed Meissner state or the $\pi$-phase SC,\cite{ber,pic} 
87: magnetic inhomogeneity at length scales $\ge \xi$ will unavoidably lead to a 
88: Josephson-junction-array-like superconductivity,\cite{xue} where $\xi$ is the coherence 
89: length. In the case of Ru1222R, the reported data seem to indicate a rather complicated 
90: magnetic structure. Both the AFM-like differential-susceptibility maximum of the Ru 
91: ($\chi_{Ru~only}$) and the hyperfine splitting of the M\"{o}ssbauer spectra, for example, 
92: occur at temperatures almost two times higher than the $T_{M}$, where an FM-like transition 
93: occurs in the low-field field-cooled magnetization ($M_{FC}$).\cite{fel,xue2} Either a 
94: phase separation or a multistage transition, therefore, should occur.\cite{fel,xue2} On the 
95: other hand, the situation of Ru1212R has been suggested to be different. The inflection point 
96: $T_{AM}$ at $\partial^{2}(T\chi_{Ru~only})/\partial T^{2} = 0$, which should be the N\'{e}el 
97: temperature in simple antiferromagnets,\cite{fis} and the $T_{M}$ are in rough agreement for 
98: a Ru1212Eu sample.\cite{but} Mean-field-like scaling has also been observed below $T_{M}$ by 
99: both neutron powder diffraction (NPD) and zero-field nuclear magnetic resonance 
100: (ZFNMR).\cite{lyn,tok} It is therefore natural that a simple CAFM was assumed in many 
101: previous investigations. This model, however, faces a dilemma in accommodating the 
102: magnetizations and the ZFNMR and NPD data. The NPD, for example, indicated that the Ru spins 
103: are AFM-aligned (G-type) along the $c$ axis with a very tight upper limit of the FM 
104: components, i.e. $< 0.1$ and $\approx 0.2$~$\mu_{B}$/Ru at $H = 0$ and 0.4~T~$\le H \le$ 7~T, 
105: respectively.\cite{lyn} The spontaneous magnetization of the sample, however, reaches 
106: $M_{r} \approx 800$~emu/mole, i.e. an FM component 0.28~$\mu_{B}$/Ru at $H = 0$. The 
107: extrapolated zero-field magnetization of 0.6~$\mu_{B}$/Ru at 50~K,\cite{but} which may serve 
108: as a lower limit for the FM component at 5~T, is again three times larger. The ZFNMR data, 
109: in addition, demonstrate that the Ru-spins should be aligned perpendicular to the $c$ with a 
110: major (or dominant) FM component.\cite{tok} This unusual magnetic structure, which appears as 
111: G-type AFM along $c$ in NPD but ordered along $a,b$ with a large FM component in both 
112: magnetization and NMR, suggested that the magnetic structure of Ru1212R deserved a 
113: reexamination. It should be pointed out that both the extremely broad $C_{p}$ peak and the 
114: superparamagnet-like $M(H)$ up to 2~$T_{M}$ in Ru1212Gd already suggest that its magnetic 
115: transition is far from simple:\cite{tal,but} the spin correlations may exist up to 2~$T_{M}$ 
116: with a significant entropy and a correlation size as large as 
117: $10^{2}$--$10^{3}$~$\mu_{B}$,\cite{xue2} both being characteristic of phase separation. It is 
118: interesting to note that both $T_{M}$ and $T_{AM}$ of Ru1212R can be tuned by 
119: Cu-doping.\cite{kla} The evolutions of $M$, $\chi_{Ru~only}$, and $C_{p}$ of 
120: Ru$_{1-x}$Cu$_{x}$Sr$_{2}$EuCu$_{2}$O$_{8+\delta}$ with $0 \le x \le 0.15$, therefore, were 
121: measured. The $T_{M}$ drops more than 25~K with $x$ while the variation in $T_{AM}$ is 
122: negligibly small. A separation between $T_{M}$ and $T_{AM}$ is developed with $x$. This 
123: separation is further accompanied by a magnetic $C_{p}/T$ with a flat plateau between $T_{M}$ 
124: and $T_{AM}$. Hence, a mesoscopic phase-separation is suggested.
125: 
126: Ceramic Ru$_{1-x}$Cu$_{x}$Sr$_{2}$EuCu$_{2}$O$_{8+\delta}$ samples with $x$ between 0 and 0.15 
127: were synthesized following the standard solid-state-reaction procedure. Precursors were first 
128: prepared by calcinating commercial oxides at 600--900~$^{\circ}$C under flowing O$_{2}$ at 
129: 1~atm. Mixed powder with a proper cation-ratio was then pressed into pellets and sintered at 
130: 960~$^{\circ}$C. The final heat treatment was done at 1065--1070~$^{\circ}$C for 7 d in oxygen 
131: after repeatedly sintering and regrinding.\cite{xue} The structure of the samples was 
132: determined by powder X-ray diffraction (XRD) using a Rigaku DMAX-IIIB diffractometer. The 
133: $x$ dependence of the lattice parameters, i.e. the $c \approx 11.553(2)$ to 11.550(2) {\AA} for 
134: $x = 0$ and 0.15, respectively, is slightly weaker than that reported for 
135: Ru$_{1-x}$Cu$_{x}$Sr$_{2}$GdCu$_{2}$O$_{8+\delta}$.\cite{kla} Minor impurity phases, likely 
136: SrRuO$_{3}$ or oxides of (Sr,Cu), are below 5\% at $x \le 0.15$ (Fig.~\ref{fig1}). The 
137: composition was measured by a JEOL JXA 8600 electron microprobe with attached wavelength 
138: dispersive spectrometers (WDS). The local inhomogeneity of $1-x$ is within the experimental 
139: resolution of $\pm 0.05$.\cite{lor} The magnetizations were measured using a Quantum Design 
140: SQUID magnetometer with an $ac$ attachment and the specific heat was measured in a 
141: Quantum Design PPMS with a specific-heat attachment.
142: 
143: \begin{figure*}
144: \includegraphics[scale=.5]{fig1.eps}% Here is how to import EPS art
145: \caption{\label{fig1}The XRD of a (Ru$_{0.9}$Cu$_{0.1}$)Sr$_{2}$EuCu$_{2}$O$_{8+\delta}$ 
146: sample. $+$: data; solid line: the Rietveld fit; *: the impurity lines.}
147: \end{figure*}
148: 
149: Superconductivity appears in all the samples below a critical temperature $T_{c} \approx$ 
150: 20--30~K. A single-step jump of $M_{FC}$ also appears with cooling at a higher temperature 
151: (Fig.~\ref{fig2}a). According to the scaling correlation 
152: $(HM_{0}/MH_{0})^{1/\gamma} = t + (M/M_{0})^{1/\beta}$, the $\partial M/\partial T$ of an 
153: ideal ferromagnet should decrease with $t = (T-T_{M})/T_{M}$ as $1/t^{\gamma+1}$ above 
154: $T_{M}$, but increase as $(-t)^{1-\beta}$ below, where $0 < \beta < 1$, $\gamma > 0$, $H_{0}$, 
155: and $M_{0}$ are two critical exponents and two critical amplitudes, respectively. The 
156: situation for a CAFM magnet should be similar. Therefore, the inflection point of $M_{FC}(T)$ 
157: at 5~Oe, i.e. the temperature at which $\partial M_{FC}/\partial T$ peaks, is used as the 
158: $T_{M}$ (Fig.~\ref{fig2}a). The well defined $T_{M}$ and the large FM component below $T_{M}$ 
159: are in rough agreement with those reported for Ru1212Eu,\cite{but} but rather different from 
160: those of Ru$_{1-x}$Cu$_{x}$Sr$_{2}$GdCu$_{2}$O$_{8+\delta}$,\cite{kla} where no clear FM 
161: transition can be identified with $x \ge 0.1$. Differences in both the rare-earth elements and 
162: the synthesis procedures may contribute to the variation. It should be pointed out that the 
163: well defined $T_{M}$ and the large $M_{FC}$ of our samples make the analysis of $M_{FC}$ and 
164: $C_{p}$ easier and without significant interference from the minor impurities. A systematic 
165: decrease of $T_{M}$ with $x$ is observed, e.g. $T_{M} \approx 134$~K and 117~K at $x = 0$ and 
166: 0.1, respectively (Fig.~\ref{fig2}a). It is also interesting to note that the reported 
167: bifurcation point between $M_{ZFC}$ and $M_{FC}$, which should be very close to $T_{M}$ if 
168: the domain pinning is strong, in Ru$_{1-x}$Cu$_{x}$Sr$_{2}$GdCu$_{2}$O$_{8+\delta}$ shows 
169: almost the same $x$ dependence, i.e. down to $\approx 115$~K and 100~K with $x = 0.1$ and 0.2, 
170: respectively.\cite{kla}
171: 
172: \begin{figure*}
173: \includegraphics[scale=.5]{fig2.eps}% Here is how to import EPS art
174: \caption{\label{fig2}a) The magnetizations. For the $x = 0$ sample, $\circ$: $M_{FC}$(5~Oe); 
175: $+$: $H/M_{FC}$(1~T); $\bullet$: $1/\chi_{Ru~only}$; solid line: C-W fit. For the $x = 0.1$ 
176: sample, $\bigtriangledown$: $M_{FC}$(5~Oe); $\blacktriangledown$: $1/\chi_{Ru~only}$. b) The 
177: spin entropy. $\circ$: the magnetic $C_{p}/T$ of the $x = 0$ sample; $\bigtriangledown$: that 
178: of the $x = 0.1$ sample; solid line: $\partial(T\chi)/\partial T$ of the $x = 0$ sample; 
179: dashed 
180: line: that of the $x = 0.1$ sample. c) The evolution of $T_{AM}-T_{M}$ with $T_{M}$. 
181: $\bullet$: Cu-doped Ru1212Eu; $\blacksquare$: Ru1212Eu of Butera \textit{et al.};\cite{but} 
182: $\bigtriangleup$: annealed Ru1222Gd; $\square$: as-synthesized Ru1222Eu.}
183: \end{figure*}
184: 
185: The $\chi = M/H$ of a simple AFM magnet, which will be $H$-independent far above its AFM 
186: transition, should have a maximum slightly above the N\'{e}el temperature, $T_{AM}$. It has 
187: been suggested, in fact, that the magnetic energy, $E_{m}$, and $\chi$ should both depend on 
188: the pair correlation functions $\Gamma(r) = 3[S^{z}(0)S^{z}(r)]/S(S+1)$ as 
189: $E_{m} \propto \Gamma_{1}$ and 
190: $\chi \propto [1+ \Sigma_{r} \Gamma(r)]/T \approx [1+f(T)\Gamma_{1}]/T$, where $\Gamma(r)$, 
191: $\Gamma_{1}$, and $f(T)$ are the pair-correlation with the pair distance $= r$, the 
192: correlation with the nearest-neighbor, and a slowly varying function of $T$, 
193: respectively.\cite{fis} This leads to an approximation of 
194: $C_{p} \propto \partial(T\chi)/\partial T$ if the short-range correlation $\Gamma_{1}$ 
195: is dominant. $T_{AM}$, therefore, can be defined as the temperature of the 
196: $\partial(T\chi)/\partial T$ peak,\cite{fis} which is observed close to the $\chi$-maximum 
197: temperature in 3D but much lower in 2D.\cite{jon} For CAFM magnets, an FM-like $M_{FC}$ step 
198: may coexist with a $\partial(T\chi)/\partial T$ peak. However, $T_{AM} \approx T_{M}$ is 
199: expected, except for the possible $H$-induced transition shifts.\cite{note} 
200: 
201: To analyze the magnetization of Ru$_{1-x}$Cu$_{x}$Sr$_{2}$EuCu$_{2}$O$_{8+\delta}$, the 
202: Eu/CuO$_{2}$ contributions were first eliminated using the procedure previously 
203: proposed,\cite{but} i.e. with a Van Vleck susceptibility of free Eu$^{3+}$ and a 
204: $T$-independent $\chi_{0}$ of $8.7 \times 10^{-4}$~emu/mole for CuO$_{2}$. For the undoped 
205: sample with $x = 0$, the Ru contribution is $H$-independent and follows a Curie-Weiss (C-W) 
206: fit only above 250~K with a C-W constant $\approx 2.6$~$\mu_{B}$/Ru and a Curie temperature of 
207: 127~K. Deviation from the C-W fit and large superparamagnetic $M(H)$, however, develop at 
208: lower temperatures (Fig.~\ref{fig2}a). The Ru contribution to $\chi$ at 1~T, for example, is 
209: more than 10\% higher than that expected between 180~K and $T_{AM}$ (Fig.~\ref{fig2}a), 
210: indicating a dominant FM interaction. The 5~T differential Ru-susceptibility after subtracting 
211: the Eu/CuO$_{2}$ contributions ($\chi_{Ru~only}$), however, shows an opposite downturn, 
212: suggesting significant AFM interactions (Fig.~\ref{fig2}a). In particular, a minimum of 
213: $1/\chi_{Ru~only}$ and a $\partial(T\chi)/\partial T$ peak appear around 157~K 
214: (Fig.~\ref{fig2}a) and $T_{AM} \approx 138$~K (Fig.~\ref{fig2}b), respectively, for the 
215: $x = 0$ sample. Undoped Ru1212Eu, therefore, might be interpreted as a simple CAFM by 
216: either ignoring the 4~K difference between $T_{M}$ and $T_{AM}$,\cite{but} or by regarding it 
217: as a small $H$-induced transition shift. 
218: 
219: To further confirm the presumed $T_{AM}$, the magnetic specific heat was measured at a zero 
220: field using a nonsuperconducting YBa$_{2}$(Cu$_{2.73}$Zn$_{0.27}$)O$_{7}$ (YBCO) ceramic as 
221: the reference (Fig.~\ref{fig2}b). The raw specific heat of Ru1212 is well above that of YBCO 
222: between 80 and 180~K, but the two merge outside this region, a situation similar to the data 
223: of Ru1212Gd.\cite{tal} The magnetic $C_{p}/T$, i.e. the difference between Ru1212R and YBCO, 
224: shows a well defined peak at 133~K, which is only slightly lower than the 138~K 
225: $\partial(T\chi)/\partial T$ peak observed. This agreement between the $C_{p}/T$ peak at 
226: zero field and the $\partial(T\chi)/\partial T$ peak at 5~T again demonstrates that the 
227: procedure of Fisher\cite{fis} works reasonably well and that the $H$-induced transition shift 
228: is small in our case. It is also interesting to note the high-$T$ tail of $C_{p}/T$ and 
229: the non-C-W magnetization up to 180~K or higher (Figs.~\ref{fig2}a,b). Significant short-range 
230: spin orders, therefore, should occur far above $T_{M}$ and $T_{AM}$.
231: 
232: With the Cu doping, however, the $T_{M}$ and the $T_{AM}$ evolve in different ways and the 
233: separation between them broadens. At $x = 0.1$, for example, the $T_{M}$ is quickly suppressed 
234: to 117~K but the $\partial(T\chi)/\partial T$ peak remains at 138~K (Figs.~\ref{fig2}a,b). 
235: The accompanying $C_{p}/T$ appears to broaden with $x$ as well (Fig.~\ref{fig2}b). In 
236: particular, the well defined peak evolves into a flat plateau between $T_{M}$ and $T_{AM}$ 
237: (Fig.~\ref{fig2}b). It should also be pointed out that the separation at $x = 0.1$ is larger 
238: than the transition width in $M_{FC}$. Neither the sample inhomogeneity nor the experimental 
239: resolution, therefore, can account for the separation (Figs.~\ref{fig2}a,b). The AFM-like 
240: $\partial(T\chi)/\partial T$ peak and the FM-like $M_{FC}$ jump seem to carry 
241: distinct spin entropies of comparable strength.
242: 
243: It is therefore interesting to compare the data with that of Ru1222R, where two separate 
244: transitions have been observed in both magnetizations and M\"{o}ssbauer 
245: spectra.\cite{fel,xue2} The $T_{M}$ and $T_{AM}$ of 
246: Ru$_{1-x}$Cu$_{x}$Sr$_{2}$EuCu$_{2}$O$_{8+\delta}$ samples with $0 \le x \le 0.15$, the 
247: O$_{2}$/Ar-annealed RuSr$_{2}$(Gd$_{1.4}$Ce$_{0.6}$)Cu$_{2}$O$_{10+\delta}$, and two 
248: as-synthesized Ru1222Eu samples are shown in Fig.~\ref{fig2}c.\cite{xue2} The separation 
249: $T_{AM}-T_{M}$ increases systematically with decreasing $T_{M}$ in the Cu-doped Ru1212Eu: from 
250: an extrapolated zero-separation at $T_{M} \approx 140$~K to 25~K at $T_{M} \approx 110$~K, 
251: where the data smoothly evolve into that of Ru1222R (Fig.~\ref{fig2}c). The observation of 
252: $T_{AM} = T_{M}$ in the Ru1212Eu sample,\cite{but} therefore, may be only a coincidence. 
253: Distinct AFM and FM transitions may coexist in both Ru1212R and Ru1222R. 
254: 
255: These two transitions, as has been argued in the case of Ru1222R,\cite{fel,xue2} may be due to 
256: either a mesoscopic phase-separation or a multistage transition. The magnetic properties 
257: between $T_{M}$ and $T_{AM}$, however, will be different in these two scenarios: some parts 
258: of Ru1212R 
259: should be in superparamagnetic states during phase separation, but should stay in a long-range 
260: spin-order state during a multistage transition. Evidence for the possible phase-separation in 
261: Ru1222R, for example, is found in both the superparamagnetic $M(H)$ with a magnetic 
262: cluster-size of $10^3$~$\mu$B and the slow spin dynamics far above $T_{FM}$.\cite{xue2} 
263: Similar properties were therefore tested in the Cu-doped Ru1212Eu.
264: 
265: The Langevin function with an additional linear term, 
266: $a \cdot H+m \cdot [c\tanh(\mu H/k_{B}T)-k_{B}T/\mu H]$, was used to fit the average 
267: magnetization in a $M$-$H$ loop (inset, Fig.~\ref{fig3}a).\cite{xue2} The fit is reasonably 
268: good with the deduced $\mu$ between 100 and 700~$\mu_{B}$/cluster (closed symbols in 
269: Fig.~\ref{fig3}a), which is 4--5 times smaller than those deduced in Ru1222R, but still far 
270: larger than that expected based on the spin-fluctuations. A cluster of 
271: 400~$\mu_{B} \approx 200$~Ru ions, for example, would be 4--5~nm or larger in an RuO layer. It 
272: should be further noted that the $\mu$ so-deduced may be only a lower limit of the actual 
273: cluster/spin-correlation length.\cite{all} The existence of such large clusters at 
274: $T/T_{M} > 1.1$ will be difficult to be interpreted as a simple fluctuation. This deduced 
275: size, on the other hand, appears to be too small for a 
276: crystalline magnet, as is suggested in the multistage transition model. 
277: 
278: \begin{figure*}
279: \includegraphics[scale=.5]{fig3.eps}% Here is how to import EPS art
280: \caption{\label{fig3}a) The cluster sizes for samples with $\bullet$: $x = 0$; 
281: $\blacktriangledown$: $x = 0.05$; $\blacksquare$: x=0.10; and $\blacklozenge$: x=0.15 and for 
282: $\circ$: as-synthesized Ru1222Eu ($\times 4$). Inset: The isothermal $M(H)$ of the $x = 0$ sample. 
283: b) The relaxation of the remnant magnetizations at 160, 150, 140, and 130~K (from top to 
284: bottom) after field-cooling at 50~Oe.}
285: \end{figure*}
286: 
287: The dynamic spin-response was also studied. The logarithmic increase of $M_{ZFC}$ at 5~Oe with 
288: time is almost unobservable, with the deduced rate of $dlnM/dlnt < 10^{-3}$ well within our 
289: experimental resolution, where 60~s~$< t <$ 3600~s is the time after the field switch. This is 
290: rather different from that of Ru1222Eu,\cite{xue2} but in agreement with the unobservable 
291: relaxation of Ru1212R $ac$-susceptibility reported between 1~s and 100~s.\cite{ziv} The lack 
292: of relaxations under the above conditions is apparently related to the 
293: cluster size in Ru1212R (Fig.~\ref{fig3}a), which is 4--5 times smaller and leads to a quicker 
294: equilibrium. The slow spin dynamics, therefore, should either be explored in a shorter time 
295: window or after an enhancement of the energy barriers. Several different experimental 
296: conditions were then tested, and significant nonlogarithmic relaxations were observed in the 
297: remnant magnetization after a 50~Oe field-cooling (Fig.~\ref{fig3}b). It is interesting to 
298: note that the energy barriers are $\approx KV_{c}-\mu H$ and $KV_{c}$, respectively, for the 
299: $M_{ZFC}$ and the remnant magnetization, where $K$ and $V_{c}$ are the magnetic anisotropy 
300: and the coherent volume, respectively. This may make the remnant magnetization a more 
301: favorable candidate for investigating the slow dynamics. The strong $T$ dependence of the 
302: relaxation observed (Fig.~\ref{fig3}b) suggests, in our opinion, that the relaxation observed 
303: is unlikely an artifact of the SQUID magnetometer, but supports the existence of 
304: superparamagnetic clusters.
305: 
306: As pointed out earlier, the phase-separation model may also offer a consistent interpretation 
307: for the conflicting NPD/NMR and superconductivity data reported previously.\cite{xue2} The 
308: conflict between the NPD and NMR data for the magnetic structure, for example, may be 
309: attributed to the fact that the two probes have different sensitivities to various magnetic 
310: species, like those well documented in manganites.\cite{kap} Similarly, the spatial 
311: separation between AFM and FM species offers a natural mechanism for the unusual 
312: superconductivity observed.\cite{xue} Superconductivity can coexist with the AFM matrix. The 
313: finely dispersed FM clusters, on the other hand, depress the local SC order parameter and 
314: serve as tunnel barriers for the Cooper pairs. The superconductivity, therefore, may retain a 
315: significant part of the condensation energy, but appears only as a Josephson-junction array. 
316: Similarly, the critical temperature observed in the transport will naturally be much lower 
317: than 
318: that associated with the corresponding $C_{p}$ anomaly,\cite{ber,tal} and can be easily 
319: suppressed by external fields.\cite{lor} The intragrain penetration depth will also be much 
320: longer than those expected based on the proposed universal $1/\lambda_{2}(T_{c})$.\cite{xue}
321: 
322: In summary, a systematic separation between $T_{M}$ and $T_{AM}$ is observed in Ru1212Eu with 
323: Cu-doping, suggesting the coexistence of FM and AFM orders and the occurrence of a 
324: mesoscopic phase-separation in the compound. The superparamagnetic $M(H)$ as well as the slow 
325: spin-dynamics further support the interpretation.
326: 
327: \begin{acknowledgments}
328: The work in Houston is supported in part by 
329: NSF Grant No. DMR-9804325, the T.~L.~L. Temple Foundation, the John J. and Rebecca 
330: Moores Endowment, and the State of Texas through the Texas Center for 
331: Superconductivity at the University of Houston; and at Lawrence Berkeley 
332: Laboratory by the Director, Office of Science, Office of Basic Energy Sciences, 
333: Division of Materials Sciences and Engineering of the U.S. Department of Energy 
334: under Contract No. DE-AC03-76SF00098.
335: \end{acknowledgments}
336: 
337: \begin{thebibliography}{99}
338: \bibitem{fel} I. Felner, U. Asaf, Y. Levi and O. Millo, Phys. Rev. B {\bf 55}, 3374 (1997); 
339: I. Felner, U. Asaf and E. Galstyan, cond-mat/0111217 (2001).
340: 
341: \bibitem{ber} C. Bernhard, J. L. Tallon, E. Br\"{u}cher and R. K. Kremer, Phys. Rev. B 
342: {\bf 61}, 14960 (2000).
343: 
344: \bibitem{tal} J. L. Tallon, J. W. Loram, G. V. M. Williams and C. Bernhard, Phys. Rev. B 
345: {\bf 61}, 6471 (2000).
346: 
347: \bibitem{xue} Y. Y. Xue, B. Lorenz, R. L. Meng, A. Baikalov and C. W. Chu, Physica C 
348: {\bf 364-365}, 251 (2001); Y. Y. Xue, B. Lorenz, A. Baikalov, D. H. Cao, Z. G. Li and 
349: C. W. Chu, Phys. Rev. B {\bf 66}, 014503 (2002).
350: 
351: \bibitem{pic} W. E. Pickett, R. Weht and A. B. Shick, Phys. Rev. Lett. {\bf 83}, 3713 (1999).
352: 
353: \bibitem{xue2} Y. Y. Xue, D. H. Cao, B. Lorenz and C. W. Chu, Phys. Rev. B {\bf 65}, 020511 
354: (2001); Y. Y. Xue, B. Lorenz, D. H. Cao and C. W. Chu, cond-mat/0211342 (2002).
355: 
356: \bibitem{fis} M. E. Fisher, Philos. Mag. {\bf 7}, 1731 (1962).
357: 
358: \bibitem{but} A. Butera, A. Fainstein, E. Winkler and J. Tallon, Phys. Rev. B {\bf 63}, 054442 
359: (2001).
360: 
361: \bibitem{lyn} J. W. Lynn, B. Keimer, C. Ulrich, C. Bernhard and J. L. Tallon, Phys. Rev. B 
362: {\bf 61}, 14964 (2000).
363: 
364: \bibitem{tok} Y. Tokunaga, H. Kotegawa, K. Ishida, Y. Kitaoka, H. Takagiwa and J. Akimitsu, 
365: Phys. Rev. Lett. {\bf 86}, 5767 (2001).
366: 
367: \bibitem{kla} P. W. Klamut, B. Dabrowski, S. Kolesnik, M. Maxwell and J. Mais, Phys. Rev. B 
368: {\bf 63}, 224512 (2001).
369: 
370: \bibitem{lor} B. Lorenz, Y. Y. Xue, R. L. Meng and C. W. Chu, Phys. Rev. B {\bf 65}, 174503 
371: (2002).
372: 
373: \bibitem{jon} L. J. de Jongh and A. R. Miedema, Adv. Phys. {\bf 23}, 1 (1974).
374: 
375: \bibitem{note} This is true below a critical field $H_{c}$ since $\partial(T\chi)/\partial T$ 
376: and $\partial M_{FC}/\partial T$ have similar $T$-dependences. Above $H_{c}$, however, the 
377: field may break the AFM spin-correlations and a magnetization jump appears. A. Herweijer 
378: \textit{et al.} reported in Phys. Rev. B {\bf 5}, 4618 (1972), for example, that the 
379: magnetizations of CsCoCl$_{3}$~$\cdot$~2H$_{2}$O (a 1D CAFM) are linear in $H$ below 0.3~T 
380: with identical $T_{AM}$ and $T_{M}$.
381: 
382: \bibitem{all} P. Allia, M. Coisson, P. Tiberto, F. Vinai, M. Knobel, M. A. Novak and 
383: W. C. Nunes, Phys. Rev. B {\bf 64}, 144420 (2001), and references therein.
384: 
385: 
386: \bibitem{ziv} I. \v{Z}ivkovi\'{c}, Y. Hirai, B. H. Frazer, M. Prester, D. Drobac, D. Ariosa, 
387: H. Berger, D. Pavuna, G. Margaritondo, I. Felner and M. Onellion, Phys. Rev. B {\bf 65}, 
388: 144420 (2002).
389: 
390: 
391: \bibitem{kap} Cz. Kapusta, P. C. Riedi, M. Sikora and M. R. Ibarra, Phys. Rev. Lett. {\bf 84}, 
392: 4216 (2000).
393: 
394: \end{thebibliography}
395: 
396: \end{document}
397: %
398: % ****** End of file apssamp.tex ******
399: