1: \documentclass[twocolumn,showpacs,preprintnumbers,a4,prb,superscriptaddress]{revtex4}
2: \usepackage{graphicx}%
3: \usepackage{dcolumn}
4: \usepackage{amsmath}
5: \usepackage{amssymb}
6: \usepackage{bm}
7:
8:
9: \begin{document}
10: \title{Crystal and magnetic structure of La$_{1-x}$Ca$_{x}$MnO$_{3}$ compound $(0.11\leq x\leq 0.175)$}
11: \author{M. Pissas\cite{cor}}
12: \affiliation{Institute of Materials Science, National Center for
13: Scientific Research, Demokritos, 15310 Ag. Paraskevi, Athens,
14: Greece}
15: \author{I. Margiolaki}
16: \affiliation{European Synchrotron Radiation Facility (ESRF), Pluo
17: 30.1.08 6 Rue J. Horowitz, 38043 Grenoble, Cedex 9, France}.
18: \author{G. Papavassiliou}
19: \author{D. Stamopoulos}
20: \affiliation{Institute of Materials Science, National Center for
21: Scientific Research, Demokritos, 15310 Ag. Paraskevi, Athens,
22: Greece}
23: \author{D. Argyriou}
24: \affiliation{Berlin Neutron Scattering Center,
25: Hahn-Meitner-Institut, Glienicker Str. 100 D-14109, Berlin,
26: Germany}
27:
28: \date{\today }
29:
30:
31: \begin{abstract}
32: We studied the crystal and magnetic structure of the
33: La$_{1-x}$Ca$_{x}$MnO$_{3}$ compound for $(0.11\leq x\leq 0.175)$
34: using stoichiometric samples. For $x<0.13$ the system's ground
35: state is insulating canted antiferromagnetic. For $0.13\le x \le
36: 0.175$ e below the Jahn Teller transition temperature ($T_{\rm
37: JT}$) the crystal structure undergoes a monoclinic distortion. The
38: crystal structure can be described with $P 2_1/c$ space group
39: which permits two Mn sites. The unit cell strain parameter
40: $s=2(a-c)/(a+c)$ increases for $T<T_{\rm JT}$, taking the maximum
41: value at the Curie point, and then decreases. Below $T_{\rm
42: M^/M^{//}}\approx 60 $ K $s$ abruptly changes slope and finally
43: approaches $T=0$ K with nearly zero slope. The change of $s$ at
44: $T_{\rm M^/M^{//}}$ is connected to a characteristic feature in
45: the magnetic measurements. As $x$ increases towards the
46: ferromagnetic metallic boundary, although $s$ is reduced
47: appreciably, the monoclinic structure is preserved. The monoclinic
48: structure is discussed with relation to the orbital ordering,
49: which can produce the ferromagnetic insulating ground state. We
50: also studied samples that were prepared in air atmosphere. This
51: category of samples shows ferromagnetic insulating behavior
52: without following the particular variation of the $s$ parameter.
53: The crystal structure of these samples is related to the so-called
54: O$^{*}$ ($c>a>b/\sqrt{2}$) structure.
55: \end{abstract}
56:
57: \pacs{75.47.Lx,75.47.GK,75.25.+z,75.30Cr} \maketitle
58:
59:
60: \section{Introduction} The nature of the ferromagnetic insulating
61: state (FMI) in lightly doped La$_{1-x}$A$_x$MnO$_3$ (A=Sr, Ca,
62: $x=0.1-0.2$) rare-earth manganite perovskites continues to offer
63: significant challenges in our understanding of the physical
64: properties of these materials. Typically, insulating compounds of
65: this category are antiferromagnetic, and ferromagnetism is
66: associated with a metallic behavior that arises from doping and
67: double exchange. The FMI state can be realized with a certain
68: orbital ordering favorable for ferromagnetism.\cite{khomskii97} In
69: the intermediate doping levels ($x=0.23-0.4$) where the
70: ferromagnetic metallic (FMM) state is observed the double-exchange
71: mechanism is able to explain some of the experimental results.
72: Despite of the numerous studies concerning the FMI phase of
73: La$_{1-x}$Sr$_x$MnO$_3$ compound, for the Ca case only a few
74: exist. Furthermore, the tendency of this system in particular to
75: produce cation deficient samples complicates the situation. For
76: the Sr case the $x=0.125$ sample shows a transition to a FMM state
77: from paramagnetic insulating state, followed by an abrupt
78: first-order transition to the FMI phase. The FMI state coexists
79: with a weak antiferromagnetic component with the $A-$type magnetic
80: structure as was found in LaMnO$_3$.\cite{pinsard97} In addition
81: in some other studies\cite{yamada00,endoh99} charge or orbital
82: ordering have been proposed to occur in the FMI state.
83:
84: \begin{figure}[htbp] \centering%
85: \includegraphics[angle=0,width=1.0 \columnwidth]{fig1.eps}
86: \caption{Ferromagnetic insulating part of the phase diagram of
87: La$_{1-x}$Ca$_{x}$MnO$_{3}$ compound. The light yellow region
88: corresponds to the CAF regime, and the light blue to the FMI
89: regime. The symbols O*, O$^/$, O$^{//}$, M$^/$ and M$^{//}$ have
90: the following meaning: [O* orthorhombic $Pnma$ symmetry with
91: $c\geq a\geq b/\sqrt{2}$],[O$^/$,(O$^{//}$) orthorhombic $Pnma$
92: symmetry with $a>c>b/\sqrt{2}$ and strong (intermediate)
93: anisotropic bond lengths] and [M$^/$, (M$^{//}$) monoclinic $P
94: 2_1/c$ symmetry with $a>c>b/\sqrt{2}$ and strong (intermediate)
95: anisotropic bond lengths]. The $T_{\rm M^/M^{//}}$ is defined at
96: the temperature where the slope of the $a(T)$ and $b(T)$ curves
97: change slope abruptly upon heating from $T=0$. The $T_{\rm JT}$
98: curve is defined at the temperature where the $m(T)$ curves
99: display a jump-like or slope-change behavior. The $T_{\rm C}$ and
100: $T_{\rm N}$ are both estimated from neutron and magnetization
101: data.}
102: \label{diagram}%
103: \end{figure}%
104:
105: \begin{figure}[htbp] \centering%
106: \includegraphics[angle=0,width=0.7 \columnwidth]{fig2.eps}
107: \caption{ (upper panel) Temperature variation of the lattice
108: parameters and unit-cell volume for
109: La$_{0.87}$Ca$_{0.13}$MnO$_{3}$ compound. (lower panel)
110: Temperature variation of the ordered ferromagnetic and
111: antiferromagnetic moments. The ferromagnetic components are
112: denoted by $F$ and the antiferromagnetic with $A$. The inset shows
113: the temperature variation of the spontaneous orthorhombic strain
114: $(s=2(a-c)/a+c))$. Lines through the experimental points are
115: guides to the eye.}
116: \label{abc-r013}%
117: \end{figure}%
118:
119: \begin{figure}[tbp] \centering%
120: \includegraphics[angle=0,width=0.7 \columnwidth]{fig3.eps}
121: \caption{ (upper panel) Temperature variation of the lattice
122: parameters and unit-cell volume for
123: La$_{0.85}$Ca$_{0.15}$MnO$_{3}$ compound. Lines through the
124: experimental points are guides to the eye. (lower panel)
125: Temperature variation of the ordered ferromagnetic moment. The
126: solid line is the calculated ordered moment supposing a power law.
127: The inset shows the temperature variation of the spontaneous
128: orthorhombic strain $(s=2(a-c)/a+c))$. }
129: \label{abc-r015}%
130: \end{figure}%
131:
132: \begin{figure}[tbp] \centering%
133: \includegraphics[angle=0,width=0.7 \columnwidth]{fig4.eps}
134: \caption{ (upper panel) Temperature variation of the lattice
135: parameters and unit-cell volume for
136: La$_{0.825}$Ca$_{0.175}$MnO$_{3}$ compound. Lines through the
137: experimental points are guides to the eye. (lower panel)
138: Temperature variation of the ordered ferromagnetic moment. The
139: solid line is the calculated ordered moment supposing a power law.
140: The inset shows the temperature variation of the spontaneous
141: orthorhombic strain $(s=2(a-c)/a+c))$. }
142: \label{abc-r0175}%
143: \end{figure}%
144:
145: The manganite perovskites of interest to this work consist of a 3D
146: network of corner-sharing BX$_6$ octahedra located at the nodes of
147: a simple cubic lattice. At the center of the unit cell there is
148: room for the A cation which will fit perfectly if $d_{\rm
149: A-X}=d_{\rm B-X}\sqrt{2}$. When $d_{\rm A-X}<d_{\rm B-X}\sqrt{2}$
150: the BX$_6$ octahedra are tilted to fill the extra interstitial
151: space (steric effect). The steric effect in
152: La$_{1-x}$Ca$_x$MnO$_3$ compound leads to a lower symmetry
153: structure with space group $Pnma$ (in comparison to the aristotype
154: cubic perovskite $Pm\overline{3}m$ structure) which according to
155: the Glazer's clasification\cite{glazer} belongs to the $\alpha^-
156: \beta^+\alpha^-$ tilt system (adopted by GdFeO$_3$ and CaTiO$_3$
157: perovskites). The adoption of the GdFeO$_3$ tilting scheme in
158: La$_{1-x}$Ca$_x$MnO$_3$ is due to the small radius of the La and
159: Ca, with respect to its surrounding cage. Thus, in order that the
160: La ion can be accommodated in the structure, the MnO$_6$-octahedra
161: tilt. Practically MnO$_6$ octahedra rotation is not rigid. Below a
162: characteristic temperature the La$_{1-x}$Ca$_x$MnO$_3$ compound
163: for $x<0.23$ displays cooperative JT distortion arising from
164: ordering of the $e_g$ orbital of the Mn$^{+3}$ ion. This structure
165: is denoted as\cite{monoclinic} O$^{/}$ and the unit cell satisfies
166: the inequality $a>c>b/\sqrt{2}$. For this phase, Mn$^{+3}$ $e_g$
167: orbitals are ordered in such a way so that in neighboring Mn sites
168: the $|3x^2-r^2\rangle$ and $|3z^2-r^2\rangle$ orbitals are
169: alternatively occupied, forming a "canted" orbital structure. This
170: configuration does not change along $b-$axis.
171:
172: Two mechanisms have been proposed to explain the orbital ordering
173: in degenerated transition metal orbital states. The first is
174: related with lattice distortions\cite{kanamori60} and the second
175: is a direct generalization of the usual superexchange which in the
176: case of orbital degeneracy is present.\cite{kugel82} The orbital
177: ordering can be described by using the pseudospin-$1/2$, $T_i$
178: operator in such a way so that the state $| T^z=\pm 1/2\rangle$
179: corresponds to the orbitals $|3z^2-r^2\rangle$ and
180: $|x^2-y^2\rangle$, respectively. An arbitrary distortion can be
181: described by linear superposition of the states $| T^z=\pm
182: 1/2\rangle$, e.g. $|\theta\rangle=\cos(\theta/2)|
183: 1/2\rangle+\sin(\theta/2)|-1/2\rangle$, where $\theta$ is an angle
184: in the $(T^z, T^x)$ plane.
185:
186: Finally, there is a third orthorhombic phase, O*
187: ($c>a>b\sqrt{2}$), related to orbital disordered manganites
188: ($x>0.23$). The O*-structure, is pseudocubic with the Mn-O bond
189: lengths almost equal, indicating that it lacks a clear static JT
190: distortion. Both phases exhibit the orthorhombic $Pnma$ symmetry.
191:
192: In this article we present a synchrotron x-ray and neutron
193: diffraction study, using carefully prepared stoichiometric
194: La$_{1-x}$Ca$_x$MnO$_3$ $(0.11\leq x\leq 0.175)$ samples in an
195: effort to isolate the structural and magnetic features that
196: characterize the FMI state. Our diffraction results on
197: stoichiometric samples shows that the crystal structure at FMI
198: regime is monoclinic $P 2_1/c$. The temperature variation of the
199: unit cell parameters is non monotonic, defining the $T_{\rm JT}$
200: and $T_{\rm M^/M^{//}}$ temperatures. In addition, this monoclinic
201: crystal structure permits two Mn crystallographic sites. This
202: crystal structure is different from it of the cation deficient
203: samples and samples that belong in the ferromagnetic metallic
204: regime.
205:
206:
207: \section{Experimental details}
208:
209: La$_{1-x}$Ca$_{x}$MnO$_{3}$ samples ware prepared by thoroughly
210: mixing high purity stoichiometric amounts of CaCO$_{3}$,
211: La$_{2}$O$_{3}$, and MnO$_{2}$. The mixed powders reacted in air
212: up to 1400$^{\text{o}}$C for several days
213: with intermediate grinding. Finally the samples were slowly cooled to 25$^{%
214: {\rm o}}$C. These samples are cation
215: deficient.\cite{pissas04a,dabrowski99a,roosmalen94} In order to
216: produce stoichiometric samples additional annealing in He
217: atmosphere was applied. More specifically, in the final
218: preparation step the samples were heated under He flow up to
219: 1100$^{\rm o}$C for 48 h followed by rapid cooling to room
220: temperature.
221: %*****************************************************************
222: It is important to stress here that in the doping range examined
223: in this work the samples fired in air in all preparation steps are
224: not stoichiometric.
225: %********************************************************
226: Although these samples are ferromagnetic and insulating for
227: $x<0.2$, they do not exhibit all the features of the
228: stoichiometric samples, especially the cooperative Jahn-Teller
229: (JT) distortion feature for $T<T_{\rm JT}$.
230:
231: Neutron powder diffraction (NPD) data were collected on the E6 and
232: E9 neutron diffractometers at the Berlin Neutron Scattering Center
233: (BENSC), located at the Hahn-Meitner-Institut, Berlin. The neutron
234: powder diffraction measurements on the E6 diffractometer were
235: performed as a function of temperature at low angles using (002)
236: reflection of pyrolytic graphite monochromator, giving a
237: wavelength $\lambda =2.4$\AA . For high resolution crystal
238: structure determinations we measured NPD data using the E9
239: diffractometer with a wavelength of $\lambda =1.798$\AA\ ((511)
240: reflection of a vertically bent Ge monochromator) and collimation
241: $\alpha _{1}=18^{/}$(in pile collimator), $\alpha
242: _{2}=20^{/}$(second collimator after monochromator), $\alpha
243: _{3}=10^{/}$(third collimator after sample). The powdered samples
244: were contained in a cylindrical vanadium can ($D=8$ mm) mounted in
245: an ILL orange cryostat.
246:
247: For high resolution x-ray diffraction data we used the ID31 high
248: resolution powder diffraction beamline of the European Synchrotron
249: Radiation Facility (ESRF). For our measurements we select a wave
250: length $\lambda=0.50019$\AA. Nuclear and magnetic structures were
251: determined and refined from the NPD data by using the Rietveld
252: method with the {\sc FULLPROF} suite of programs.\cite{fullprof}
253: For synchrotron x-ray diffraction data, the GSAS suite was
254: used.\cite{toby,larson} As peak shape function we used the
255: gaussians convoluted with axial divergence asymmetry function\cite
256: {finger94} for neutron data and the constant wave length profile
257: function no 4 of the GSAS for the x-ray data. DC magnetization
258: measurements were performed in a superconducting quantum
259: interference device (SQUID) magnetometer (Quantum Design).
260: Finally, for the ac susceptibility measurements we used a home
261: made ac-susceptometer.
262:
263: \begin{table*}[tbp] \centering%
264: \caption{La$_{1-x}$Ca$_{x}$MnO$_{3}$ ($x=0.125$) structural
265: parameters at several temperatures, as determined from Rietveld
266: refinements based on neutron powder diffraction data of E9
267: diffractometer ($\lambda=1.79784$\AA ). The monoclinic space group
268: $P 2_1/c\equiv P\ 1\ 1\ 2_1/a$ (no 14, unique $c$-axis) was used.
269: In this space group La, Ca O1, O2A and O2B occupy the general $4e$
270: position $(x,yz)$. Mn1 and Mn2 occupy the $2c$ (1/2,0,0) and $2d$
271: (1/2,1/2,0) positions respectively. Numbers in parentheses are
272: statistical errors of the last significant digit.}
273: \begin{ruledtabular}
274: \begin{tabular}{ccccccc}
275: $T$ (K) & & 300 & 198.4 & 121 & 48 & 2 \\
276: \tableline
277: $a$ (\AA ) & & 5.5754(1) & 5.5855(1) & 5.5829(1) & 5.5709(1)&5.5697(1)\\
278: $b$ (\AA ) & & 7.7423(1) & 7.7213(1) & 7.7186(1) & 7.7288(1)&7.7296(1)\\
279: $c$ (\AA ) & & 5.5269(1) & 5.5253(1) & 5.5234(1) & 5.5206(1)&5.5205(1)\\
280: $\gamma$ (deg.)& & 90.125(3) & 90.130(5) & 90.127(4) & 90.106(6)&90.107(5)\\
281: \tableline
282: La,Ca &$x$ & 0.0319(4) & 0.0344(3) & 0.0348(6)& 0.0342(3)&0.0341(3) \\
283: &$y$ & 0.2501(8) & 0.2487(5) & 0.2477(7)& 0.2479(9)&0.2485(8)\\
284: &$z$ & 0.0058(5) & 0.0049(4) & 0.0055(4)& 0.0064(4)& 0.0061(4) \\
285: &$B$ & 0.69(4) & 0.57(3) & 0.43(3) & 0.33(3) &0.37(3)\\
286: \tableline
287: %--------------------------------------------------------------
288: Mn1 &$B$ & 0.38(7) & 0.30(6) & 0.24(6) & 0.19(5) &0.17(6) \\
289: Mn2 &$B$ & 0.38(7) & 0.30(6) & 0.24(6) & 0.19(5) &0.17(6) \\
290: %--------------------------------------------------------------
291: \tableline
292: O1 &$x$ & 0.4909(6) & 0.4898(5) & 0.4906(5)& 0.4909(5)&0.4901(5)\\
293: &$y$ & 0.244(1) & 0.247(1) & 0.246(1) & 0.248(1)&0.247(1) \\
294: &$z$ & -0.0670(6) &-0.0679(4) &-0.0688(5)&-0.0687(4)&-0.0683(5)\\
295: &$B$ & 1.01(6) & 0.87(5) & 0.74(5) & 0.60(4) &0.62(5)\\
296: %--------------------------------------------------------------
297: \tableline
298: O2A&$x$ & 0.213(1) & 0.211(1) & 0.209(1)& 0.215(1)&0.223(1)\\
299: &$y$ &-0.0380(9) &-0.0377(8) &-0.0392(7)&-0.0376(8)&-0.0374(9)\\
300: &$z$ & 0.784(1) & 0.784(1) & 0.784(1)& 0.788(1)&0.783(1) \\
301: &$B$ & 1.06(5) & 0.83(4) & 0.66(4) & 0.76(4) &0.73(4)\\
302: \tableline
303: %--------------------------------------------------------------
304: O2B&$x$ &-0.207(1) &-0.207(1) &-0.208(1)&-0.205(1)&-0.207(1)\\
305: &$y$ & 0.4656(9) & 0.4640(7) & 0.4655(7)& 0.4644(8)&0.4641(8)\\
306: &$z$ &-0.764(1) &-0.763(1) &-0.764(1)&-0.778(1)&-0.766(1)\\
307: &$B$ & 1.06(5) & 0.83(3) & 0.66(4) & 0.76(4) &0.73(4)\\
308: %--------------------------------------------------------------
309: %\tableline
310: %&$S_{400}$ & 0.40(6) &1.02(6) & 1.14(6) &1.06(5)& 0.40(4)\\
311: %&$S_{040}$ & 0.19(1) &0.23(1) & 0.23(1) &0.20(1)& 0.07(1)\\
312: %&$S_{004}$ & 0.05(4) &0.10(3) & 0.30(4) &0.15(3)& 0.02(3)\\
313: %&$S_{220}$ & 0.56(1) &0.61(1) & 0.61(1) &0.58(1)& 0.44(4)\\
314: %&$S_{202}$ & 0.8(1) &3.0(2) & 3.4(2) &3.0(1) & 2.1(1)\\
315: %&$S_{022}$ &-0.17(8) &0.05(7) &-0.16(7) &0.04(6)& 0.44(4)\\
316: \tableline
317: &$R_{p}$ & 7.0 & 5.9 & 6.4 & 5.7&5.9\\
318: &$R_{wp}$ & 9.2 & 7.6 & 8.2 & 7.5&7.8 \\
319: &$R_{B}$ & 6.2 & 5.3 & 5.6 & 4.7&4.8\\
320: %\tableline
321: %--------------------------------------------------------------
322: %&$F _{x}$ &0 & 0 & 0.57(9) &3.48(7)&3.48(7)\\
323: %&$F _{y}$ &0 & 0 & 0.04(4) &0.52(4)&0.52(4) \\
324: %&$F _{z}$ &0 & 0 & 0.0 &0.0 &0.0\\
325: %&$F $ &0 & 0 & 0.57(9) &3.52(3)&3.52(1)\\
326: %&$R_{B}(F)$ &0 & 0 & 4.5 &2.0 &2.0 \\
327: %\tableline
328: %--------------------------------------------------------------
329: \tableline
330: Mn1-O1 $\times 2$& &1.928(9) &1.947(8) & 1.942(8) &1.95(1)& 1.948(1)\\
331: Mn1-O2A$\times 2$& &2.013(7) & 2.026(6) & 2.035(6) &2.022(8)& 2.014(7)\\
332: Mn1-O2A$\times 2$& &1.994(8) & 1.986(6) & 1.981(7) &1.971(9)& 1.987(7) \\
333: Mn2-O1 $\times 2$& &2.014(9) & 1.986(8)& 1.993(8) &1.98(1)& 1.990(1)\\
334: Mn2-O2B$\times 2$& &2.102(7) & 2.111(6) & 2.099(6) &2.082(8)& 2.095(7)\\
335: Mn2-O2B$\times 2$& &1.884(8) & 1.881(6) & 1.887(6) &1.911(9)& 1.891(7) \\
336: \end{tabular}
337: \end{ruledtabular}\label{table1}
338: \end{table*}%
339:
340:
341:
342: \begin{table*}[tbp] \centering%
343: \caption{La$_{1-x}$Ca$_{x}$MnO$_{3}$ ($x=0.15$) structural
344: parameters at several temperatures, as determined from Rietveld
345: refinements based on neutron powder diffraction data of E9
346: diffractometer ($\lambda=1.79784$\AA ). At $T=320$ K the space
347: group $Pnma$ (No 62) was used. La, Ca and apical oxygen (O1)
348: occupy the $4c$, $(x,1/4,z)$ site, Mn the $4b$ $(0,0,1/2)$ site
349: and the plane oxygen (O2) the general $8d$ site. For other
350: temperatures the monoclinic space group $P 2_1/c\equiv P\ 1\ 1\
351: 2_1/a$ (no 14, unique $c$-axis) was used. In this space group La,
352: Ca O1, O2A and O2B occupy the general $4e$ position $(x,yz)$. Mn1
353: and Mn2 occupy the $2c$ (1/2,0,0) and $2d$ (1/2,1/2,0) positions
354: respectively. Numbers in parentheses are statistical errors of the
355: last significant digit.}
356: \begin{ruledtabular}
357: \begin{tabular}{ccccccc}
358: $T$ (K) & & 320 &250 & 170 & 100 & 2 \\
359: \tableline
360: $a$ (\AA )& & 5.5246(1) & 5.5573(1) & 5.5614(1)&5.5480(1)&5.5300 \\
361: $b$ (\AA )& & 7.7877(1) & 7.7387(1) & 7.7266(1)&7.7386(1)&7.7508 \\
362: $c$ (\AA )& & 5.5219(1) & 5.5213(1) & 5.5195(1)&5.5159(1)&5.5081 \\
363: $\gamma$ (deg.)& & 90 & 90.140(1) & 90.134(2)&90.0965(1)&90.0752 \\
364: \tableline
365: %--------------------------------------------------------------
366: La,Ca &$x$ & 0.0263(4) & 0.0300(3) & 0.0311(3)& 0.0309(3)&0.0303(2) \\
367: &$y$ & 0.25 & 0.2489(5) & 0.2501(7)& 0.2493(6)&0.2454(6)\\
368: &$z$ & -0.0083(5) & 0.0060(4) & 0.0060(4)& 0.0058(4)&-0.0065(4) \\
369: &$B$ & 0.54(5) & 0.40(3) & 0.42(3) & 0.12(2) &0.15(3)\\
370: \tableline
371: %--------------------------------------------------------------
372: Mn1 &$B$ & 0.36(7) & 0.25(5) & 0.19(5) & 0.19(5) &0.16(5) \\
373: Mn2 &$B$ & & 0.25(5) & 0.19(5) & 0.19(5) &0.16(5) \\
374: \tableline
375: %--------------------------------------------------------------
376: O1 &$x$ & 0.4910(8) & 0.4912(4) & 0.4914(4)& 0.4905(5)&0.4909(4)\\
377: &$y$ & 0.25 & 0.2450(8) & 0.246(1) & 0.2429(8)&0.259(1) \\
378: &$z$ & 0.0684(7) &-0.0668(4) &-0.0675(1)&-0.0693(4)&-0.0680(5)\\
379: &$B$ & 0.94(7) & 0.69(4) & 0.58(4) & 0.45(4) &0.484(4)\\
380: \tableline
381: %--------------------------------------------------------------
382: O2A&$x$ & 0.2817(6) & 0.2180(7) & 0.2188(8)& 0.2212(9)&0.2219(9)\\
383: &$y$ & 0.0353(3) &-0.0381(5) &-0.0364(7)&-0.0375(6)&-0.0384(6)\\
384: &$z$ & 0.7238(6) & 0.721(1) & 0.780(1)& 0.780(1)&0.782(1) \\
385: &$B$ & 0.88(5) & 0.63(3) & 0.58(3) & 0.40(3) &0.32(3)\\
386: \tableline
387: %--------------------------------------------------------------
388: O2B&$x$ & &-0.2075(8) &-0.2059(9)&-0.206(1)&-0.2085(0)\\
389: &$y$ & & 0.4659(6) & 0.4639(7)& 0.4657(7)&0.4666(6)\\
390: &$z$ & &-0.766(1) &-0.768(1)&-0.768(1)&-0.768(1)\\
391: &$B$ & & 0.63(3) & 0.58(3) & 0.40(3) &0.32(3)\\
392: %--------------------------------------------------------------
393: %\tableline
394: %&$S_{400}$ & 0.40(6) &1.02(6) & 1.14(6) &1.06(5)& 0.40(4)\\
395: %&$S_{040}$ & 0.19(1) &0.23(1) & 0.23(1) &0.20(1)& 0.07(1)\\
396: %&$S_{004}$ & 0.05(4) &0.10(3) & 0.30(4) &0.15(3)& 0.02(3)\\
397: %&$S_{220}$ & 0.56(1) &0.61(1) & 0.61(1) &0.58(1)& 0.44(4)\\
398: %&$S_{202}$ & 0.8(1) &3.0(2) & 3.4(2) &3.0(1) & 2.1(1)\\
399: %&$S_{022}$ &-0.17(8) &0.05(7) &-0.16(7) &0.04(6)& 0.44(4)\\
400: \tableline
401: &$R_{p}$ & 6.9 & 5.1 & 6.2 & 6.0&6.1\\
402: &$R_{wp}$ & 9.0 & 6.8 & 7.9 & 8.0&7.9 \\
403: &$R_{B}$ & 4.3 & 4.1 & 5.6 & 3.7&3.6\\
404: %\tableline
405: %--------------------------------------------------------------
406: %&$F _{x}$ &0 & 0 & 0.57(9) &3.48(7)&3.48(7)\\
407: %&$F _{y}$ &0 & 0 & 0.04(4) &0.52(4)&0.52(4) \\
408: %&$F _{z}$ &0 & 0 & 0.0 &0.0 &0.0\\
409: %&$F $ &0 & 0 & 0.57(9) &3.52(3)&3.52(1)\\
410: %&$R_{B}(F)$ &0 & 0 & 4.5 &2.0 &2.0 \\
411: %\tableline
412: %--------------------------------------------------------------
413: \tableline
414: Mn1-O1 $\times 2$& &1.9839(8) &1.932(6) & 1.941(8) &1.919(6)& 1.973(7)\\
415: Mn1-O2A$\times 2$& &2.006(3) & 1.999(5) & 1.996(6) &1.986(6)& 1.973(6)\\
416: Mn1-O2A$\times 2$& &1.9634(3) & 1.992(5) & 1.991(6) &1.995(7)& 2.002(7) \\
417: Mn2-O1 $\times 2$& & & 2.007(6)& 1.994(8) &2.026(6)& 1.975(7)\\
418: Mn2-O2B$\times 2$& & & 2.092(5) & 2.094(6) &2.086(6)& 2.071(6)\\
419: Mn2-O2B$\times 2$& & & 1.886(2) & 1.893(6) &1.891(7)& 1.893(6) \\
420: \end{tabular}
421: \end{ruledtabular}\label{table3}
422: \end{table*}%
423:
424: \begin{table*}[tbp] \centering%
425: \caption{La$_{1-x}$Ca$_{x}$MnO$_{3}$ ($x=0.175$) structural
426: parameters at several temperatures, as determined from Rietveld
427: refinements based on neutron powder diffraction data of E9
428: diffractometer ($\lambda=1.79784$\AA ). At $T=300$ K the space
429: group $Pnma$ (No 62) was used. La, Ca and apical oxygen (O1)
430: occupy the $4c$, $(x,1/4,z)$ site, Mn the $4b$ $(0,0,1/2)$ site
431: and the plane oxygen (O2) the general $8d$ site. For other
432: temperatures the monoclinic space group $P 2_1/c\equiv P\ 1\ 1\
433: 2_1/a$ (no 14, unique $c$-axis) was used. In this space group La,
434: Ca O1, O2A and O2B occupy the general $4e$ position $(x,yz)$. Mn1
435: and Mn2 occupy the $2c$ (1/2,0,0) and $2d$ (1/2,1/2,0) positions
436: respectively. Numbers in parentheses are statistical errors of the
437: last significant digit.}
438: \begin{ruledtabular}
439: \begin{tabular}{cccccccc}
440: $T$ (K) & & 300 & 200 & 170 & 120 & 70 & 2 \\
441: \tableline
442: $a$ (\AA )& & 5.5113(1) &5.5275(1) & 5.5323(1) & 5.5242(1)& 5.5165(1)&5.5152(1)\\
443: $b$ (\AA )& & 7.7778(1) &7.7449(2) & 7.7414(1) & 7.7453(1) & 7.7553(1)&7.7568(1)\\
444: $c$ (\AA )& & 5.5142(1) &5.5114(1) & 5.5104(1) & 5.5080(1) & 5.5050(1)&5.5042(1)\\
445: $\gamma$ (deg.)& 90.0 & &90.094(6) & 90.09(1) & 90.063(5) &90.059(4) &90.056(4) \\
446: \tableline
447: %--------------------------------------------------------------
448: La,Ca &$x$ & 0.0252(4)&0.0283(3)& 0.0292(5)& 0.0290(3) & 0.0280(3)& 0.0278(3) \\
449: &$y$ & 0.25 &0.248(1)& 0.244(1)& 0.247(1) & 0.249(1)& 0.248(1) \\
450: &$z$ & -0.008(1) &0.0063(5)&0.0087(7)& 0.0074(5) & 0.0070(6)&0.0077(6) \\
451: &$B$ & 0.57(4) &0.44(3) & 0.70(4) & 0.37(3) & 0.28(3) & 0.28(3)\\
452: \tableline
453: %--------------------------------------------------------------
454: Mn1 &$B$ & 0.37(7) &0.1(1) &0.34(8) & 0.08(6) & 0.11(6) & 0.13(6)\\
455: Mn2 &$B$ & &0.1(1) &0.34(8) & 0.08(6) & 0.11(6) & 0.13(65)\\
456: %--------------------------------------------------------------
457: \tableline
458: O1 &$x$ & 0.4927(8)&0.4914(6)&0.4910(8) & 0.4927(6) & 0.4913(6)& 0.4913(5)\\
459: &$y$ & 0.25 &0.247(1)&0.246(1) & 0.248(1) & 0.250(1)& 0.251(1)\\
460: &$z$ & 0.0680(8)&-0.0669(5)&-0.0646(7)&-0.0671(5) &-0.0665(6)&-0.0673(3)\\
461: &$B$ & 0.86(7) &0.65(5) &0.77(6) & 0.46(5) & 0.38(5) & 0.37(5) \\
462: %--------------------------------------------------------------
463: \tableline
464: O2A&$x$ & 0.2792(7)&0.221(1)&0.221(2) & 0.224(1) & 0.221(1)&0.221(1)\\
465: &$y$ & 0.0344(4)&-0.035(1)&-0.0360(9)&-0.0369(8) &-0.033(1)&-0.033(1)\\
466: &$z$ & 0.7240(7)&0.781(1)&0.771(2) & 0.770(1) & 0.781(1)& 0.778(2)\\
467: &$B$ & 0.7224(7)&0.62(3) &1.04(5) & 0.64(4) & 0.50(3) & 0.56(4) \\
468: \tableline
469: O2A&$x$ & &-0.209(1)&-0.209(2) &-0.208(1) &-0.213(1)&-0.214(1)\\
470: &$y$ & &0.464(1)&0.4676(9) & 0.4666(9) & 0.462(1)& 0.462(1)\\
471: &$z$ & &-0.766(1)&-0.771(2) &-0.780(1) &-0.770(1)&-0.774(2)\\
472: &$B$ & &0.62(4) &1.04(5) & 0.64(4) & 0.50(3) & 0.56(4) \\
473: \tableline
474: %--------------------------------------------------------------
475: &$R_{p}$ & 6.0 &5.4 & 6.8 & 5.8 & 6.1 & 6.2\\
476: &$R_{wp}$ & 8.0 &7.0 & 9.4 & 7.6 & 8.0 & 8.0 \\
477: &$R_{B}$ & 3.3 &3.6 & 5.6 & 4.2 & 3.8 & 3.8\\
478: %\tableline
479: %--------------------------------------------------------------
480: %&$F _{x}$ &0 &0 &1.50(5) &2.82(3) & 3.32(2) &3.54(2)\\
481: %&$F _{y}$ &0 &0 &0.0 &0.24 & 0.24 &0.24\\
482: %&$F _{z}$ &0 &0 &0.0 &0.0 & 0.0 &0.0\\
483: %&$F $ &0 &0 &1.5 &2.83(3) & 3.33(3) &3.55(2)\\
484: %&$R_{B}(F)$ &- &- &3.2 &3.8 & 2.8 &1.4 \\
485: Mn1-O1$\times 2$&&1.9809(1) & 1.949(1)& 1.93(1)& 1.95(1) &1.979(1) & 1.98(1) \\
486: Mn1-O2A$\times 2$&&1.996(4) & 1.973(8) & 2.00(1) & 2.000(8) &1.968(8) & 1.980(9)\\
487: Mn1-O2A$\times 2$&&1.973(4) & 1.995(8) & 1.95(1) & 1.959(8) &1.989(9) & 1.97(1) \\
488: Mn2-O1$\times 2$&& & 1.993(1)& 1.99(1)& 1.98(1) &1.96(1) & 1.96(1) \\
489: Mn2-O2B$\times 2$&& & 2.074(8) & 2.04(1) & 2.029(8) &2.043(8) & 2.028(9)\\
490: Mn3-O2B$\times 2$&& & 1.891(8) & 1.92(1) & 1.944(9) &1.922(8) & 1.93(1) \\
491:
492: \end{tabular}\end{ruledtabular}\label{table4}
493: \end{table*}%
494:
495: %\begin{table}[tbp] \centering%
496: %\caption{ Selected bond lengths (\AA) for La$_{1-x}$Ca$_x$MnO$_3$
497: %($x=0.175$) sample calculated from the structural model of Table
498: %\ref{table6}.}
499: %\begin{ruledtabular}
500: %\begin{tabular}{ccccccc}
501: %$T$ (K) & 300 & 200 &170 & 120 & 70 & 1.5\\
502: %\tableline
503: %Mn-O1$\times 2$&1.9809(1) & 1.9725(6)& 1.9694(7)& 1.9725(5) &1.9746(6)& 1.9752(6) \\
504: %Mn-O2$\times 2$&1.996(4) & 2.022(2) & 2.033(3) & 2.015(2) &2.009(2) & 2.007(2)\\
505: %Mn-O2$\times 2$&1.973(4) & 1.945(2) & 1.931(3) & 1.950(2) &1.951(2) & 1.953(2) \\
506: %\end{tabular}
507: %\end{ruledtabular}
508: %\label{table7}%
509: %\end{table}%
510:
511: \section{Crystal structure refinements}
512: Our structural and magnetization results on stoichiometric
513: La$_{1-x}$Ca$_{x}$MnO$_{3}$, with an emphasis on the FMI regime,
514: are summarized in Fig. \ref{diagram} in the form of a phase
515: diagram. The particular phase diagram is in good agreement with
516: the one estimated by Biotteau et al., \cite{biotteau01} using
517: single crystal samples. The $T_{\rm JT}$ curve is defined at the
518: temperature where the magnetization curves display a jump-like or
519: slope-change behavior and the unit cell parameter $a$ ($b$)
520: increases (decreases) abruptly on cooling. The particular behavior
521: of the unit cell constants resemble the situation observed in
522: LaMnO$_3$ compound on cooling through the $T_{\rm JT}$
523: transition.\cite{carvajal98} Depending on $x$, for $T<T_{\rm JT}$
524: a symmetry change from $Pnma$ to $P 2_1/c$ occurs. The $T_{\rm
525: C}(x)$ and $T_{\rm N}(x)$ lines concern the Curie and Neel
526: temperatures as were estimated both from neutron and magnetization
527: data. The line $T_{\rm M^/M^{//}}(x)$ is defined at the
528: temperature where the $a(T)$ and $b(T)$ curves display an abrupt
529: slope change, upon heating from $T=0$. Based on the transition
530: lines mentioned above the phase diagram of the
531: La$_{1-x}$Ca$_{x}$MnO$_{3}$ compound depicted in Fig.
532: \ref{diagram} displays two distinct regions as the ground state is
533: concerned. In the first region a canted antiferromagnetic (CAF)
534: orthorhombic structure is present. In the second one, depending on
535: $x$, a ferromagnetic insulating (FMI) monoclinic and orthorhombic
536: structures have been observed. The change from CAF to FMI ground
537: states occurs in a narrow concentration regime $0.12<x\lessapprox
538: 0.14$, where in neutron diffraction patterns (vide infra) a
539: ferromagnetic transition ($T_{\rm C}$) is observed firstly and
540: after an antiferromagnetic transition ($T_{\rm N}$) have been
541: observed upon cooling. Our high resolution synchrotron x-ray
542: diffraction patterns revealed that the true symmetry of the
543: samples in the boundary of the CAF and FMI regimes and partially
544: in the FMI regime is monoclinic with very small monoclinic angle
545: ($\gamma=0.04-0.1^{\rm o}$). This regime is denoted by symbol
546: M$^/$ in the phase diagram. The temperature variation of the unit
547: cell parameters in the FMI regime revealed an additional
548: structural transformation which occurs upon heating from $T=0$ at
549: temperature $T_{\rm M^/M^{//}}$. The magnitude of the unit cell
550: changes at the $T_{\rm M^/M^{//}}$ reduces when the $x$ is in the
551: CAF-FMI(M) and FMI(M$^/$)-FMI(O$^/$) phase boundaries. In this
552: phase diagram the ferromagnetic to paramagnetic ($T_{\rm C}$), JT
553: ($T_{\rm JT}$) and $T_{\rm M^/M^{//}}$ transition lines are all
554: estimated from neutron and magnetic measurements.
555: %******************************************************************
556: \begin{figure*}[tbp] \centering%
557: \includegraphics[angle=0,width=5in]{fig5.eps}
558: \caption{ Rietveld plot for La$_{0.85}$Ca$_{0.15}$MnO$_{3}$ at
559: $T=5$ K using the high resolution x-ray diffraction data from ID31
560: beam line of the ESRF. Show are details of the diffraction peaks
561: where the characteristic peak splitting compatible with $P 2_1/c$
562: space group is expected.}
563: \label{esrf}%
564: \end{figure*}%
565:
566: \begin{figure}[tbp] \centering%
567: \includegraphics[angle=0,width=0.8 \columnwidth]{fig6.eps}
568: \caption{ Rietveld plot for La$_{0.85}$Ca$_{0.15}$MnO$_{3}$ at
569: $T=320$ K. The observed data points are indicated by open circles,
570: while the calculated and difference patterns are shown by solid
571: lines. The positions of the reflections are indicated by vertical
572: lines below the patterns.}
573: \label{rietveld-320-r015}%
574: \end{figure}%
575:
576: \begin{figure}[tbp] \centering%
577: \includegraphics[angle=0,width=0.8 \columnwidth]{fig7.eps}
578: \caption{ Rietveld plot for La$_{0.85}$Ca$_{0.15}$MnO$_{3}$ at
579: $T=2$ K. The observed data points are indicated by open circles,
580: while the calculated and difference patterns are shown by solid
581: lines. The positions of the reflections are indicated by vertical
582: lines below the patterns.
583: %The inset, at the lower panel, shows the
584: %high angle observed neutron diffraction patterns for $T=320, 170$
585: %K and 2 K for a direct comparison.
586: }
587: \label{rietveld-5-r015}%
588: \end{figure}%
589: Figures \ref{abc-r013}, \ref{abc-r015} and \ref{abc-r0175}
590: (upper panels) show the temperature variation of the unit cell
591: parameter for $x=0.13, 0.15$ and 0.175 samples using the pseudo
592: orthorhombic description of the unit cell.
593: %********************************************************************
594: For $T<T_{\rm JT}$, the unit cell parameters follow the inequality
595: %********************************************************************
596: $a>c>b/\sqrt{2}$, a fact implying that in this pseudo orthorhombic
597: description the ${\rm O^/}$ structure is realized in the whole
598: temperature regime studied in this work. For all the samples there
599: is a temperature region where the unit cell parameters change
600: abruptly ($a$ increases, $b/\sqrt{2}$ decreases and $c$ decreases
601: slightly) resembling the situation observed\cite{carvajal98} in
602: the LaMnO$_3$ compound due to the orbital ordering state. As it
603: is mentioned above, we define the transition temperature $T_{\rm
604: JT}$ as the point where an extreme in the $da/dT$ curve occurs, in
605: the high temperature regime of the data. This particular
606: assignment agrees with the magnetic measurements (vide infra). The
607: orthorhombic strain parameter $s=2(a-c)/(a+c)$ (see Figs.
608: \ref{abc-r013}, \ref{abc-r015} and \ref{abc-r0175}, lower panels)
609: becomes maximum at the ferromagnetic transition temperature.
610: Furthermore, for $T<T_c$ $s$ decreases during cooling and finally
611: for $T<T_{\rm M^/M^{//}} \sim 60$ K (for the $x=0.13$ sample) it
612: becomes almost temperature independent. The particular behavior
613: reveal that the FMI state does not concern only the spin degrees
614: of freedom but is also related with the lattice and orbital ones.
615: The temperature variation of the unit cell volume $V(T)$ exhibits
616: a discontinuous change at $T_{\rm JT}$ implying that the JT
617: distortion is first order. As $x$ increases the volume jump is
618: difficult to be identified, on the basis of the available
619: diffractometer resolution. It is very likely that the JT
620: structural transition becomes second order for $x\ge 0.175$. Above
621: and below $T_{\rm JT}$ the unit cell volume decreases with
622: temperature nearly linearly, showing a small slope change at
623: $T_{\rm C}$. Finally, at $T_{\rm M^/M^{//}}$ $V(T)$ displays a
624: pronounced slope change for all samples.
625:
626: Let us present now the high resolution synchrotron x-ray data. The
627: samples for $0.125\leq x\leq 0.175$ for $T<T_{\rm JT}$ exhibited
628: additional reflections that can not be accounted for the $Pnma$
629: space group. More specifically, we observed splitting of the
630: $(hk0)$, and $(hkl)$ diffraction peaks not allowed in the $Pnma$
631: space group. This splitting is compatible with the $P 1 1
632: 2_1/a\equiv P 2_1/c$ (no 14, unique $c$-axis) space group. Fig.
633: \ref{esrf} depicts the high resolution synchrotron x-ray
634: diffraction data for the La$_{0.85}$Ca$_{0.15}$MnO$_3$ sample at
635: $T=5$ K, as an example. One can see very clearly the splitting of
636: the diffraction peaks with non zero $h$ and $k$. We obtained
637: similar data also for the other samples $x=0.125-0.175$. In this
638: monoclinic distortion $a$ glide-plane element is preserved, while
639: the $n$ and $m$ elements are lost. The loss of the mirror plane
640: may be signaling a change of the JT order from C type orbital
641: ordering (that is followed by LaMnO$_3$) to a new one. We note
642: that similar monoclinic distortion have been detected in
643: La$_{1-x}$Sr$_x$MnO$_3$ ($x=0.11-0.5$) compounds.\cite{cox01}
644: Based on this critical information about the symmetry of the
645: structure we refine the neutron data using the $P 2_1/c$ space
646: group.
647: %***************************************************************
648: The mirror plane at $y=1/4$ being present for $Pnma$ space group
649: is now absent, resulting in two Mn sites along the $b$ axis, the
650: first at $(1/2,0,0)$ and the second at $(1/2,1/2,0)$. Similarly,
651: instead of one plane oxygen site (O2, $P nma$ symmetry) there are
652: two now in the $P 2_1/c$ symmetry. Thus, in this new symmetry two
653: crystallographically independent Mn sites produce a layer type
654: arrangement of two different MnO$_{6}$ octahedra along the $b$
655: axis.
656: %***************************************************************
657: The refinement strategy employed in this work was to first
658: optimize the scale factor, background, unit cell parameters and
659: zero-shift errors. Subsequently, we refined the atomic positional
660: and isotropic thermal parameters. In order to reduce the number of
661: parameters we decided to refine the thermal parameters for all the
662: sites isotropically and we kept equal the thermal parameters of
663: the Mn and O2 sites. Figures \ref{rietveld-320-r015} and
664: \ref{rietveld-5-r015} show two representative Rietveld plots of
665: neutron data for La$_{0.85}$Ca$_{0.15}$MnO$_{3}$ sample at 320 K
666: and 2 K, respectively. The corresponding structural parameters for
667: all the studied samples are reported in Tables
668: \ref{table1}-\ref{table4}.
669:
670: Fig. \ref{structure} shows projections of the monoclinic structure
671: of the La$_{0.85}$Ca$_{0.15}$MnO$_{3}$ compound for $T=170$ K on
672: the $bc$ and $ab$ planes, respectively. All the samples show
673: qualitatively the same behavior, as far as the variation of the
674: structural parameters with temperature is concerned. Nevertheless,
675: there are significant quantitative differences related to the
676: influence of the hole doping through Ca substitution.
677:
678: \begin{figure}[tbp] \centering%
679: \includegraphics[angle=0,width=3in]{fig8a.eps}
680: \includegraphics[angle=0,width=3in]{fig8b.eps}
681: %\includegraphics[angle=0,width=3in]{ac.eps}
682: \caption{Crystal structure of La$_{0.85}$Ca$_{0.15}$MnO$_{3}$
683: compound at $T=170$ K. The upper and lower panels show the
684: projections along the $a$ and $c$ axis, respectively. The large
685: hatched, small filled and intermediate open spheres represent La,
686: Mn and O ions respectively. The Mn1, Mn2 sites are located at
687: $y=0$ and $y=1/2$ planes respectively ($b-$axis is the largest
688: one). The sort Mn1-O, Mn2-O bonds denoted by grey rods are
689: directed along and perpendicular to the $b$-axis, respectively.}
690: \label{structure}%
691: \end{figure}%
692:
693: The Bragg peaks are quite sharp for $T>T_{\rm C}(x)$ but for
694: $T<T_{\rm JT}$ they develop a pronounced selective peak
695: broadening. In order to account for the selective peak broadening
696: the Stephens\cite{stephens99} microstrain model was used. As our
697: Rietveld refinement results reveal the microstrain parameters
698: follow the temperature variation of the unit cell parameters. The
699: most significant broadening is related to $S_{202}$ and $S_{400}$
700: parameters. The $S_{202}$ strain parameter describes a correlated
701: type random strain between $a$ and $c$ crystallographic directions
702: and it is reasonable to be attributed to some kind of local static
703: orbital ordering correlations. The average structure revealed a
704: cooperative JT distortion which increases as temperature
705: decreases. Parallel to this behavior, there is an antagonistic
706: interaction which produces correlated fluctuation of the cell
707: constants. These fluctuations are directly connected with the
708: development of the ferromagnetic long range order and temperature
709: variation of the unit cell parameters. The microstrain parameters
710: take the maximum value at $T_{\rm C}$. Most importantly the
711: microstrain parameters are reduced about 30\% at $T_{\rm
712: M^/M^{//}}$ and below this temperature remain nearly temperature
713: independent. Consequently, our structural data reveal a complex
714: magneto-elastic behavior for $T<T_{\rm C}$, that characterizes the
715: FMI regime of the La$_{1-x}$Ca$_{x}$MnO$_{3}$ compound.
716:
717: Anisotropic broadening is usually connected with the presence of
718: correlated structural defects, due to the preparation method, or
719: is inherent to structural or magnetic transformations. Since we
720: observe temperature dependance in the microstrain parameters, the
721: case of the structural defects caused by the preparation method
722: should be ruled out, because in this case a temperature
723: independent behavior is expected. It is natural one to ask what
724: is the physical origin of the particular temperature variation of
725: the microstrain parameters. Essentially, the anisotropic peak
726: broadening concerns the lattice planes $(h00)$ and $(h0l)$,
727: meaning that there is some modulation along the $a$-axis and ${\bf
728: a}+{\bf c}$ -direction.
729: %*******************************************
730: Our findings may be related with recent theoretical studies where
731: the important role of the elastic energies (strain) in determining
732: the ground state of the system has been pointed
733: out.\cite{calderon03,ahn04} Furthermore, the anisotropic
734: broadening may originate from grains which contain different
735: structural twin domains corroborated by the successive
736: magneto-structural transformation occurring upon cooling.
737: %*******************************************************
738: Finally, one cannot disregard the case where phase separation is
739: present especially near the CAF-FMI phase boundary. In such a case
740: several phases with nearly similar parameters occur producing
741: indirectly a selective peak broadening, which is accounted for by
742: microstrain formalism.
743:
744:
745: \section{Bond lengths}
746: \begin{figure}[tbp] \centering%
747: \includegraphics[angle=0,width=0.8 \columnwidth]{fig9.eps}
748: \caption{ Mn-O$_i$ ($i=1,2$) bond distances as a function of
749: temperature for $x=0.125, 0.15$ and 0.175 samples. The upper panel
750: shows the Mn1-O bonds and lower panel the Mn2-O bonds (Mn1-O1,
751: Mn2-O1 are the apical bonds and Mn1-O2A, Mn2-O2B are the in-plane
752: bonds). Lines through the experimental points are guides to the
753: eye. The solid, open and "crossed" symbols correspond to
754: $x=0.125$, 0.15 and 0.175 respectively. The data for the LaMnO$_3$
755: (thin lines) and La$_{0.75}$Ca$_{0.25}$MnO$_3$ (thick lines)
756: samples are also included for a direct comparison.}
757: \label{bond}%
758: \end{figure}%
759:
760: In Fig. \ref{bond} are depicted the temperature variation of the
761: Mn-O bond lengths for $x=0.125, 0.15$ and 0.175 samples, deduced
762: from neutron diffraction data. For $T<T_{\rm JT}$ the crystal
763: structure of the La$_{1-x}$Ca$_{x}$MnO$_{3}$ in the FMI regime
764: based on the $P 2_1/c$ space group have two Mn sites.
765: %We include the MnO bond lengths that appear in LaMnO$_3$
766: %with thin lines in Fig. \ref{bond} for a direct comparison.
767: As the Ca content increases the long Mn2-O bond length (which has
768: been attributed to the $d_{z^2}$ orbital) decreases. The short
769: Mn2-O bonds are approximately at the same level as in the
770: LaMnO$_3$ case. Concerning the intermediate Mn2-O bond, its
771: magnitude is above the corresponding one of LaMnO$_3$. Is is
772: interesting to note that the long and short bonds follow the
773: temperature variation of the $a$ and $c$ cell parameters or in
774: other words these bonds define the temperature variation of the
775: unit cell parameters. Based on the similarities of the Mn2-O bond
776: lengths with those of LaMnO$_3$, it is reasonable to attribute the
777: Mn2 site to a Mn$^{+3}$ like site in the
778: La$_{1-x}$Ca$_{x}$MnO$_{3}$ ($0.125\le x \le 0.175$) compound.
779:
780: Let us turn now to the Mn1-O bond lengths. As their magnitude is
781: concerned these bond lengths, are more close to the bond lengths
782: of MnO, observed in the FMM region of the phase diagram. In Fig.
783: \ref{bond} (upper panel) with bold lines, the Mn-O bond lengths
784: observed in La$_{0.75}$Ca$_{0.25}$MnO$_3$ compound are depicted.
785: One can see that as $x$ increases the Mn1-O bond lengths approach
786: these lines. It is plausible to attribute the Mn1 site to a mixed
787: valence Mn like site as observed in the
788: La$_{0.75}$Ca$_{25}$MnO$_3$ compound. As $x$ reduces towards the
789: CAF regime the Mn1-O bond lengths become non degenerated and move
790: away from those of La$_{0.75}$Ca$_{0.25}$MnO$_3$
791: compound.\cite{radaelli97a}
792:
793: Except for the metrical variation of the bond lengths there is an
794: orientation difference, concerning the short Mn$i$-O bonds
795: ($i=1,2$), between the two Mn-sites. In the Mn2-site (which
796: resembles the LaMnO$_3$ case) the short bond lies in the $ac$
797: plane, contrary to the Mn1-site, where the short Mn1O-bond is
798: directed, approximately, along the $b$-axis.
799: %******************************************
800: According to the authors' opinion this is the
801: key point to which one can be based in order
802: to explain the FMI phase. It is important to note the fundamental
803: difference between the conventional JT orthorhombic O$^/$ and the
804: monoclinic M$^/$ structures. The latter cannot be the usual C-type
805: antiferrodistortive arrangement found in the O$^/$ structure in
806: LaMnO$_3$ and CAF part of the phase diagram. As proposed by Cox et
807: al\cite{cox01} in the case of La$_{1-x}$Sr$_x$MnO$_3$, the loss of
808: the mirror plane would be consistent with a change in JT order
809: from C-type to G-type at which there is antiferrodistortive
810: coupling of MnO$_6$ octahedra along the $b$ direction instead of
811: parallel coupling (see Fig. \ref{structure}).
812:
813:
814:
815: \section{Magnetic measurements and magnetic structure}
816:
817: Fig.~\ref{moment15} and Fig.~\ref{moment175} (upper panels) show
818: the temperature variation of the magnetic moment per Mn ion for
819: the samples $x=0.15$ and 0.175, respectively, for several dc
820: magnetic fields. We obtained similar magnetic measurements data
821: also for $x=0.125,0.13$ and 0.14 samples. In the lower panels of
822: Figs.~\ref{moment15}, and \ref{moment175} we plot the temperature
823: dependance of the normalized real part of the ac susceptibility,
824: and the low field dc magnetic moment (left axis). In addition, we
825: plot the temperature variation of the inverse dc susceptibility
826: ($H/m$) (right axis) measured under a dc-field $H=5$ kOe. The
827: abrupt increasing of the magnetic signal in the dc and ac
828: measurements at $T_{\rm C}=168$ K and $T_{\rm C}=187$ K for
829: $x=0.15$ and $x=0.175$ respectively marks a transition from a
830: paramagnetic to ferromagnetic state, in agreement with neutron
831: data (vide infra). The sharp peak at $T_{\rm C}$ in the $\chi'$ is
832: related with the Hopkison effect, as we discussed in Ref.
833: \onlinecite{pissas04a}. Upon heating from $T=0$, $\chi'(T)$
834: slightly increases up to $T\approx T_{\rm M^/M^{//}}$. For
835: $T>T_{\rm M^/M^{//}}$ $\chi'(T)$ increases in a step wise fashion
836: as the dashed vertical lines in Figs. \ref{moment15} and
837: \ref{moment175} show. This peculiarity reveals the magnetic
838: signature of the structural change that occurs at $T_{\rm
839: M^/M^{//}}$. The particular anomaly is present and in the dc
840: magnetization measurements as it is seen for example, in the upper
841: panel of the Fig. \ref{moment15} where a step like slope change
842: occurs at $T_{\rm M^/M^{//}}$.
843:
844: Furthermore, a close inspection of the $m(T)$ curves at the
845: paramagnetic regime reveals an additional anomaly which is related
846: with JT distortion observed in the crystal structure data (see
847: inset of Fig. \ref{moment15}). This anomaly is more pronounced for
848: $x=0.11$ sample, its signature reduces as $x$ increases and
849: finally becomes hard to discern for the samples with $x\ge 0.175$.
850: The anomaly in the magnetic measurements at $T_{\rm JT}$ is seen
851: more clearly in the inverse susceptibility data (see lower panels
852: of Fig.~\ref{moment15} and Fig.~\ref{moment175}). Comparing these
853: data one can clearly see that the jump of $H/m$ at $T_{\rm JT}$ is
854: transformed to a slope change as $x$ increases. The particular
855: change at $T_{\rm JT}$ for the $x> 0.175$ samples may mark a
856: change of JT transition from first to second order.
857: \begin{figure}[tbp] \centering%
858: \includegraphics[angle=0,width=0.85 \columnwidth]{fig10.eps}
859: \caption{ (upper panel) Temperature variation of the bulk magnetic
860: moment per ion Mn for La$_{0.85}$Ca$_{0.15}$MnO$_{3}$. The data
861: were collected under the indicated magnetic fields using the
862: standard ZFC-FC procedure. The inset shows the details of the
863: measurements at the $T_{\rm JT}$ region. (lower panel) Temperature
864: variation of the inverse dc magnetic moment, normalized real part
865: of the fundamental ac-susceptibility (bold line) and normalized
866: ZFC and FC bulk magnetic moment measured under a dc field 100 Oe
867: (thin line). The dot lines corresponds to the Curie-Weiss law used
868: to fit the data above and below $T_{\rm JT}$. The vertical dashed
869: line indicates the $T_{\rm O^/O^{//}}$. }
870: \label{moment15}%
871: \end{figure}%
872:
873: \begin{figure}[tbp] \centering%
874: \includegraphics[angle=0,width=0.85 \columnwidth]{fig11.eps}
875: \caption{ (upper panel) Temperature variation of the bulk magnetic
876: moment per ion Mn for La$_{0.825}$Ca$_{0.175}$MnO$_{3}$. The data
877: were collected under the indicated magnetic fields using the
878: standard ZFC-FC procedure. (lower panel) Temperature variation of
879: $H/m$, normalized real part of the fundamental ac-susceptibility
880: (bold line) and normalized ZFC and FC bulk magnetic moment
881: measured under a dc magnetic field 100 Oe (thin line). The dot
882: lines corresponds to the Curie-Weiss law used to fit the data
883: above and below $T_{\rm JT}$. The vertical dashed line indicates
884: the $T_{\rm M^/M^{//}}$.}
885: \label{moment175}%
886: \end{figure}%
887:
888:
889: Above and below $T_{\rm JT}$, $\chi^{-1}$ can be fitted by a
890: straight line implying a Curie-Weiss law
891: ($\chi^{-1}=(T-\Theta)/C)$. The Curie constant is given by
892: $C=(N/V_c)g^2\mu_B^2S(S+1)/3k_B$, where $N$ is the number of
893: magnetic ions inside the unit cell, $V_c$ is the unit cell volume,
894: $g\approx 2$ ($S=J, L=0)$ is the Land\'{e} factor, $\mu_B$ is Bohr
895: magneton, $S$ the spin of the ion, $k_{\rm B}$ is Boltzmamn's
896: constant and $\Theta$ is the Weiss constant. Since $\chi^{-1}$ is
897: nearly parallel above and below $T_{\rm JT}$ it is reasonable to
898: attribute this change to $\Theta$. Therefore, since $\Theta\propto
899: J$ (where $J$ is the exchange coupling) the abrupt change in the
900: $\chi^{-1}$ may be related to a discontinuous change of $J$ above
901: and below $T_{\rm JT}$. The O$^/$ phase has larger exchange
902: constants in comparison to the M$^{/}$ one. The intersection of
903: $H/m$ with the temperature-axis, for $T<T_{\rm JT}$ gives a
904: $\Theta=175 $ K, in good agreement with the Curie temperature
905: estimated from neutron data (vide infra). From the Curie constant
906: we estimate magnetic moment per ion which is $6.01 \mu_B$. This
907: value is in disagreement with the one theoretically expected for
908: an appropriate, random mixture of Mn$^{+3}$ and Mn$^{+4}$ ions,
909: $\mu^2=(1-x)g^2S_3(S_3+1)+xg^2S_4(S_4+1)$, where $S_{3,4}=2$ and
910: $3/2$ for Mn$^{+3}$ and Mn$^{+4}$ respectively. Enhanced
911: paramagnetic effective moment has also been observed in Sr-based
912: compound and has been attributed to magnetic
913: clusters.\cite{nogues01,nojiri99}
914: %In the lower panel of Fig.~\ref{diagram}%
915: % is depicted the%
916: %concentration variation of the%
917: %effective magnetic moment.%
918: The larger effective magnetic moment of the M$^{/}$ phase for
919: $x>0.15$ may be related with the idea of magnetic polaron that
920: "survives" in the regime $T_{\rm C}\le T\le T_{\rm JT}$. We must
921: note that this state is more relevant when the system moves away
922: from the CAF regime. Finally, it is interesting to note the
923: similarity of our bulk magnetic moment data with those of Refs.
924: \onlinecite{nogues01,nojiri99} referring to the Sr-compound. There
925: is however a distinct difference below $T_{\rm C}$. Our ac
926: susceptibility, reduces significantly for $T< T_{\rm M^/M^{//}}$,
927: while in the Sr-case it increases in a stepwise fashion. This
928: "fine" difference may imply that the M$^{//}$ states of Ca and Sr
929: compounds are different. The orthorhombic strain and the long
930: in-plane Mn-O2 distance reduction is very similar but the ordered
931: magnetic moment, bulk magnetic moment and ac susceptibility show
932: opposite change at $T_{\rm M^/M^{//}}$.
933:
934: \begin{figure}[tbp] \centering%
935: \includegraphics[angle=0,width=1.0 \columnwidth]{fig12.eps}
936: \caption{ (Upper panel) Selected portion of the neutron powder
937: diffraction patterns for La$_{0.89}$Ca$_{0.11}$MnO$_{3}$ for $2\le
938: T \le 210$ K. (Lower panel) Decomposition of the neutron
939: diffraction pattern as $T=2$ K. Shown are the profiles coming from
940: the A- antiferromagnetic, the ferromagnetic and crystal
941: structures. The inset shows the observed and calculated profiles
942: for the Bragg peaks $(210), (012)$ and $(030)$ based on the models
943: $F_xF_yA_z$ and $A_xF_yF_z$ . }
944: \label{m1}%
945: \end{figure}%
946:
947: Let us present now the neutron diffraction data related with the
948: magnetic structure. The magnetic structures for all the samples
949: are best monitored by observing the neutron diffraction patterns,
950: using data from E6 instrument with $\lambda=2.4$\AA. Fig. \ref{m1}
951: show portions of neutron diffraction patterns in the region where
952: the magnetic coherent intensity is maximum for $x=0.11$ sample.
953: Similar data were also collected for the other samples. In all
954: samples we observed an increasing of the Bragg intensity for the
955: peaks $(101)$ and $(020)$ for $T<T_{\rm C}$, something which is
956: compatible with the development of long range ferromagnetic order.
957: In addition, the samples $x=0.11, 0.125$ and 0.13 show the
958: characteristic magnetic peaks of the $A$ antiferromagnetic
959: structure. The most pronounced magnetic Bragg peak of the
960: A-magnetic structure is the $(010)$ located at $\approx 18^{\rm
961: o}$. As $x$ increases, the area of this peak at $T=2$ K reduces,
962: implying that the antiferromagnetic ordered moment reduced and
963: finally becomes zero for $x\gtrapprox 0.14$.
964: %**********************************************
965: We refined the $x=0.11$ sample, where the pronounced
966: orthorhombicity enables us to safely determine the spin components
967: along the three axes. The lower panel of Fig. \ref{m1} shows a
968: decomposition of the neutron diffraction pattern at $T=2$ K, which
969: consists of the patterns coming from the A-type antiferromagnetic,
970: the ferromagnetic and the crystal structure patterns. The inset of
971: the lower panel of Fig. \ref{m1} shows the calculated patterns of
972: models $F_xF_yA_z$ and $A_xF_yF_z$ ($F_x$ denotes ferromagnetic
973: moment along $x$-axis and $A_x$ magnetic moment along $x$-axis
974: which follows the A-antiferromagnetic structure). The agreement
975: between the theoretical and experimental profile is better for
976: the $A_xF_yF_z$ model. This magnetic model is closely connected
977: with the canted magnetic structure observed in the LaMnO$_3$
978: compound.\cite{moussa96} Similarly, for $x=0.125$ sample, except
979: for the major ferromagnetic component, small intensity coming from
980: $A$-type antiferromagnetic structure is present. In addition, the
981: negligible intensity of the $(200)$ magnetic peak, implies that
982: the ferromagnetic component cannot be parallel with $b$-axis.
983: Therefore there are two possibilities. The first is the
984: antiferromagnetic component to be parallel with the $c$-axis,
985: while the ferromagnetic component parallel with $a$ and $b$ axis.
986: The second possibility is both the antiferromagnetic and part of
987: the ferromagnetic component to be parallel with $a$-axis. Since it
988: is not possible to have simultaneous ferromagnetic and
989: antiferromagnetic spin components along the $a$-axis in the same
990: domain these two different magnetic states concern different
991: domains. This finding is not strange since in this concentration
992: regime the phase boundary separating the CAF and FMI ground states
993: exists.
994:
995: For the sample $x=0.15$ the NDP at $T=2$ K shows only
996: ferromagnetic magnetic peaks. The orthorhombicity of the sample
997: permits the calculation of the orientation of the magnetic moment
998: in respect of the crystallographic axes. Thus, the case ${\bf
999: m}||b$-axis should be excluded. If the magnetic moment was along
1000: the $b$-axis the magnetic contribution at the $(020)$ peak should
1001: be zero. Based on the best agreement between observed and
1002: theoretically calculated profiles we conclude that the ordered
1003: magnetic moment should be lie on the $ac$-plane with its larger
1004: component along the $a$-axis. Nevertheless, a small component
1005: along the $b$-axis cannot be totally excluded. The ordered
1006: magnetic moments as estimated from Rietveld refinements for the
1007: $x=0.13, 0.15$ and 0.175 samples are plotted in the lower panels
1008: of Figs. \ref{abc-r013}, \ref{abc-r015} and \ref{abc-r0175},
1009: respectively.
1010: %*******************************
1011: A weak but distinctive anomaly is present in the temperature
1012: variation curves of the ordered magnetic moments at $T_{\rm
1013: M^/M^{//}}$. It is imperative to note that for $T\approx T_{\rm
1014: M^/M^{//}}$ the ordered magnetic moment curve, for $x=0.15$ and
1015: 0.175 samples, displays a small slope change. For samples
1016: $x=0.11, 0.125$ and 0.13 where we can detect the magnetic peak
1017: from the $A-$type magnetic structure, this anomalous behavior
1018: certainly is related with the antiferromagnetic component.
1019: %***********************
1020: For $x\ge 0.15$ it is possible that a small percentage of the
1021: spin of Mn ions may prefer a glass like behavior for $T<T_{\rm
1022: M^/M^{//}}$, as the ac susceptibility measurements
1023: show.\cite{markovich02a} Concerning this "strange" magnetic
1024: component it is not clear how it is connected with the structural
1025: transition at the same temperature. One could invoke the idea of
1026: the phase separation to explain this behavior, however, the bulk
1027: character of the concomitant structural transition at $T_{\rm
1028: M^/M^{//}}$ is not so favorable for this interpretation. Based on
1029: the experimental data one may say that the structural transition
1030: prevents some part of the Mn-spin to participate in the long-range
1031: ordering.\cite{markovich02a,pissas04a} This structural transition
1032: should be related with the orbital degree of freedom in agreement
1033: with M\"{o}ssbauer\cite{pissas04b} and NMR\cite{papavassiliou03}
1034: data.
1035:
1036: \section{Discussion}
1037:
1038: Unexpectedly, the space group $P 2_1/c$ has been used by Lobanov
1039: et al.,\cite{lobanov00} in order to describe the crystal structure
1040: for a sample with nominal stoichiometry
1041: La$_{0.85}$Ca$_{0.15}$MnO$_3$ prepared by liquid road and final
1042: reaction temperature 900$^{\rm o}$C. However, although this space
1043: group is the same with that used in our study, several differences
1044: exist. More specifically: (a) The ac susceptibility and
1045: resistivity data are different. (b) The sample of Lobanov et
1046: al.,\cite{lobanov00} does not show the pseudo-orthorhombic
1047: splitting seen in our data and in the La$_{1-x}$Sr$_x$MnO$_3$
1048: compound. The lattice parameters follow a monotonic temperature
1049: variation, contrary to ours. (c) The particular arrangement of the
1050: short bonds of Mn1-O and Mn2-O is different with it deduced from
1051: our data. Since no more data, from samples prepared with method of
1052: Ref. \onlinecite{lobanov00}, are available, we cannot safely
1053: further comment on the possible relation between the two studies.
1054:
1055: It is interesting to discuss our crystal structure data in terms
1056: of the normal coordinates\cite{kanamori60} of the octahedral
1057: complex MnO$_6$, $Q_3$ and $Q_2$. An arbitrary $e_g$ orbital
1058: state can be written as
1059: $|\theta\rangle=\cos(\theta/2)|d_{z^2}\rangle+\sin(\theta/2)|d_{x^2-y^2}\rangle$.
1060: The orbitals $|d_{z^2}\rangle$ and $|d_{x^2-y^2}\rangle$
1061: correspond to the states $|\theta=0\rangle$ and
1062: $|\theta=\pi\rangle$ respectively. It is also possible, the
1063: orbitals ($|d_{x^2}\rangle$, $|d_{y^2}\rangle$) and (
1064: $|d_{z^2-x^2}\rangle$, $|d_{z^2-y^2}\rangle$) to be formed. In
1065: this case we have $|\theta=\pm (2/3)\pi\rangle$ and $|\theta=\pm
1066: (1/3)\pi\rangle$, respectively. Experimentally the orbital
1067: filling can be estimated from crystallographic data, through the
1068: Mn-O bond lengths using the equation\cite{kanamori60,kugel82}
1069: $\tan\theta=Q2/Q3$, where $Q_3=2/\sqrt{6}(2m-l-s)$ and
1070: $Q_2=\sqrt{2}(l-s)$. Here $l,m$ and $s$ mean long, medium and
1071: short Mn-O bond lengths, respectively. $Q_2$
1072: mode mixes the $|d_{x^2-y^2}\rangle$
1073: and $|d_{z^2}\rangle$ orbitals and produces a two sublattice
1074: orbital ordering pattern, with a lower energy orbital
1075: $\Psi_g=c_1|d_{x^2-y^2}\rangle\pm c_2|d_{z^2}\rangle$ and a higher
1076: energy orbital $\Psi_e=c_2|d_{x^2-y^2}\rangle\mp
1077: c_1|d_{z^2}\rangle$ ($\pm$ is referring to the two sublattices).
1078:
1079: Using the crystallographic data at $T=300$ K of LaMnO$_3$ (Ref.
1080: \onlinecite{carvajal98}) one finds, $Q_3=-0.121$\AA\ and
1081: $Q_2=0.383$\AA. These values correspond to a canted
1082: antiferromagnet like orbital arrangement with two sublattices,
1083: with $\theta=\pm 72.38^{\rm o}$, respectively. Based on the
1084: crystallographic data of the present study we calculated the $Q_2$
1085: and $Q_3$ parameters for the two cites predicted from $P 2_1/c$
1086: structural model. Fig. \ref{q2q3} shows the temperature variation
1087: of $Q_2$ and $Q_3$ for the $x=0.125, 0.15$ and 0.175 samples. With
1088: solid lines we also plot the values which correspond to the
1089: LaMnO$_3$ compound. The first comment which can be made is that
1090: both $Q_2$ and $|Q_3|$ parameters for the two sites have lower
1091: values, in comparison to these of the LaMnO$_3$ compound. The
1092: second one is that, both Mn sites have $Q_3\approx 0 $ with the
1093: first displaying a high value while the second a moderate one.
1094: Since we only have a small number of data it is difficult to
1095: recognize any change at $T_{\rm M^/M^{//}}$. The third comment is
1096: that, only the sample $x=0.125$ shows non monotonous variation of
1097: $Q_2$ at $T_{\rm M^/M^{//}}$. We note that for $T>T_{\rm JT}$ the
1098: two sites pattern collapse into one site, with both $Q_2$ and
1099: $Q_3$ modes takng values near zero.
1100: \begin{figure*}[tbp] \centering%
1101: \includegraphics[angle=0,width=5 in]{fig13.eps}
1102: \caption{ Temperature variation of the $Q_2$ and $Q_3$ parameters
1103: for La$_{1-x}$Ca$_x$ MnO$_3$ ($x=0.125, 0.15$ and 0.175) samples.
1104: The data for the LaMnO$_3$ compound are also included for a direct
1105: comparison (thin line).}
1106: \label{q2q3}%
1107: \end{figure*}%
1108:
1109: \subsection{The comparison with Sr-compound}
1110:
1111: First of all, we must point out the similarities between the Sr
1112: and Ca based compounds, by comparing the neutron data from the
1113: stoichiometric La$_{1-x}$Sr$_x$MnO$_3$ ($0.1\le x\le 0.2$) of
1114: Dabrowski et al.\cite{dabrowski99,xiong99}with ours. As Dabrowski
1115: et al. have pointed out,\cite{dabrowski99} the phase diagrams of
1116: the Sr case reported in the recent literature were constructed
1117: using physical and structural information obtained for dissimilar
1118: samples. Specifically, in the doping regime $0\le x<0.3$ of
1119: Sr-compound the structure can form vacancies on both the A(La,Sr)
1120: and B(Mn) sites during synthesis under oxidizing conditions.
1121: Formation of equal number of vacancies on the A and B sites would
1122: increase the average Mn oxidation state. Let us discuss our
1123: results with those of single crystal data of
1124: La$_{7/8}$Sr$_{1/8}$MnO$_3$ compound (this concentration is the
1125: most representative and widely studied of the particular regime)
1126: where very interesting results are available.
1127:
1128: In a recent resonant x-ray scattering study on a
1129: La$_{7/8}$Sr$_{1/8}$MnO$_{3}$ single crystal, Geck et
1130: al.,\cite{geck04} have shown that the orbital state occurring in
1131: the interval $T_{\rm MT}\le T\le T_{JT}$, (M means monoclinic and
1132: T triclinic) is similar to that found in LaMnO$_3$. For $T<T_{\rm
1133: MT}$ an orbital reordering is compatible with the resonant data.
1134: This group\cite{geck04} observed non zero resonant and nonresonant
1135: intensity at (3/2, 3/2, 3) superlattice reflection, only for
1136: $T<T_{\rm MT}$. The nonresonant scattering in the particular
1137: reflection implies some kind of a superstructure modulation that
1138: develops at $T=T_{\rm MT}$. Such behavior is related to structural
1139: and electronic degrees of freedom, speculatively attributed to the
1140: orbital polarons formation.\cite{mizokawa00,kilian98}
1141: %**********************
1142: The work of Geck et al.,\cite{geck04} further supports the aspect
1143: that the CAF and FMI phases of Ca and Sr based
1144: La$_{1-x}$(Ca,Sr)$_x$MnO$_3$ compounds have both common
1145: characteristics, and distinct differences. It seems that the
1146: vertical boundary of the Sr-case that separates the O$^/$ CAF
1147: ground state from the triclinic new unknown orbital state exists
1148: even in the Ca-case. Our results show that: (a) Similar to the Sr
1149: compound for $T_{\rm MT}\le T\le T_{\rm JT}$ the structure is
1150: monoclinic $(P 2_1/c)$. (b) Based on our high resolution
1151: synchrotron x-ray diffraction data, the crystal structure of the
1152: Ca case remains monoclinic, while in the Sr case it is triclinic
1153: and (c) both families of compounds exhibit significantly lower JT
1154: distortion in comparison with the M$^/$ phase which occurs for
1155: $T_{\rm M^/M^{//}}<T<T_{\rm TJ}$.
1156:
1157:
1158: \subsection{Comparison between stoichiometric and cation deficient samples}
1159:
1160: We would like to emphasize once again
1161: that our results concern stoichiometric samples. The
1162: stoichiometric samples have properties different from those
1163: prepared in air. \cite{huang98,pissas04a,dai96} The most prominent
1164: differences between the two families of samples are that the
1165: oxygen deficient samples do not display the CAF structure and do
1166: not undergo the characteristic cooperative JT distortion,
1167: irrespectively of the nominal calcium concentration. The second
1168: difference makes the cation deficient samples to be in an orbital
1169: liquid state below $T_{\rm C}$, in the FMI regime, contrary to the
1170: stoichiometric ones where the ferromagnetic phase takes place in
1171: the cooperative ordered JT state at least in the interval $T_{\rm
1172: M^/M^{//}}<T<T_{\rm C}$. In addition, the cation deficient samples
1173: (e.g. $x=0.1, 0.15$ samples\cite{huang98,dai96b}) show an almost
1174: monotonous variation with temperature of the unit cell parameters,
1175: satisfying also the inequality $c>a>b/\sqrt{2}$ (for $T<T_{\rm
1176: C}$), contrary to the behavior observed in our stoichiometric
1177: samples where a pronounced non monotonous variation is observed.
1178: It is now well established that for low Ca concentrations the air
1179: prepared samples have cation
1180: vacancies.\cite{pissas04a,dabrowski99a,roosmalen94} The amount of
1181: vacancies increases as the nominal Ca concentration is near $x=0$.
1182: These samples are ferromagnetic irrespectively of $x$. The ordered
1183: magnetic moment at zero temperature is significantly lower than
1184: the corresponding moment of the stoichiometric samples. For the
1185: $x=0.1$ and 0.15 non stoichiometric samples the ordered magnetic
1186: moment was found $\sim 3 \mu_{\rm B}$ (Refs.
1187: \onlinecite{huang98,dai96b,algarabel03}), while in our
1188: stoichiometric samples are obtained $\sim 3.7\mu_{\rm B}$.
1189:
1190: In order to check once more the situation we selected the $x=0.15$
1191: sample, prepared in air for a detailed structural characterization
1192: using neutron diffraction data. Fig. \ref{ap015} shows the
1193: temperature variation of the unit cell parameters (upper panel),
1194: the ordered magnetic moment and the Mn-O bond lengths (lower
1195: panel), as they were deduced from neutron diffraction data using
1196: E9 instrument. In perfect agreement with the above, this sample is
1197: ferromagnetic insulator without the orthorhombic splitting that
1198: characterizes the cooperative JT distortion of the reduced
1199: samples. Interestingly, the ordered magnetic moment is lower than
1200: the one of the reduced samples (e.g 3.2 vs 3.7$\mu_B$).
1201:
1202: \begin{figure}[tbp] \centering%
1203: \includegraphics[angle=0,width=0.7 \columnwidth]{fig14.eps}
1204: \caption{ (upper panel) Temperature variation of the lattice
1205: parameters and unit-cell volume for
1206: La$_{0.85}$Ca$_{0.15}$MnO$_{3}$ compound prepared in air
1207: atmosphere. (lower panel) Temperature variation of the ordered
1208: ferromagnetic moment and Mn-O$_i$ bond lengths. Lines through the
1209: experimental points are guides to the eye.}
1210: \label{ap015}%
1211: \end{figure}%
1212: In the work of Algarabel et al.\cite{algarabel03} that concerns a
1213: cation deficient sample with nominal $x=0.1$, significant small
1214: angle neutron scattering intensity ( due to nanometric clusters)
1215: has been found to coexist with the long-range ferromagnetic phase.
1216: In our opinion, it is difficult to correlate this nanometric
1217: clusters with the anomalies observed for $T<T_{\rm M^/M^{//}}$ in
1218: the stoichiometric samples because these clusters may be related
1219: with cation vacancies. It seems that in samples prepared in air
1220: the ratio $R(P_{\rm O_2},T)= \rm{Mn}^{+4}/\rm{Mn}^{+3}$ is
1221: controlled by the partial oxygen pressure. When these samples are
1222: cooled slowly during the final preparation step the $R(P_{\rm
1223: O_2},T)$ is different from the nominal Ca concentration,
1224: $R_x=x/(1-x)$.
1225: %**************
1226: In order the electric charge neutrality to be maintained, cation
1227: vacancies are created.
1228: %**************
1229: This effect is significant for $x<0.2$. Therefore, in order to
1230: produce stoichiometric samples we must use reduced atmosphere
1231: conditions or quench the sample from high temperature to liquid
1232: N$_2$.\cite{pissas04a,dabrowski99a,roosmalen94} The presence of
1233: cation vacancies completely destroys the coherent JT distortion
1234: even for the $x=0$ sample.\cite{pissas04a} In the cation deficient
1235: samples firstly, it seems that the $R= \rm{Mn}^{+4}/\rm{Mn}^{+3}$
1236: is locked to a nearly independent $x$ value, and secondly the
1237: M$^/$ phase which concerns a coherent JT distortion cannot be
1238: formed due to the cation vacancies. In this context numerous
1239: experiment results in the particular Ca regime should be
1240: interpreted. It is a mystery why the cation deficient samples show
1241: similar bulk magnetic measurements near $T=100$ K as the
1242: stoichiometric ones. The stoichiometric samples display a
1243: structural transition at $T_{\rm M^/M^{//}}$, therefore it is
1244: reasonable to correlate this anomaly with the particular shape of
1245: the curves obtained in dc magnetic
1246: and ac-susceptibility measurements. For non stoichiometric samples such
1247: structural anomaly is absent leaving open the question about the
1248: relation of the bulk magnetization anomaly among the two families
1249: of samples.
1250:
1251: Summarizing, the crystal structure refinements , from
1252: stoichiometric samples of the La$_{1-x}$Ca$_x$MnO$_3$ compound in
1253: the FMI regime, are presented. All the samples ($0.11\le x\le
1254: 0.175$) undergo a structural transition from O* to M$^/$ structure
1255: at $T_{\rm JT}$. High resolution x-ray diffraction data revealed
1256: that for $T<T_{\rm JT}$, La$_{1-x}$Ca$_x$MnO$_3$ compounds
1257: ($0.11\le x\le 0.175$) have monoclinic $P 2_1/c$ symmetry.
1258: %----------------------------------------------------------------
1259: The major structural characteristic of this model is the two Mn
1260: sites. As the bond lengths are concerned the Mn2 site displays
1261: properties similar with those of the Mn site of the LaMnO$_3$
1262: compound, while the Mn1 site, resembles the Mn site in the FMM
1263: regime. By further cooling, below the characteristic temperature
1264: $T_{\rm M^/M^{//}}$, although, the structure remains the same, a
1265: strong reduction of the strain parameter $s=(a-c)/(a+c)$ is
1266: observed. This change has been accompanied by significant
1267: reduction of the monoclinic $\gamma$ angle and is correlated with
1268: distinct changes in dc and ac magnetic measurements.
1269:
1270:
1271: \acknowledgements{This work was partial supported from EU through
1272: the CHRX-CT93-0116 access to large-scale facilities project. }
1273:
1274: \begin{references}
1275: \bibitem[*]{cor} Author to whom correspondence should be
1276: addressed. email: mpissas@ims.demokritos.gr
1277:
1278: \bibitem{khomskii97} D. Khomskii and G. Sawatzky, Solid State Comm. {\bf 102}, 87 (1997).
1279:
1280: \bibitem{pinsard97} L. Pinsard, J. Rodr\'{i}quez-Carvajal, A. H.
1281: Moudden, A. Anane, A. Revcolevschi, C. Dupas, Physica B {\bf
1282: 234-236}, 856 (1997).
1283:
1284: \bibitem{endoh99} Y. Endoh, K. Hirota, S. Ishihara, S. Okamoto, Y.
1285: Murakami, A. Nishizawa, T. Fukuda, H. Kimura, H. Nojiri, K.
1286: Kaneko, and S. Maekawa, Phys. Rev. Lett. {\bf 82},4328 (1999).
1287:
1288: \bibitem{yamada00} Y. Yamada, J. Suzuki, K. Oikawa, S. Katano, J. A.
1289: Fernandez-Baca, Phys. Rev. B {\bf 62}, 11600 (2000).
1290:
1291: \bibitem{glazer} A. M. Glazer, Acta Cryst. B {\bf 28}, 3384 (1972);
1292: A. M. Glazer, Acta Cryst. A {\bf 31}, 756 (1975);C. J. Howard and
1293: H. T. Stokes, Acta Cryst. B {\bf 54}, 782 (1998); P. M. Woodward,
1294: Acta Cryst. B{\bf 53}, 32 (1997); P. M. Woodward, Acta Cryst. B
1295: {\bf 53}, 44 (1997).
1296:
1297: \bibitem{monoclinic} As the present study reveals that the true
1298: structure below the $T_{\rm JT}$ is monoclinic, we prefer to keep
1299: the symbols O$^/$ and O$^{//}$ in order to help the reader to
1300: compare our results with those of other studies.
1301:
1302: \bibitem{kanamori60} J. Kanamori, J. Appl. Phys. {\bf 31}, 14S (1960).
1303:
1304: \bibitem{kugel82} K. I. Kugel' and D. I. Khomskii, Sov. Phys. Usp.
1305: {\bf 25}, 231 (1982).
1306:
1307: \bibitem{pissas04a} M. Pissas and G. Papavassiliou J. Phys.: Condens. Matter {\bf 16}, 6527 (2004).
1308:
1309:
1310: \bibitem{dabrowski99a} B. Dabrowski, R. Dybzinski, Z. Bukowski, and
1311: O. Chmaissem, J. Solid State. Chem. {\bf 146} 448, (1999).
1312:
1313: \bibitem{roosmalen94} J. A. M. van Roosmalen, E. H. P. Cordfuke,
1314: E. B. Helmholdt and H. W. Zandbergen, J. Solid State. Chem. {\bf
1315: 146} 448, (1999).
1316:
1317: \bibitem{fullprof} J. Rodr\'{i}quez-Carvajal, Physica B {\bf 192}, 55 (1993).
1318:
1319: \bibitem{toby} B. H. Toby, J. Appl. Cryst. {\bf 34},210 (2001).
1320:
1321: \bibitem{larson} A. C. Larson and R. B. Von Dreele, {\it General
1322: Structure Analysis System (GSAS)}, Los Alamos National Laboratory
1323: Report LAUR 86-748 (2000).
1324:
1325: \bibitem{finger94} L. W. Finger, D. E. Cox, A. P. Jephcoat, J. Appl. Cryst.
1326: {\bf 27},892 (1994).
1327:
1328: \bibitem{biotteau01} G. Biotteau, M. Hennion, F. Moussa, J. Rodr\'{i}guez-Carvajal, L. Pinsard, A. Revcolevschi,
1329: Y. M. Mukovskii, and D. Shulyatev, Phys. Rev. B {\bf 64}, 104421
1330: (2001).
1331:
1332:
1333: \bibitem{carvajal98} J. Rodr\'{i}guez-Carvajal, M. Hennion, F. Moussa,
1334: A. H. Moudden, L. Pinsard and A. Revcolevschi, Phys. Rev. B {\bf
1335: 57}, R3189, (1998).
1336:
1337:
1338: \bibitem{cox01} D. E. Cox, T. Iglesias, and E. Moshopoulou, K.
1339: Hirota, K. Takahashi, and Y. Endoh, Phys. Rev. B {\bf 64}, 024431
1340: (2001)
1341:
1342: \bibitem{stephens99} P. W. Stephens J. Appl. Cryst., {\bf 32},281 (1999).
1343:
1344:
1345: \bibitem{calderon03} M. J. Calderon, A. J. Millis, K. H. Ahn,
1346: Phys. Rev. B, {\bf 68} 100401(R) (2003).
1347:
1348: \bibitem{ahn04} K.H. Ahn, T. Lookman and A. R. Bishop, Nature,
1349: {\bf 428}, 401 (2004).
1350:
1351: \bibitem{radaelli97a} P. G. Radaelli, D. E. Cox, M. Marezio and S-W. Cheong,
1352: Phys. Rev. B {\bf 55}, 3015 (1997); P. G. Radaelli, G. Iannone, M.
1353: Marezio, H. Y. Hwang, S-W. Cheong, J. D. Jorgensen, D. N. Argyriou, Phys. Rev. B {\bf %
1354: 56}, 8265 (1997).
1355:
1356: \bibitem{nogues01} J. Nogu\'{e}īs, Vassil Skumryev, J. S. Mu\~{n}oz,
1357: B. Mart\'{i}nez, J. Fontcuberta, L. Pinsard and A. Revcolevschi,
1358: Phys. Rev. B {\bf 64}, 024434 (2001).
1359:
1360: \bibitem{nojiri99} H. Nojiri, K. Kaneko, M. Motokawa, K. Hirota, Y. Endoh,
1361: K. Takahashi, Phys. Rev. B {\bf 60}, 4142 (1999). 980-8578,
1362:
1363:
1364: \bibitem{moussa96} F. Moussa, M. Hennion, J. Rodr\'{i}guez-Carvajal, H.
1365: Moudden, L. Pinsard and A. Revcolevschi, Phys. Rev. B {\bf 54},
1366: 15149 (1996); Q. Huang, A. Santoro, J. W. Lynn, R. W. Erwin, J. A.
1367: Borchers, J. L. Peng and R. L. Greene, Phys. Rev. {\bf 55}, 14987
1368: (1997).
1369:
1370: \bibitem{markovich02a} V. Markovich, E. Rozenberg, A. I. Shames, G. Gorodetsky, I.
1371: Fita, K. Suzuki, R. Puzniak, D. A. Shulyatev and Ya. M.
1372: Mukovskii, Phys. Rev. B {\bf 65}, 144402 (2002).
1373:
1374: \bibitem{pissas04b} M. Pissas, G. Papavassiliou, E. Devlin, A.
1375: Simopoulos, V. Likodimos, cont-mat/0403446.
1376:
1377: \bibitem{papavassiliou03} G. Papavassiliou, M. Pissas, M. Belesi, M. Fardis, J.
1378: Dolinsek, C. Dimitropoulos, and J. P. Ansermet, Phys. Rev. Lett.
1379: {\bf 91}, 147205 (2003).
1380:
1381: \bibitem{lobanov00} M. V. Lobanov, A. M. Balagurov, V. Ju. Pomjakushin, P.
1382: Fischer, M. Gutmann, A. M. Abakumov, O. G. Dyachenko, E. V.
1383: Antipov, O. I. Lebedev, and G. Van Tendeloo, Phys. Rev. B,{\bf
1384: 61},8941 (2000).
1385:
1386: \bibitem{dabrowski99} B. Dabrowski, X. Xiong, Z. Bukowski, R. Dybzinski, P. W. Klamut, J. E. Siewenie,
1387: O. Chmaissem, J. Shaffer, C. W. Kimball, J. D. Jorgensen and S.
1388: Short, Phys. Rev. B {\bf 60}, 7006 (1999).
1389:
1390: \bibitem{xiong99} X. Xiong, B. Dabrowski, O. Chmaissem, Z. Bukowski, S. Kolesnik,
1391: R. Dybzinski,C. W. Kimball, and J. D. Jorgensen, Phys. Rev. B {\bf
1392: 60}, 10186 (1999).
1393:
1394: \bibitem{geck04} J. Geck, P. Wochner, D. Bruns, B. B\"{u}chner, U. Gebhardt, S. Kiele, P. Reutler, and A.
1395: Revcolevschi, Phys. Rev. B {\bf 69}, 104413 (2003).
1396:
1397: \bibitem{mizokawa00} T. Mizokawa, D.I. Khomskii, and G.A. Sawatzky, Phys. Rev. B {\bf 61},
1398: R3776 (2000).
1399:
1400:
1401: \bibitem{kilian98} R. Kilian and G. Khaliullin, Phys. Rev. B {\bf 58},
1402: 11841(R) (1998).
1403:
1404: \bibitem{huang98} Q. Huang, A. Santoro, J. W. Lynn, R. W. Erwin, J. A.
1405: Borchers, J. L. Peng, K. Ghosh and R. L. Greene, Phys. Rev. B {\bf
1406: 58}, 2684 (1998); Q. Huang, A. Santoro, J. W. Lynn, R. W. Erwin,
1407: J. A. Borchers, J. L. Peng, and R. L. Greene, Phys. Rev. B {\bf
1408: 55}, 14987 (1998); J. W. Lynn, R. W. Erwin, J. A. Borchers, Q.
1409: Huang, A. Santoro, J-L. Peng and Z. Y. Li, Phys. Rev. Lett. {\bf
1410: 76}, 4046 (1996).
1411:
1412:
1413: \bibitem{dai96} P. Dai, J Zhang, H. A. Mook, S.-H. Liou, P. A. Dowben, E.
1414: W. Plummer, Phys. Rev. B {\bf 54,} R3694, (1996).
1415:
1416: \bibitem{dai96b} P. Dai, J. Zhang, H.A. Mook, F. Foong, S.-H. Liou,
1417: P. A. Dowben and E. W. Plummer, Sol. St. Comm.{\bf 100}, 865
1418: (1996).
1419:
1420: \bibitem{algarabel03} P.A. Algarabel, J. M. De Teresa, J. Blasco,
1421: M. R. Ibarra, Cz. Kapusta, M. Sikora, D. Zajac, P. C. Riedi and C.
1422: Ritter, Phys. Rev. B {\bf 67}, 134402 (2003).
1423:
1424:
1425:
1426: \end{references}
1427:
1428:
1429:
1430: \end{document}
1431: