1:
2:
3: %% ****** Start of file template.aps ****** %
4: %%
5: %%
6: %% This file is part of the APS files in the REVTeX 4 distribution.
7: %% Version 4.0 beta 3 of REVTeX, November 30, 1999.
8: %%
9: %%
10: %% Copyright (c) 1999 The American Physical Society.
11: %%
12: %% See the REVTeX 4 README file for restrictions and more information.
13: %%
14: %
15: % This is a template for producing files for use with REVTEX 4.0 beta
16: % Copy this file to another name and then work on that file.
17: % That way, you always have this original template file to use.
18: %
19: % Group addresses by affiliation. Use superscriptaddress for long
20: % author lists or if there are many overlapping affiliations
21: % For Phys. Rev. look and feel change preprint to twocolumn
22: % showpacs will print out PACS numbers
23: \documentclass[aps,twocolumn,groupedaddress,showpacs]{revtex4}
24: %\documentclass[aps,preprint,superscriptaddress]{revtex4}
25: %\documentclass[aps,twocolumn,groupedaddress]{revtex4}
26: \usepackage{graphicx}
27: \begin{document}
28: % You should use BibTeX and revtex.bst for references
29: \bibliographystyle{apsrev}
30:
31: % Use the \preprint command to place your local institutional report
32: % number on the title page in preprint mode.
33: % Multiple \preprint commands are allowed.
34: %\preprint{}
35:
36: %Title of paper
37: \title{Kondo Lattice Behavior in the Ordered Dilute Magnetic Semiconductor Yb$_{14-x}$La$_x$MnSb$_{11}$}
38: % Optional argument for running titles on pages
39: %\title[]{}
40:
41: % repeat the \author .. \affiliation etc. as needed
42: % \email, \thanks, \homepage, \altaffiliation all apply to the current
43: % author. Explanatory text should go in the []'s, actual e-mail
44: % address or url should go in the {}'s for \email and \homepage.
45: % Please use the appropriate macro for the type of information
46:
47: % \affiliation command applies to all authors since the last
48: % \affiliation command. The \affiliation command MUST follow the
49: % other information
50: \author{B.C. Sales$^1$}
51: \author{P. Khalifah$^{1,2}$}
52: \author{T. P. Enck$^1$}
53: \author{E. J. Nagler$^1$}
54: \author{R. E. Sykora$^3$}
55: \author{R. Jin$^1$}
56: \author{D. Mandrus$^1$}
57:
58: \affiliation{$^1$Condensed Matter Sciences Division, Oak Ridge
59: National Laboratory, Oak Ridge, TN 37831}
60: \affiliation{$^2$Chemistry Department, University of
61: Massachusetts, Amherst, MA 01003}
62: \affiliation{$^3$Chemical
63: Sciences Division, Oak Ridge National Laboratory, Oak Ridge, TN
64: 37831}
65: %Collaboration name if desired (requires use of superscriptaddress
66: %option in \documentclass). \noaffiliation is required (may also be
67: %used with the \author command).
68: %\collaboration{}
69: %\noaffiliation
70:
71: \date{\today}
72:
73:
74:
75: \begin{abstract}
76: We report Hall, magnetic, heat capacity and doping studies from
77: single crystals of Yb$_{14}$MnSb$_{11}$ and
78: Yb$_{13.3}$La$_{0.7}$MnSb$_{11}$. These heavily doped
79: semiconducting compounds are ferromagnetic below 53 K and 39 K
80: respectively. The renormalization of the carrier mass from 2m$_e$
81: near room temperature to 15m$_e$ at 5 K , plus the magnetic
82: evidence for partial screening of the Mn magnetic moments suggest
83: that these compounds represent rare examples of an underscreened
84: Kondo lattice with T$_K \approx$ 350 K.
85: \end{abstract}
86: % insert suggested PACS numbers in braces on next line
87: \pacs{75.30.Mb, 72.15.Qm, 75.50.Cc}
88: %\maketitle must follow title, authors, abstract and PACS
89: \maketitle
90:
91: Ferromagnetic semiconductors are envisioned as a key component of
92: many proposed spintronic devices that functionalize both the
93: electron charge and spin.\cite{1,2} There have been many theories
94: of the origin of ferromagnetism in doped semiconductors,\cite{3}
95: but it has been difficult to obtain a clean comparison between
96: theory and experiment because of problems associated with
97: clustering or phase separation of magnetic dopants in the
98: semiconducting host. In the present article we show that it is
99: productive to investigate the ferromagnetism in stoichiometric
100: compounds such as Yb$_{14}$MnSb$_{11}$ (T$_c$=53K). In this
101: heavily doped semiconducting compound the magnetic Mn atoms are at
102: a unique crystallographic site in the structure resulting in a
103: minimum Mn-Mn separation of 10 \AA. Using a combination of Hall,
104: heat capacity, magnetic data, and doping studies we are able to
105: show that the ferromagnetism is likely mediated by relatively
106: heavy quasiparticles that form due to the Kondo interaction. This
107: leads to the non-intuitive and surprising result that using
108: carrier tuning to increase T$_c$ can result in a reduced
109: saturation moment. Since the local environment of each Mn atom is
110: similar to that found in the heavily studied III-V semiconductors
111: \cite{4} (such as GaAs:Mn), the present results may also have
112: significant implications for III-V materials as well.
113:
114: The compound Yb$_{14}$MnSb$_{11}$ was first synthesized by Chan
115: $\it {et}$ $\it {al.}$ in 1998.\cite{5} It is isostructural with a
116: large family of related Zintl compounds such as
117: Ca$_{14}$AlSb$_{11}$ and Ca$_{14}$MnSb$_{11}$\cite{6}
118: crystallizing with a tetragonal lattice in the space group $\it
119: {I41/acd}$. The structure of Yb$_{14}$MnSb$_{11}$ is shown in
120: Fig.\ 1, where for clarity only the Mn and four nearest neighbor
121: Sb atoms are shown. Previous magnetization and X-ray Magnetic
122: Circular Dichroism (XMCD) measurements indicate that Yb is
123: divalent in this compound with no magnetic moment.\cite{5,7}
124: Initial magnetization and magnetic susceptibility measurements
125: suggested a Mn$^{+3}$ ($\it {d}$$^4$) configuration.\cite{5,8} The
126: MnSb$_4$ tetrahedron is somewhat distorted with angles of
127: 105.6$^\circ$ and 117.5$^\circ$ which were previously interpreted
128: as a Jahn-Teller distortion associated with a d$^4$
129: configuration.\cite{5} However, more recent XMCD measurements are
130: more consistent with a Mn$^{+2}$ (d$^5$) configuration with the
131: moment of one spin compensated by the antialligned spin of an Sb
132: 5p hole from the Sb$_4$ cage surrounding the Mn.\cite{7} There is
133: also a substantial distortion of the AlSb$_4$ tetrahedra in the
134: isostructural compound Ca$_{14}$AlSb$_{11}$ implying that steric
135: effects cause distortions even in the absence of Jahn-Teller
136: effects.\cite{9} A d$^5$+hole (d$^5$+h) picture is also expected
137: from detailed electronic structure calculations on related
138: Ca$_{14}$MnBi$_{11}$ compound.\cite{10} This is in accord with the
139: understanding of the most heavily studied dilute magnetic
140: semiconductor (DMS) GaAs:Mn which is suggested by multiple
141: experiments to have also a d$^5$+h Mn configuration.\cite{2,3,11}
142:
143: We grew relatively large (up to 0.5 g) single crystals of
144: Yb$_{14}$MnSb$_{11}$ with and without various dopants (La, Te, Y,
145: Sc etc.) capable of modifying the carrier concentration using the
146: conditions worked out by Fisher $\it {et}$ $\it {al.}$\cite{8} for
147: pure Yb$_{14}$MnSb$_{11}$. The partial substitution of La for Yb
148: had the largest effect on T$_c$ and the hole carrier concentration
149: and hence of the doped samples we investigated these crystals in
150: the most detail. A single crystal structure refinement of a La
151: doped crystal (Bruker SMART APEX CCD X-ray diffractometer)
152: indicates a composition of Yb$_{13.3}$La$_{0.7}$MnSb$_{11}$ with
153: all of the La atoms substituting on an Yb site (a = 16.6613(6)
154: \AA, c = 21.9894(7) \AA). This is expected since La$^{+3}$ and
155: Yb$^{+2}$ ions are about the same size and are chemically similar.
156: Hall effect, resistivity, and heat capacity data were obtained
157: using a Physical Property Measurement System (PPMS) from Quantum
158: Design on oriented and thinned single crystals. Hall and
159: resistivity data were obtained using a standard six lead method
160: and either rotating the sample by 180$^\circ$ in a fixed magnetic
161: field or by sweeping the direction of the field from positive to
162: negative values. The Hall data were qualitatively the same
163: regardless of the orientation of the crystal with respect to the
164: current or field directions. Magnetization data were obtained
165: using a commercial SQUID magnetometer from Quantum Design.
166:
167: Magnetic data for single crystals of Yb$_{14}$MnSb$_{11}$ and
168: Yb$_{13.3}$La$_{0.7}$MnSb$_{11}$ are shown in Fig.\ 2. The
169: partial substitution of La for Yb lowers T$_c$ by about 14 K, but
170: results in an increase of the saturation moment from $\sim$
171: 4.2$\mu_B$ to 4.5$\mu_B$ per Mn.
172:
173: \begin{figure}
174: \includegraphics[keepaspectratio=true, totalheight =3.0 in, width =
175: 2.5 in]{Fig1.eps}
176: \caption{Structure of the tetragonal compound Yb$_{14}$MnSb$_{11}$ with $\it {a}$ = 16.61 \AA
177: and $\it {c}$ = 21.95 \AA. For clarity only the Mn atoms (red) and
178: the nearest neighbor Sb atoms (black) are shown. The closest Mn-Mn
179: distance is 9.98 \AA. }
180: \end{figure}
181:
182: Heat capacity data from both materials are shown in Fig.\ 3. The
183: transitions to the ferromagnetic phases are evident at T$_c$ = 53
184: K (Yb$_{14}$MnSb$_{11}$ crystal) and T$_c$=39 K
185: (Yb$_{13.3}$La$_{0.7}$MnSb$_{11}$ crystal). The small
186: concentration of magnetic Mn atoms, and the relatively large
187: lattice contribution to the heat capacity data near T$_c$ make it
188: difficult to accurately determine the magnetic contribution to the
189: heat capacity data below T$_c$. The electronic contribution to the
190: heat capacity, $\gamma$T, can be reliably estimated, however by
191: using the heat capacity data from the lowest temperatures (T $\le$
192: 5 K) and plotting C/T vs. T$^2$. This analysis yields $\gamma$
193: $\sim$ 58 mJ/K$^2$-mole Mn for the La doped crystal and $\gamma$
194: $\sim$ 120 mJ/K$^2$-mole Mn for the undoped crystal. The value of
195: $\gamma$ for the undoped crystal is very similar to the values
196: previously measured by Fisher $\it {et}$ $\it {al.}$\cite{8} and
197: Burch $\it {et}$ $\it {al.}$\cite{12} However, to determine the
198: significance of $\gamma$ requires a measure of the carrier
199: concentration and an estimate of the expected band mass.
200:
201: \begin{figure}
202: \includegraphics[keepaspectratio=true, totalheight =3.0 in, width =
203: 2.5 in]{Fig2.eps} \caption{Normalized magnetization (left scale)
204: vs. temperature for crystals cooled in a field of 100 Oe. The
205: Curie temperature of the Yb$_{14}$MnSb$_{11}$ crystal is T$_c$ =
206: 53 $\pm$ 1 K, a value consistent with values previously
207: reported.\cite{1,7} Magnetization (right scale) vs. applied
208: magnetic field for Yb$_{14}$MnSb$_{11}$ and
209: Yb$_{13.3}$La$_{0.7}$MnSb$_{11}$ single crystals. Although the
210: T$_c$ of the La doped crystal is lower, the saturation
211: magnetization is about 4.5$\mu_B$ per Mn as compared to about
212: 4.2$\mu_B$ per Mn for the undoped crystals.}
213: \end{figure}
214:
215: The carrier concentrations for the crystals are determined from a
216: detailed analysis of Hall data that takes into account the
217: anomalous Hall effect (AHE) associated with the ferromagnetism.
218: For many ferromagnets it has been experimentally
219: demonstrated\cite{13} that the Hall resistivity, $\rho_{xy} =
220: R_0B + R_sM$, where R$_0$ is the "normal" Hall coefficient, B is
221: the magnetic field, M is the magnetization and R$_s$ is the
222: "anomalous" Hall coefficient. In the simplest situations R$_0$ is
223: proportional to 1/n, where n is the effective carrier
224: concentration. There are a variety of theories and models for the
225: AHE, which can be viewed as an additional current that develops in
226: the y direction in response to an electric field in the x
227: direction (magnetization in z direction).\cite{14,15,16} There can
228: be both intrinsic ($\rho_{xy} \propto \rho_{xx}^2$) and extrinsic
229: contributions ( $\rho_{xy} \propto \rho_{xx}$) to the AHE that are
230: usually separated based on how $\rho_{xy}$ depends on the normal
231: resistivity, $\rho_{xx}$.\cite{13} Interest in spin currents for
232: spin based electronics, coupled with new theoretical insights into
233: the microscopic origin of R$_s$, have stimulated substantial
234: recent interest in the AHE. While the AHE of Yb$_{14}$MnSb$_{11}$
235: and related alloys are quite interesting, these data will be
236: discussed in more detail in a future publication. The measured
237: Hall resistivity for Yb$_{14}$MnSb$_{11}$ is shown in Fig.\ 4.
238: Analysis of these data (as described in the figure caption)
239: indicates that the carrier concentration is about 1.0 $\times$
240: 10$^{21}$ to 1.35 $\times$ 10$^{21}$ holes/cm$^3$ above T$_c$,
241: depending on how the data are analyzed, but increasing to 1.9
242: $\times$ 10$^{21}$ holes/cm$^3$ for temperatures well below T$_c$.
243: All of the measured carrier concentrations are typical of a
244: heavily doped semiconductor and the values correspond to
245: approximately 1 hole/Mn (1 hole per Mn gives 1.3 $\times$
246: 10$^{21}$ holes/cm$^3$). Doping the crystals with a small amount
247: of La ($\it {i.e.}$ Yb$_{13.3}$La$_{0.7}$MnSb$_{11}$) lowers the
248: measured hole concentration to about 4 $\times$ 10$^{20}$
249: holes/cm$^3$. This observation is consistent with the simple idea
250: that the extra electrons contributed to the compound when part of
251: the Yb$^{+2}$ ions are replaced by La$^{+3}$ ions fill some of the
252: holes in the Sb 5p bands. A crude estimate for the band gap of
253: $\sim$ 1 eV is obtained from the optical data of Burch $\it {et}$
254: $\it {al.}$\cite{12} on Yb$_{14}$MnSb$_{11}$, but it is not clear
255: if the carrier concentration in these materials can be reduced
256: enough to observe a gap with electrical transport measurements.
257:
258: \begin{figure}
259: \includegraphics[keepaspectratio=true, totalheight =4.0 in, width =
260: 3.0 in]{Fig3.eps} \caption{Heat capacity divided by temperature
261: (C/T) vs. temperature (T) for single crystals of
262: Yb$_{14}$MnSb$_{11}$ and Yb$_{13.3}$La$_{0.7}$MnSb$_{11}$. The
263: ferromagnetic phase transition for each crystal is noted. Analysis
264: of these data at low temperatures yield values for the electronic
265: heat coefficient of $\gamma$ = 120 $\pm$ 10 mJ/K$^2$-mole Mn, and
266: $\gamma$ = 58 $\pm$ 5 mJ/K$^2$-mole Mn for the undoped and
267: La-doped crystals, respectively. Heat capacity data on both
268: crystals were taken down to 0.5 K and in magnetic fields up to 14
269: T (not shown). The data were corrected for a small nuclear
270: Schottky contribution that was significant for temperatures below
271: 1 K}
272: \end{figure}
273:
274: The average Fermi wavevector, k$_F$, only depends on the number of
275: carriers per unit volume and is given by k$_F =(3\pi^2n)^{1/3}
276: \sim$ 0.36 $\AA^{-1}$. The Fermi energy, E$_F$, from the free
277: electron model is given by h$^2k_F^2/(4\pi^2m_e)$ which gives
278: E$_F$ = 0.5 eV. If the band mass, m$_b \approx 1.8m_e$, is used
279: (as measured from optical data\cite{12} or estimated from
280: electronic structure calculations,\cite{10} or estimated\cite{17}
281: from the room temperature Seebeck coefficient of +44 $\mu$V/K)
282: E$_F \sim$ 0.28 eV. The measured values for the carrier
283: concentration indicate a substantial mass enhancement for the
284: carriers of about 15m$_e$ for the undoped crystal and 11m$_e$ for
285: the La doped crystal. These mass enhancement values suggest a
286: Kondo temperature T$_K \sim$ 370 K. Similar mass enhancement
287: values and estimates of the Kondo temperature were obtained by
288: Burch $\it {et}$ $\it { al.}$\cite{12} for the undoped crystal
289: using optical and heat capacity data. In addition, Burch $\it
290: {et}$ $\it {al.}$\cite{12} were able to map the evolution of the
291: enhanced carrier mass as both the temperature and frequency were
292: lowered toward zero. They also noted that the optical data were
293: qualitatively similar to that found in many Kondo lattice
294: compounds.
295:
296: \begin{figure}
297: \includegraphics[keepaspectratio=true, totalheight =3.5 in, width =
298: 3.2 in]{Fig4.eps} \caption{Hall resistivity ($\rho_{xy}$) vs.
299: temperature for Yb$_{14}$MnSb$_{11}$ measured in a field of 7
300: Tesla. The carrier concentration is estimated by fitting the data
301: above T$_c \sim$ 53 K to a constant plus a Curie-Weiss Law.
302: Similar data (not shown) were also taken with a field of 1 Tesla.
303: This analysis yields an average value for the carrier
304: concentration above T$_c$ of 1.35 $\times 10^{21}$ holes/cm$^3$.
305: Comparable values for the carrier concentration are obtained if
306: the measured magnetization data are used. If the Hall resistivity
307: above T$_c$ is analyzed using the expression derived for Kondo
308: lattice compounds above the coherence temperature\cite{22} (the
309: anomalous part of the Hall resistivity R$_s$M is replaced by
310: C$\rho_{xx}$M, where C is a constant) the data yields a
311: temperature independent (75 K $\le T \le$ 300K) carrier
312: concentration of 1.03 $\pm$ 0.05 $\times$ 10$^{21}$ holes/cm$^3$.
313: The carrier concentration well below T$_c$ is estimated by
314: plotting $\rho_{xy}$/B vs M/B using the measured magnetization
315: data as shown in the inset. This analysis yields a carrier
316: concentration of 1.9 $\times$ 10$^{21}$ holes/cm$^3$. It appears
317: that from just above T$_c$ (T $\approx$ 75 K) to well below T$_c$,
318: the carrier concentration jumps by about 0.9 $\times$ 10$^{21}$
319: holes cm$^{-3}$.}
320: \end{figure}
321:
322: \begin{figure}
323: \includegraphics[keepaspectratio=true, totalheight =3.0 in, width =
324: 3.0 in]{Fig5.eps} \caption{Effective magnetic moment versus
325: temperature extracted from DC susceptibility data taken in a field
326: of 1 T. The Curie-Weiss temperature, T$_{CW}$, was fixed for each
327: material and the data was fit to $\chi$(cm$^3$/mole Mn) = $\chi_0$
328: + C/(T-T$_{CW})$ over 50 K intervals, with
329: $\mu_{eff}$=(8C)$^{1/2}$.}
330: \end{figure}
331:
332:
333: The basic Kondo interaction describes the antiferromagnetic
334: interaction between a localized spin (in this case a Mn d$^5$
335: configuration) and the mobile carriers, which for
336: Yb$_{14}$MnSb$_{11}$ would be 5p holes from the Sb bands. A
337: material with a regular array of these spins (or Kondo centers)
338: is referred to as a Kondo lattice compound. While the Kondo
339: interaction is often associated with the formation of a
340: non-magnetic singlet and a large enhancement of the carrier mass,
341: this same interaction is also responsible for RKKY (carrier
342: mediated exchange) coupling between spins which can result in
343: magnetic order.\cite{18} In rare cases, such as CePdSb, both
344: effects can occur which results in ferromagnetic order but with a
345: reduced saturation moment.\cite{19} Dietl $\it {et}$ $\it {
346: al.}$\cite{11} have shown that in Mn doped III-V semiconductors
347: there is a strong tendency for the Kondo interaction to bind holes
348: from neighboring group V $\it {p}$ bands to the $\it {d}$ shell of
349: the Mn$^{+2}$ impurity resulting in a composite object with a net
350: magnetic moment of 4$\mu_B$. The binding energy of the holes in
351: Yb$_{14}$MnSb$_{11}$ is relatively weak and is roughly given by
352: T$_K \approx$ 370 K. The standard estimate for the size of the
353: Kondo compensation cloud is hv$_F$/2$\pi$k$_B$T$_K$ $\approx$ 40
354: \AA, which also indicates weak binding. At temperatures near T$_K$
355: the composite quasiparticles should begin to break up with less
356: compensation of the Mn $\it {d}$$^5$ moments by the 5$\it {p}$
357: holes. A careful analysis of the magnetic susceptibility data from
358: Yb$_{14}$MnSb$_{11}$ (Fig.\ 5) indicates an increase of the
359: effective moment for temperatures near T$_K$. The experimental
360: data from the doped crystals are also consistent with this
361: picture. The replacement of Yb by La reduces the number of
362: carriers and weakens the indirect exchange coupling resulting in a
363: lower T$_c$. The lower number of holes also reduces the magnetic
364: screening of the Mn $\it {d}$$^5$ moments which results in a
365: larger saturation moment and a larger effective magnetic moment,
366: as is observed. Our work suggests that the unusual $\it {d}$$^5$+h
367: magnetism of Mn atoms may be intrinsic to many dilute magnetic
368: semiconductors. This leads to the non-intuitive and surprising
369: result that using carrier tuning to increase T$_c$ can result in a
370: reduced saturation moment. The present data also support the idea,
371: proposed by Burch $\it {et}$ $\it {al.}$,\cite{12} that
372: Yb$_{14}$MnSb$_{11}$ is an underscreened Kondo lattice compound. A
373: general characteristic of these compounds is that part of the
374: entropy associated with the local magnetic moment (Mn spins) is
375: removed by a transfer of part of the local $\it {d}$ character to
376: the conduction band.\cite{20,21} This transfer begins near the
377: coherence temperature, T$^*$, and results in a large
378: renormalization of the mass of quasiparticles at the Fermi energy
379: and modification of the Fermi surface below T$^*$. Burch $\it
380: {et}$ $\it {al.}$\cite{12} estimated that T$^*$ is near the
381: ferromagnetic transition temperature (T$^* \approx$ 50 K). This is
382: consistent with the present Hall data that indicates a jump in the
383: carrier concentration (which is a measure of the topology of the
384: Fermi surface) from about 1.2 $\times$ 10$^{21}$ holes cm$^{-3}$
385: above T$_c$ to 1.9 $\times$ 10$^{21}$ holes cm$^{-3}$ well below
386: T$_c$. Further studies of these materials are clearly needed,
387: however, to substantiate the Kondo lattice interpretation of the
388: present data. We also believe that further investigations of
389: natural DMS, such as Yb$_{14}$MnSb$_{11}$, will provide additional
390: insights into to the complex physics of carrier mediated magnetism
391: that will be essential to the understanding and design of actual
392: spintronic devices.
393:
394:
395: % If in twocolumn mode, this environment will move to single column
396: % format so that long equations can be displayed. Use
397: % sparingly.
398: % Note: this may cause bad behavior if this occurs near a page
399: % break - this can be worked around by adding an explicit \pagebreak.
400: %\begin{widetext}
401: % put long equation here
402: %\end{widetext}
403:
404: % If you have acknowledgments this puts in the proper section
405: \begin{acknowledgments}
406: It is a pleasure to acknowledge enlightening discussions with
407: Kenneth Burch, David Singh, Victor Barzykin, Igor Zutic, Thomas
408: Schulthess and Steve Nagler. Oak Ridge National Laboratory is
409: managed by UT-Battelle LLC, for the U.S. Department of Energy
410: under contract DE-AC05-00OR22725.
411: \end{acknowledgments}
412:
413: %references
414:
415: \begin{references}
416:
417: \bibitem{1} S.A. Wolf $\it{et}$ $\it {al.}$, Science {\bf 294}, 1488 (2001).
418:
419: \bibitem{2} I. Zutic, J. Fabian, and Das Sarma, Rev. Mod. Phys. {\bf 76}, 323 (2004).
420:
421: \bibitem{3} A.H. MacDonald, P. Schiffer, N. Samarth, Nature Materials {\bf 4},
422: 195 (2005) and references therein.
423:
424: \bibitem{4} T. Dietl, and H. Ohno, Mat. Res. Soc. Bull. {\bf 28}, 714 (2003).
425:
426: \bibitem{5} J.Y. Chan, M.M. Olmstead, S.M. Kauzlarich, and D.J. Webb, Chem. Mat. {\bf 10}, 3583 (1998).
427:
428: \bibitem{6} S.M. Kauzlarich, $\it {Chemistry,}$ $\it {Structure}$ $\it {and}$ $\it {Bonding}$ $\it {of}$ $\it {Zintl}$ $\it {Phases}$ $\it {and}$ $\it {Ions}$, p245 (Ed. S.M. Kauzlarich; VCH: New York, 1996).
429:
430: \bibitem{7} A.P. Holm $\it {et}$ $\it {al.}$, J. Amer. Chem. Soc. {\bf 124}, 9894 (2002).
431:
432: \bibitem{8} I.R. Fisher $\it {et}$ $\it {al.}$, Phys. Rev. B {\bf 59}, 13829 (1998).
433:
434: \bibitem{9} A. Rehr $\it {et}$ $\it {al.}$, Chem. Mat. {\bf 6}, 93(1994).
435:
436: \bibitem{10} D. Sanchez-Portal, R.M. Martin, S.M. Kauzlarich, and W.E. Pickett, Phys. Rev. B {\bf 65}, 144414 (2002).
437:
438: \bibitem{11} T. Dietl, F. Matsukura, H. Ohno, Phys. Rev. B {\bf 66}, 33203 (2002).
439:
440: \bibitem{12} K.S. Burch $\it {et}$ $\it {al.}$, Phys. Rev. Lett. {\bf 95}, 46401 (2005).
441:
442: \bibitem{13} L. Berger, G. Bergmann, $\it {The}$ $\it {Hall}$ $\it {Effect}$ $\it {and}$ $\it {Its}$ $\it {Applications}$, p55 (Eds. C. L. Chien and C. R. Westgate, Plenum, New York, 1980).
443:
444: \bibitem{14} T. Jungwirth, Q. Niu, and A.H. MacDonald, Phys. Rev. Lett. {\bf
445: 88}, 207208 (2002).
446:
447: \bibitem{15} Y. Taguchi $\it {et}$ $\it {al.}$, Science {\bf 291}, 2573 (2001).
448:
449: \bibitem{16} W.L. Lee, S. Watauchi, V.L. Miller, R.J. Cava, and N.P. Ong, Science {\bf 303}, 1647 (2004).
450:
451: \bibitem{17} B.C. Sales (unpublished).
452:
453: \bibitem{18} P. Fazekas, Lecture Notes on $\it {Electron}$ $\it {Correlation}$ $\it {and}$ $\it {Magnetism}$, p650 (World Scientific, London,
454: 1999).
455:
456: \bibitem{19} S.K. Malik, and D.T. Adroja, Phys. Rev. B {\bf 43}, R6295 (1991).
457:
458: \bibitem{20} S. Nakatsuji, D. Pines, Z. Fisk, Phys. Rev. Lett. {\bf 92}, 16401 (2004).
459:
460: \bibitem{21} N.J. Curro, B.L. Young, J. Schmalian, D. Pines, Phys. Rev. B {\bf 70}, 235117 (2004).
461:
462: \bibitem{22} A. Fert, P.M. Levy, Phys. Rev. B {\bf 36}, 1907 (1987).
463:
464: \end{references}
465:
466: % figures follow here or may be put into the text as floats.
467: % Use the graphics or
468: % graphicx packages distributed with LaTeX2e. See the LaTeX Graphics
469: % Companion by Michel Goosens, Sebastian Rahtz, and Frank Mittelbach
470: % for instance.
471: %
472: % Here is an example of the general form of a figure:
473: % Fill in the caption in the braces of the \caption{} command. Put the label
474: % that you will use with \ref{} command in the braces of the \label{} command.
475: %
476: % \begin{figure}
477: % \label{}
478: % \includegraphics[]{}
479: % \caption{}
480: % \end{figure}
481:
482: % tables follow here or maybe be put in the text
483: %
484: % Here is an example of the general form of a table:
485: % Fill in the caption in the braces of the \caption{} command. Put the label
486: % that you will use with \ref{} command in the braces of the \label{} command.
487: % Insert the column specifiers (l, r, c, d, etc.) in the empty braces of the
488: % \begin{tabular}{} command.
489: %
490: % \begin{table}
491: % \label{}
492: % \caption{}
493: % \begin{tabular}{}
494: % \end{tabular}
495: % \end{table}
496:
497: \end{document}
498: %
499: % ****** End of file template.aps ******
500: