Exchange broadening of EPR line in ZnO:Co

We study the X-band EPR spectra of Co²⁺ in single crystalline Zn₁₋xCoxO (x = 0.001–0.075) thin films grown by plasma-assisted molecular beam epitaxy. By analyzing the EPR linewidth behavior we argue that the exchange-narrowing model, usually applied to Mn-based II-VI DMS, fails here and that a combi...

Full description

Saved in:
Bibliographic Details
Date:2007
Main Authors: Sati, P., Pashchenko, V., Stepanov, A.
Format: Article
Language:English
Published: Фізико-технічний інститут низьких температур ім. Б.І. Вєркіна НАН України 2007
Online Access:https://nasplib.isofts.kiev.ua/handle/123456789/7713
Tags: Add Tag
No Tags, Be the first to tag this record!
Journal Title:Digital Library of Periodicals of National Academy of Sciences of Ukraine
Cite this:Exchange broadening of EPR line in ZnO:Co / P. Sati, V. Pashchenko, A. Stepanov // Физика низких температур. — 2007. — Т. 33, № 11. — С. 1222-1226. — Бібліогр.: 33 назв. — англ.

Institution

Digital Library of Periodicals of National Academy of Sciences of Ukraine
id nasplib_isofts_kiev_ua-123456789-7713
record_format dspace
spelling Sati, P.
Pashchenko, V.
Stepanov, A.
2010-04-08T12:14:29Z
2010-04-08T12:14:29Z
2007
Exchange broadening of EPR line in ZnO:Co / P. Sati, V. Pashchenko, A. Stepanov // Физика низких температур. — 2007. — Т. 33, № 11. — С. 1222-1226. — Бібліогр.: 33 назв. — англ.
0132-6414
PACS: 76.30.Fc
PACS: 76.30.Fc
https://nasplib.isofts.kiev.ua/handle/123456789/7713
We study the X-band EPR spectra of Co²⁺ in single crystalline Zn₁₋xCoxO (x = 0.001–0.075) thin films grown by plasma-assisted molecular beam epitaxy. By analyzing the EPR linewidth behavior we argue that the exchange-narrowing model, usually applied to Mn-based II-VI DMS, fails here and that a combined effect of exchange and dipolar broadening can explain the linewidth variation with Co content and temperature.
en
Фізико-технічний інститут низьких температур ім. Б.І. Вєркіна НАН України
Exchange broadening of EPR line in ZnO:Co
Article
published earlier
institution Digital Library of Periodicals of National Academy of Sciences of Ukraine
collection DSpace DC
title Exchange broadening of EPR line in ZnO:Co
spellingShingle Exchange broadening of EPR line in ZnO:Co
Sati, P.
Pashchenko, V.
Stepanov, A.
title_short Exchange broadening of EPR line in ZnO:Co
title_full Exchange broadening of EPR line in ZnO:Co
title_fullStr Exchange broadening of EPR line in ZnO:Co
title_full_unstemmed Exchange broadening of EPR line in ZnO:Co
title_sort exchange broadening of epr line in zno:co
author Sati, P.
Pashchenko, V.
Stepanov, A.
author_facet Sati, P.
Pashchenko, V.
Stepanov, A.
publishDate 2007
language English
publisher Фізико-технічний інститут низьких температур ім. Б.І. Вєркіна НАН України
format Article
description We study the X-band EPR spectra of Co²⁺ in single crystalline Zn₁₋xCoxO (x = 0.001–0.075) thin films grown by plasma-assisted molecular beam epitaxy. By analyzing the EPR linewidth behavior we argue that the exchange-narrowing model, usually applied to Mn-based II-VI DMS, fails here and that a combined effect of exchange and dipolar broadening can explain the linewidth variation with Co content and temperature.
issn 0132-6414
url https://nasplib.isofts.kiev.ua/handle/123456789/7713
citation_txt Exchange broadening of EPR line in ZnO:Co / P. Sati, V. Pashchenko, A. Stepanov // Физика низких температур. — 2007. — Т. 33, № 11. — С. 1222-1226. — Бібліогр.: 33 назв. — англ.
work_keys_str_mv AT satip exchangebroadeningofeprlineinznoco
AT pashchenkov exchangebroadeningofeprlineinznoco
AT stepanova exchangebroadeningofeprlineinznoco
first_indexed 2025-11-25T21:12:24Z
last_indexed 2025-11-25T21:12:24Z
_version_ 1850549295981264896
fulltext Fizika Nizkikh Temperatur, 2007, v. 33, No. 11, p. 1222–1226 Exchange broadening of EPR line in ZnO:Co P. Sati1, V. Pashchenko2, and A. Stepanov1 1 Laboratoire Matériaux et Microélectronique de Provence, Case-142 Université Paul Cézanne and CNRS, 13397, Marseille Cedex 20, France E-mail: anatoli.stepanov@L2MP.fr 2 Physikalisches Institut, Johann Wolfgang Goethe Universität, 1 Max-von-Laue-Strasse Frankfurt am Main 60438 , Germany Received March 3, 2007 We study the X -band EPR spectra of Co2 � in single crystalline Zn1�xCoxO (x = 0.001–0.075) thin films grown by plasma-assisted molecular beam epitaxy. By analyzing the EPR linewidth behavior we argue that the exchange-narrowing model, usually applied to Mn-based II-VI DMS, fails here and that a combined effect of exchange and dipolar broadening can explain the linewidth variation with Co content and temperature. PACS: 76.30.Fc Iron group (3d) ions and impurities (Ti–Cu). Keywords: electron paramagnetic resonance, diluted magnetic semiconductors. Diluted magnetic semiconductors (DMS), the mag- netic properties of which are due to the substitution of cations by transition-metal (TM) ions, have become a fo- cus of considerable interest in recent years as essential materials for practical semiconductor spintronic devices such as spin filters [1] or spin polarizers [2]. The theoreti- cal predictions based on the local spin density approxima- tion (LSDA), triggered extensive studies of ZnO:TM al- loys with a special focus on ZnO:Co as the most promising candidate for a room-temperature ferromag- netic (FM) semiconductor [3]. Many experiments have been reported on this material fabricated by a variety of methods [4–6]; however, the magnetic properties of ZnO:Co still remain a controversial issue since the ob- served magnetic behavior appears to be strongly depend- ent on the preparation methods and is poorly reproduc- ible. Ferromagnetism was reported for thin films and bulk samples of ZnO:Co with a very large spread of spontane- ous magnetic moment from 6.1 � B /Co to 0.01 � B /Co ac- companied by a Curie temperature well above room tem- perature [4–9]. At the same time the absence of ferromagnetism and paramagnetic behavior down to helium temperatures in ZnO:Co were claimed by many authors [10–16]. On the theoretical side, LSDA has difficulties when applied to the magnetic state of TM-doped ZnO, since it does not account for correlations between d-electrons and leads to a semimetallic FM ground state. Quite surpris- ingly, an improved version of LSDA, LSDA + U also leads to controversial results as regards the exchange con- stant sign between Co 2 � ions in ZnO. Indeed, in their re- cent paper, Chanier et al. [17] show that the exchange constants, J out and J in , between nearest-neighbor (NN) Co ions in the ZnO wurtzite structure are both negative (AFM) and have the values �9 K and �21 K, respectively. In contrast, Lee and Chang [18] and Sluiter et al. [19] have detected a competition between FM J out and AFM J in in Co-doped ZnO. The EPR offers an interesting alternative to traditional magnetic and optical methods providing complementary information on the electronic properties of transition met- als ions. For example, EPR is able to probe the exchange interactions between TM ions by studying the well known phenomenon of the exchange narrowing of an EPR line. In this case a broadened line, say, by the magnetic dipole interaction, �d , is narrowed by a fast dynamics resulting from the exchange coupling, J . This gives finally a linewidth �H /Jd� � 2 [20]. Interestingly, the exchange narrowing picture explains the linewidth of various Mn-based II-VI DMS at high temperatures [21]. It was also established, that at low temperatures these DMS ex- hibit, quite universally, a critical line broadening — a phenomenon which is also driven by the exchange cou- pling [22–24]. More recently, the exchange-narrowing model was confirmed for Zn 1�xMn xO thin films having various Mn contents [25,26]. To our knowledge, no © P. Sati, V. Pashchenko, and A. Stepanov, 2007 detailed EPR investigations have been yet performed on Zn 1�xCo xO. In this paper we report the X -band EPR experiments on Zn 1�xCo xO single crystalline thin films, performed in a wide range of temperature and Co content, which were undertaken in order to gain more insight into the role played by the magnetic Co–Co interaction in ZnO:Co. Zn 1�xCo xO thin films with x = 0.001–0.075 were grown on sapphire substrates by plasma-assisted MBE and had thicknesses of about 1�m, the c-axis of the wurt- zite structure being perpendicular to the film plane. The conductivity of the films was n-type, with residual carrier concentrations ne � 10 18 cm �3. We first discuss the magnetic properties of the studied ZnO:Co films. The data were obtained using a commer- cial SQUID magnetometer (Quantum Design MPMS) in magnetic fields up to 50 kOe and in the temperature range 2 300� K. A pure ZnO film on a sapphire substrate and a sample holder were examined separately and their signals were subsequently subtracted from the total magnetic mo- ment. The Co content x of the studied samples was deter- mined by energy dispersive x-ray (EDX) microanalysis. For the lowest concentration (x � 0 005. ), the x value was determined by magnetic measurements. In Fig. 1 we show the temperature dependence of the inverse static magnetic susceptibility, � �1( )T , measured at H �10 kOe and H c for x � 0 052. . A linear increase of � �1( )T at higher temperatures can be fitted to the Curie–Weiss law � � � �1( ) ( ( )) / ( )T T x C x , with the usual notation for ( )x and C x( ). Note that the curve in Fig. 1 is representative of many experiments in two respects, namely, i) for all measured samples the sign of is found to be negative, ii) the continuous increase of with x is also observed. The inset of Fig. 1 shows the magnetization of the Zn1�xCoxO film with x � 0 052. as a function of magnetic field at T � 2 K. As expected from the results on weakly Co-doped ZnO films [27], M H( ) curves reveal a consid- erable magnetic anisotropy of Co 2+ in the wurtzite lattice. In order to probe the exchange interactions in ZnO:Co, we have compared the experimental data with the results of simulations in the framework of a cluster model. Our conclusion is that both | |J in and | |J out exceed10 K and are negative in ZnO:Co [28]. We turn now to our EPR results. The X -band EPR spectra were recorded using a Bruker EMX spectrometer equipped with a standard TE102 cavity and a continuous helium flow cryostat that allows temperature scans be- tween 4 and 300 K. The films with an area of 3 � 3 mm were mounted on a quartz rod sample holder. The angle between the c-axis of the films and the direction of the static magnetic field H was controlled by a goniometer with a precision better than � 0 25. � . In Fig. 2 we present the low-temperature EPR spectra of ZnO:Co samples with different Co 2 � concentration measured at 9.4 GHz for H c| | . For the lowest x the reso- Exchange broadening of EPR line in ZnO:Co Fizika Nizkikh Temperatur, 2007, v. 33, No. 11 1223 T, K 50 100 200 250 300150 0 M , /C o � B 0 10 20 30 40 50 2.0 1.5 1.0 0.5 H, kOe 4 3 2 1 � – 1 3 – 3 , 1 0 m o l cm Fig. 1. Temperature dependence of the inverse magnetic sus- ceptibility for a Zn1�xCoxO sample with x � 0052. taken at H �10 kOe and H c . The solid line is the Curie–Weiss law. The inset shows M vs H plot for the same sample at T � 2 K and for H c|| and H c . a b 2800 2900 3000 3100 3200 x < 0.001 x = 0.003 x = 0.018 x = 0.052 x = 0.075 J = 0.0085 cm–1 J = 0.0295 cm–1 1500 2250 3000 3750 4500 H, Oe Fig. 2. EPR spectra taken at 7 K and for H c|| on Zn1�xCoxO films with different Co content. Note that the upper scale cor- responds only to the sample with x � 0.001 (a). EPR spectrum of the sample with x � 0003. shown on an enlarged scale. The vertical bars indicate the positions of two exchange pair spec- tra characterized by two different exchange constants (b). nance spectrum consists of eight equally spaced compo- nents resulting from the hyperfine interaction of the elec- tron spin of Co 2 � and its nuclear spin, I � 7 2/ . The main EPR spectrum originates from the low-lying doublet S /z � � 1 2 of a S � 3 2/ ground-state manifold and its po- sition is determined by the following spin Hamiltonian �spin � � �� B z z z z zg H S DS A S I|| || 2 , (1) where g || .� 2 236, D � �2 76 1. cm and A || .� � �16 1 10 4 1cm were inferred from experiments for the lowest x [27,29,30]. Additionally EPR spectra of Co 2 � exchange pairs are observed and shown in detail in Fig. 2,b. (Details of the pairs spectra interpretation will be published in our forthcoming paper.) With increasing of the Co concentra- tion the observed hyperfine structure (HFS) first becomes poorly resolved, for x � 0 003. , and then completely disap- pears for x � 0 02. due to a line broadening. This line broadening is our main concern here. In order to separate the linewidth caused by the HFS from the total one we used the following procedure. The observed EPR spectra were modeled by eight (2I+1) Lorentzian components corresponding to the HFS transi- tions. The resonance fields of the eight Lorentzian were fixed by the position of the center of gravity of the hyperfine octet in accordance with the observed g-factor. Then, we calculate the individual position of each compo- nent using A ||. The full width at half maximum, �H, of in- dividual HFS components were adjusted to get the best agreement with the recorded signal. Fig. 3 shows the linewidth obtained in this way as a function of both Co content and temperature. As it may be seen from the inset of Fig. 3, �H grows almost linearly with x. This is in significant disagreement with previ- ously reported data on Mn-doped DMS. In this latter case and at low x � 01. a rapid decrease of �H is observed sug- gesting the exchange narrowing of an initially broadened line. Another important feature of the observed linewidth behavior is its temperature dependence shown in Fig. 3. �H decreases as the temperature is decreased from 100 K to 60 K and remains almost constant at temperatures below � 60 K down to 4 K. The fast decrease of �H T( ) can be naturally explained by the temperature dependence of the spin lattice relax- ation time of Co 2 � ions, since it is observed even at low- est x. However, the «plateau» below 60 K has clearly a magnetic origin. As we mentioned above, according to previously reported EPR data on Mn-doped DMS, one would expect that �H increases as the temperature is decreased due to the fluctuations of antiferromagne- tically coupled TM ions. Therefore the independence of �H T( ) on temperature can be interpreted as the absence of any such coupling between Co 2 � contributing to the observed line. It is clear that the exchange narrowing scenario is rather unlikely in ZnO:Co, since simple estimates of �H xM /J� 2 , in the case where the second moment M 2, stems from the magnetic dipole or Dzyaloshinskii–Moriya interaction, lead to a negligible linewidth, which is about two orders of mag- nitude smaller than the observed one. The situation changes, however, if one considers M 2 arising from the single ion anisotropy �A . In this case �H x D /J� ( )2 2 gives a linewidth of several kOe i.e. much larger then the observed one. This is an indication that �A of Co 2 � should be included in the unperturbed Hamil- tonian �0 which determines the structure of the energy lev- els between which the EPR absorption takes place. Here the exchange modulation of a broadened line cannot be effec- tive, because �A does not commute with �ex. Kopvillem has developed a theory of the second moment in the case where �0 includes the single ion anisotropy and Zeeman terms �Z while the exchange and the dipole inter- actions were considered as perturbations [31]. The underly- ing physical picture is the following. The �A and �Z split the energy spectrum of the spin system into a series of quasi-discrete bands whose widths depend on the magnetic dipole and exchange interactions among the paramagnetic centers. In its turn the EPR spectrum splits up into a series of fine-structure components corresponding to the magnetic dipole transitions | , | ,S m S mS S� � � �1 . Under these con- ditions the second moment is given by M b P b J P J Pjk k jk jk jk2 2 1 2 2 3� � ��[ ] , (2) where the summation is over all lattice sites, b jk � � �( ) ( cos ( ))||3 2 1 32 2 2 3/ g /rB jk jk� , J jk are the exchange constants and Pi are numerical constants having the fol- lowing values in the case of S � 3 2/ and the transition be- tween m /S � � 1 2: P1 1125� . , P /2 3 4� � , P /3 3 8� . The above formula clearly indicates that the exchange and magnetic dipole interactions are responsible for a linewidth broadening but it cannot be employed directly to calculate the linewidth in ZnO:Co for the following 1224 Fizika Nizkikh Temperatur, 2007, v. 33, No. 11 P. Sati, V. Pashchenko, and A. Stepanov � H , O e 500 400 300 200 100 0 20 40 60 80 100 T, K g || �H 500 � H , O e 400 300 200 100 Co content 0 0.02 0.04 0.06 0.08 g || 2.4 2.3 2.2 2.1 2.0 Fig. 3. Temperature dependence of the EPR linewidth for x � 0018. . The inset shows �H (the left scale) and g || (the right scale) in a function of the Co content. reasons. As was pointed out by Kittel and Abrahams, in the case of very diluted paramagnets (x �� 01. ), when only the magnetic dipole interaction is taken into account, the ratio M /M4 2 2 1�� (where M 4 is the fourth moment) and a resonance line is better described by a cut-off Lorentzian line with �H M�� 2 1 2/ [32]. The summation in the M 2 and M 4 formulae, which have to be carried out over the occu- pied sites only, is replaced by the sum over all crystal sites so that both moments are proportional to the con- centration of magnetic ions, as well as the linewidth �H xM M� 2 3 2 4 1 2/ // . To adapt this approach to the case treated in [31] one needs an estimate of M 4 . Let us first note that the second moment (2) is a sum of pair contributions coming from the dipole and ex- change interactions between different Co 2 � sites. It is easy to show that with the choice of �0 made above the perturbation Hamiltonians, �ex and �d , will cause additional transitions at frequencies �0 � �( )pJ qbjk jk , where �0 is the unperturbed frequency and p and q are constants of the order of unity. Thus M 2 is simply proportional to ( )pJ qbjk jk k �� 2 in agree- ment with Kopvillem’s formula. Quite similarly the leading term in M 4 , which is proportional to x, can be written as ( ) .p J qbjk jk k �� 4 Using these estimates one can recover a linear dependence of the linewidth on the magnetic ion concen- tration, �H x M� 2 1 2/ , which is in qualitative agreement with our experimental data. Another interesting point is the role played by the ex- change interaction in the mechanism of linewidth forma- tion. As we explained above, a pair of Co 2 � spins cou- pled by the exchange constant J contributes to EPR spectrum at frequencies (or magnetic fields) which are proportional to this particular value of J . Hence if one puts in (2), for example, J � 10 K (an estimate for the nearest neighbor exchange integral in ZnO:Co) �H of the order of several teslas will be obtained, which is certainly meaningless. Therefore we should restrict the sum to those pair contributions which do not exceed the experi- mentally observed �H. Some guidelines on how to do this provide our pair spec- trum presented in Fig. 2,b. In fact the exchange-pair transi- tions shown in this figure are both involved in the main reso- nance line broadening, but the ones with J � �0 0295 1. cm , which are too far in the wings, do not contribute to the ob- served linewidth. One can argue, by comparing the ex- change constants obtained in this work with those found for ZnO:Mn [33], that in the case of J � �0 0295 1. cm we are dealing with fifth or sixth neighbor in the ZnO lattice. Thus, we have arrived at an important conclusion that a large num- ber of Co 2 � neighbors in ZnO lattice (n � 5) have no effect on the observed linewidth. This could bring an explanation for the experimentally established absence of strong antiferromagnetic fluctuations at low temperatures in ZnO:Co. In summary, we have reported the X -band EPR studies of single crystalline Zn 1�xCo xO (x = 0.001–0.075) thin films grown by plasma-assisted molecular beam epitaxy. By analyzing the EPR linewidth behavior, we show that a combined effect of the exchange and dipolar broadening plays an important role in the mechanism of linewidth formation in this material. The authors thank C. Deparis and C. Morhain for the sample preparation. Two of us (V.P. and A.S.) dedicate this article to the memory of their dear mentor, Anatolii Illarionovich Zvyagin, an eminent scientist and wonder- ful man. 1. H. Ohno, Science 281, 951 (1998). 2. Y. Ohno, D.K. Young, B. Beschoten, F. Matsukura, H. Ohno, and D.D. Awschalom, Nature 402, 790 (1999); T. Dietl, H. Ohno, F. Matsukura, J. Cibert, and D. Ferrand, Science 287, 1019 (2000). 3. K. Sato and H. Katayama-Yoshida, Physica E10, 251 (2001). 4. C. Song, K.W. Geng, F. Zeng, X.B. Wang, Y.X. Shen, F. Pan, Y.N. Xie, T. Liu, H.T. Zhou, and Z. Fan, Phys. Rev. B73, 024405 (2006). 5. M. Venkatesan, C.B. Fitzgerald, J.G. Lunney, and J.M.D. Coey, Phys. Rev. Lett. 93, 177206 (2004). 6. J.R. Neal, A.J. Behan, R.M. Ibrahim, H.J. Blythe, M. Ziese, A.M. Fox, and G.A. Gehring, Phys. Rev. Lett. 96, 197208 (2006). 7. A.C. Tuan, J.D. Bryan, A.B. Pakhomov, V. Shutthanandan, S. Thevuthasan, D.E. McCready, D. Gaspar, M.H. Engelhard, J.W. Rogers, Jr., K. Krishnan, D.R. Gamelin and S.A. Cham- bers, Phys. Rev. B70, 054424 (2004). 8. K.R. Kittistved, N.S. Norberg, and D.R. Gamelin, Phys. Rev. Lett. 94, 147209 (2005). 9. M.H. Kane, K. Shalini, C.J. Summers, R. Varatharajan, J. Nause, C.R. Vestal, Z.J. Zhang, and I.T. Ferguson, J. Appl. Phys. 97, 023906 (2005). 10. S. Yin, M.X. Xu, L. Yang, J.F. Liu, H. Rosner, H. Hahn, H. Gleiter, D. Schild, S. Doyle, T. Liu, T.D. Hu, E. Ta- kayama-Muromachi, and J.Z. Jiang, Phys. Rev. B73, 224408 (2006). 11. J. Alaria, H. Bieber, S. Colis, G. Schmerber, and A. Dinia, Appl. Phys. Lett. 88, 112503 (2006). 12. M. Bouloudenine, N. Viart, S. Colis, J. Kortus and A. Dinia, Appl. Phys. Lett. 87, 052501 (2005). 13. S. Kolesnik, B. Dabrowski, and J. Mais, J. Appl. Phys. 95, 2582 (2004). 14. Jae Hyun Kim, Hyojin Kim, Dojin Kim, Soon Gil Yoon, and Woong Kil Choo, Solid State Commun. 131, 677 (2004). 15. S-J. Han, B.Y. Lee, J.-S. Ku, Y.B. Kim, and Y.H. Jeong, J. Magn. Magn. Mater. 272, 2008 (2004). 16. A.S. Risbud, N.A. Spaldin, Z.Q. Chen, S. Stemmer, and R. Seshadri, Phys. Rev. B68, 205202 (2003). 17. T. Chanier, M. Sargolzaei, I. Opahle, R. Hayn, and K. Koepernik, Phys. Rev. B73, 134418 (2006). 18. E.C. Lee and K.J. Chang, Phys. Rev. B69, 085205 (2004). 19. M.H.F. Sluiter, Y. Kawazoe, P. Sharma, A. Inoue, A.R. Raju, C. Rout, and U.V. Waghmare, Phys. Rev. Lett 94, 187204 (2005). Exchange broadening of EPR line in ZnO:Co Fizika Nizkikh Temperatur, 2007, v. 33, No. 11 1225 20. A. Abragam and B. Bleaney, Electron Paramagnetic Reso- nance of Transition Ions, Clarendon, Oxford, (1970). 21. N. Samarth and J.K. Furdyna, Phys. Rev. B37, 9227 (1988). 22. S.B. Oseroff, Phys. Rev. B25, 6584 (1982). 23. H.A. Sayad and S.M. Bhagat, Phys. Rev. B31, 591 (1985). 24. R.E. Kremer and J.K. Furdyna, Phys. Rev. B31, 1 (1985). 25. E. Chikoidze, H.J. von Bardeleben, Y. Dumont, P. Galtier, and J.L. Cantin, J. Appl. Phys. 97 10D327 (2005). 26. A. Ben Mahmoud, H.J von Bardeleben, J.L. Cantin, A. Mauger, E. Chikoidze, and Y. Dumont, Phys. Rev. B74, 115203 (2006). 27. P. Sati, R. Hayn, R. Kuzian, S. Regnier, S. Schafer, A. Stepanov, C. Morhain, C. Deparis, M. Laugt, M. Goiran, and Z. Golacki, Phys. Rev. Lett. 96, 017203 (2006). 28. P. Sati, C. Deparis, C. Morhain, S. Schafer, and A. Stepanov, Phys. Rev. Lett. 98, 137204 (2007). 29. T. Estle and M. De Wit, Bull. Am. Phys. Soc. 6, 445 (1961). 30. N. Jedrecy, H.J. von Bardeleben, Y. Zheng, and J-L. Cantin, Phys. Rev. B69, 041308(R) (2004). 31. U.Kh. Kopvillem, Soviet Phys. JETP 11, 109 (1960). 32. C. Kittel and E. Abrahams, Phys. Rev. 90, 238 (1953). 33. X. Gratens, V. Bindilatti, N.F. Oliveira, Jr., Y. Shapira, S. Foner, Z. Golacki, and T.E. Haas, Phys. Rev. B69, 125209 (2004). 1226 Fizika Nizkikh Temperatur, 2007, v. 33, No. 11 P. Sati, V. Pashchenko, and A. Stepanov