PHYSICAL REVIEW B, VOLUME 64, 106401

Comment on ‘‘Paramagnetic anomalies above the Curie temperature and colossal magnetoresistance in optimally doped manganites’’ F. Rivadulla,1 L. E. Hueso,2 M. A. Lo´pez-Quintela,1 J. Rivas,2 and M. T. Causa3 1

Departamento de Quı´mica-Fı´sica, Universidad de Santiago de Compostela, E-15782 Santiago de Compostela, Spain 2 Departamento de Fı´sica Aplicada, Universidad de Santiago de Compostela, E-15782 Santiago de Compostela, Spain 3 Centro Ato´mico de Bariloche, 8400-San Carlos de Bariloche, Argentina 共Received 19 February 2001; published 23 August 2001兲 In a recently published paper 关Phys. Rev. B 62, 5313 共2000兲兴, Yuan et al. studied the magnetoresistance 共MR兲 and electron paramagnetic resonance 共EPR兲 of ceramic samples of (La1⫺x Yx ) 1/3Ca2/3MnO3 (x ⫽0.15,0.20). They argue that the EPR linewidth and the MR exhibit similar temperature dependences in the range T C ⭐T⭐T onset , and conclude that the same mechanism underlies both effects. In our opinion, they made a mistake in analyzing the origin of EPR line broadening below T onset . Similar experiments carried out in single crystals showed that while the peak of MR around T C is intrinsic to these manganites, the observed line broadening is not. This effect has been previously related to the presence of pores between grains in polycrystalline samples and surface irregularities in single crystals, and is common to other magnetic materials that do not show colossal magnetoresistance. DOI: 10.1103/PhysRevB.64.106401

PACS number共s兲: 76.30.⫺v, 75.30.Vn, 76.50.⫹g, 71.30.⫹h

The possibility of a dynamic electronic phase segregation as the origin of colossal magnetoresistance 共CMR兲 in manganites is now under discussion.1,2 In fact, through scanning tunneling spectroscopy experiments, Fa¨th et al.3 visualized ferromagnetic 共FM兲 regions embedded in a paramagnetic 共PM兲 matrix around T C in La2/3Ca1/3MnO3 , and studied the evolution of these FM clusters with temperature and magnetic field. Moreover, muon spin relaxation and neutron spin echo measurements in ferromagnetic (La,Ca)MnO3 indicated the presence of two spatially separated regions possessing very different Mn-ion spin dynamics.4 In the paper here commented on, Yuan et al.5 basically observed that the temperature below which MR becomes obin ceramic samples of servable (T onset ) (La1⫺x Yx ) 2/3Ca1/3MnO3 is approximately the same below which the electron paramagnetic resonance 共EPR兲 linewidth (⌬H pp ) starts to increase. When the temperature dependence of MR and ⌬H pp are plotted on the same graph, they observe a similar rate of increase for both magnitudes in the range T C ⬍T⬍T onset . From these results they conclude that the same mechanism is responsible for the peak of MR and the EPR line broadening around T C . We disagree with this interpretation about the origin of the observed line broadening below T onset . The typical CMR response around T C in Mn3⫹ /Mn4⫹ perovskites was first observed in thin films of these oxides6 but also occurs in polycrystalline materials7 and single crystals,8 and nowadays there is no doubt that it is intrinsic to these materials. So the CMR peak around T C and the minimum in the EPR linewidth at T onset showed by Yuan et al. are expected and previously observed results.9 On the other hand, as we have reported in another paper,10 the increment of ⌬H pp below T onset is a well-known effect that is not intrinsic to manganites. This has been observed in pyroclores, ferrites, garnets, and several other magnetic materials that do not show CMR11,12 and is related to a two-magnon scattering relax0163-1829/2001/64共10兲/106401共3兲/$20.00

ation mechanism induced by the demagnetization fields of the pores between crystallites. The Gilbert equation, which accounts for the ferromagnetic resonance 共FMR兲, predicts a temperature-independent value of the linewidth below T C . 11 This result is in clear contradiction with some experimental data, like those presented in the paper by Yuan et al. We previously reported a constant linewidth below T C for single crystals of Pr2/3Sr1/3MnO3 and La2/3Sr1/3MnO3 , and observed that a similar broadening to that measured in similar ceramic samples can be induced by surface polishing.10 In Fig. 1 we show this effect for a single crystal of La2/3Sr1/3MnO3 . On the other hand, surface polishing does not produce any change in the MR of these samples. From these results it follows that the observed increment of the EPR linewidth below T onset is a surface effect, not intrinsic to the physics of manganites. On approaching T C from above 共near T onset ) the conductivity and magnetic permeability of the samples increase as we approach the paramagnetic-insulator to ferromagnetic-metallic phase transition 共in fact, some parts of the sample are ferromagnetic and metallic between T C and T onset ), and the absorption/dispersion ratio decreases 共as the skin depth does兲, producing Dysonian-like line shapes. Although the authors observe the distortions in symmetry of the Lorentzian line below T onset , they measure the linewidth directly from the plot. This is not correct. As we comment on in Ref. 10, for Dysonnian shapes, the line must be fitted to an absorption/dispersion function in order to obtain the parameters of the absorption line. So the similar increment rate of CMR and ⌬H pp below T onset seems to be a coincidence 共in fact, Yuan et al. are comparing a dimensionless magnitude, like MR, with ⌬H pp in Oe, on the same plot兲. Moreover, the MR presented by Yuan et al. is measured at 50 kOe, far above the resonance field for the X band (⬃3.3 kOe).

64 106401-1

©2001 The American Physical Society

COMMENTS

PHYSICAL REVIEW B 64 106401

FIG. 1. Effect of surface polishing on the thermal evolution of the linewidth of La0.67Sr0.33MnO3 below T onset . The size of the surface pits induced is indicated. For unpolished crystals, the linewidth follows the temperature-independent behavior predicted by the Gilbert equation, while a dramatic increment similar to ceramic samples 共also shown兲 can be induced by the surface polish.

As we have previously shown, the temperature dependence of ⌬H pp below ⬃1.1 T C for ceramic samples is successfully described by ⌬H pp 共 ⬍1.1T C 兲 ⫽4 ␲ M p ␤ ,

共1兲

where M is the magnetization extracted from initial magnetization curves for H⫽H r , the resonance field; p is the porosity of the sample 共measured as the diference between the x-ray and macroscopic densities兲 and ␤ , following the theory developed by Schlo¨mann13 and Sparks,14 can be related to the shape and size of the pores between grains. In Fig. 2 we show the experimental values of the linewidth for various samples at various frequencies, along with the fit to Eq. 共1兲. As can be seen, good correlation between experimental data and Eq. 共1兲 is obtained for different samples and frequencies. The usual FMR linewidth is determined by the relaxation frequency of the uniform precession spin wave

1

A. Moreo, M. Mayr, A. Feiguin, S. Yunoki, and E. Dagotto, Phys. Rev. Lett. 84, 5568 共2000兲. 2 M. Uehara, S. Mori, C.H. Chen, and S.-W. Cheong, Nature 共London兲 399, 560 共1999兲. 3 M. Fa¨th, S. Freisem, A.A. Menovsky, Y. Tomioka, J. Aarts, and J.A. Mydosh, Science 285, 1540 共1999兲. 4 R.H. Heffner, J.E. Sonier, D.E. MacLaughlin, G.J. Nierwenhuis, G. Ehlers, F. Mezei, S.-W. Cheong, J.S. Gardner, and H. Ro¨der, Phys. Rev. Lett. 85, 3285 共2000兲. 5 S.L. Yuan, J.Q. Li, Y. Jiang, Y.P. Yang, X.Y. Zeng, G. Li, F. Tu, G.Q. Zhang, C.Q. Tang, and S.Z. Jin, Phys. Rev. B 62, 5313 共2000兲.

FIG. 2. Linewidth for various frequencies for ceramic samples of different compositions. Solid lines represent the fits to Eq. 共1兲, with M extracted from the initial magnetization curves at the resonance field for each frequency.

共i.e., k⫽0,␭⫽⬁). But the presence of pores between the grains in a ceramic sample, or the surface pits created by surface polishing in single crystals, introduces a demagnetizing energy in the Hamiltonian of the system that produces relaxation of the uniform precession into other spin-wave modes (␭⫽⬁). This gives place to a nonintrinsic contribution to the linewidth of several magnetic materials. In fact, this theory of porosity broadening was originally developed by Scho¨loman and Sparks for ferrites and has been recently applied to the case of Sr2 FeMoO6 .15 Surface polishing effects were also observed in yttrium iron garnet 共YIG兲 and ferrite single crystals.16 In summary, we have demonstrated that the sudden increase of ⌬H pp (T⬍T onset ) observed in CMR manganites is not an intrinsic property to these materials, and hence it cannot be related to the phase-segregation scenario now believed to underlie CMR. On the other hand, our results indicate that the evolution of the linewidth below this temperature is due to a complex two-magnon relaxation mechanism, common to other magnetic materials.

6

S. Jin, T.H. Tiefel, M. McCormack, R.A. Fastnacht, R. Ramesh, and L.H. Chen, Science 264, 413 共1994兲. 7 P. Schiffer, A.P. Ramirez, W. Bao, and S.-W. Cheong, Phys. Rev. Lett. 75, 3336 共1995兲. 8 A. Urushibara, Y. Moritomo, T. Arima, A. Asamitsu, G. Kido, and Y. Tokura, Phys. Rev. B 51, 14 103 共1995兲. 9 G. Alejandro, M.T. Causa, M. Tovar, J. Fontcuberta, and X. Obradors, J. Appl. Phys. 87, 5603 共2000兲, and references therein. 10 F. Rivadulla, M.A. Lo´pez-Quintela, L.E. Hueso, J. Rivas, M.T. Causa, C. Ramos, R.D. Sa´nchez, and M. Tovar, Phys. Rev. B 60, 11 922 共1999兲. 11 A. H. Morrish, The Physical Principles of Magnetism 共Wiley,

106401-2

COMMENTS

12

PHYSICAL REVIEW B 64 106401

New York, 1965兲. ˜ ez, C.A. Ramos, A. Butera, F. M. Tovar, M.T. Causa, G. Iban Rivadulla, B. Alascio, S.B. Oseroff, S.-W. Cheong, X. Obradors, ˜ ol, J. Appl. Phys. 83, 7201 共1998兲. and S. Pin

E. Schlo¨mann, J. Phys. Chem. Solids 6, 242 共1958兲. M. Sparks, J. Appl. Phys. 36, 1570 共1965兲. 15 D. Niebieskikwiat et al., Phys. Rev. B 62, 3340 共2000兲. 16 G.F. Dionne, J. Appl. Phys. 43, 1221 共1972兲. 13 14

106401-3