Abstract

Using original modulation technique, the longitudinal electron spin-lattice relaxation time T₁ has been measured in a series of nearly cubic La_1−xCa_xMnO₃ (x=0.1; 0.2; 0.25; 0.33) and layered La_2−2xSr_1+2xMn₂O₇ (x=0.4 and 0.5) manganites both in paramagnetic state and around the temperature (T_c) of the magnetic ordering. The data are compared with the evolution of the transverse relaxation time T₂ as determined from the ESR linewidth. Well above T_c, the T₁=T₂ equality is confirmed, whereas a pronounced slowing down of T₁ is observed in all the materials as T_c is approached. The temperature dependence of T₁ is found to be consistent with that of T(T), where (T) represents the temperature dependence of the electron-spin susceptibility determined from the ESR absorption area. The interpretation suggests freezing the local field fluctuations due to polarization of superparamagnetic clusters in the external magnetic field.

Author Keywords: Electron spin resonance; Spin relaxation; Manganites; Critical phenomena

Article Outline

• References

Recently much attention is drawn to properties of manganites with the perovskite structure. The explanation of the most intriguing features of these materials, such as the colossal magnetoresistance (CMR), simultaneous magnetic and conduction phase transitions, and an extremely rich and peculiar phase diagram, is based on the "double exchange" model including the interaction between Mn³⁺ and Mn⁴⁺ ions mediated by mobile charge carriers. The Jahn–Teller effect and polaron formation might be also taken into account, but the whole physical picture is not clear yet (see the review articles [1 and 2]).

Electron spin resonance (ESR) is a powerful tool in investigation of internal fields and spin dynamics in solids. A number of papers on ESR in manganites were published, concerned with both nearly cubic materials such as La_1−xMe_xMnO₃ where Me=Ca, Sr, … (see, for example, Refs. [3, 4, 5, 6, 7 and 8]) and the layered ones (La_2−2xSr_1+2xMn₂O₇) [9, 10 and 11]. A single intense ESR line with g≈2 and Lorentzian shape is commonly observed in the paramagnetic phase. It is commonly accepted that this line corresponds to a combined spin system including both Mn³⁺ (effective spin S=2) and Mn⁴⁺ (S=3/2) ions coupled by strong exchange interaction. More detailed information on magnetic and electronic state can be extracted from the measurements of electron spin relaxation. The transverse relaxation time T₂ can be easily measured from the peak-to-peak ESR linewidth Δ_pp according to the equation T₂⁻¹=(3/2)Δ_pp, where is the electronic gyromagnetic ratio. However, the contribution of the inhomogeneous broadening can mask the true value of the relaxation rate. Much more useful can be measurements of the longitudinal relaxation time, T₁. The corresponding rate, T₁⁻¹, is proportional to the spectral density of the internal field fluctuations at the ESR frequency ₀ and so can tell about magnetic correlations and other dynamical processes of interest. Unlike T₂, measuring T₁ in manganites presents considerable difficulties because of unrealistic microwave power needed for pronounced saturation of the ESR line. To overcome this problem, we have employed an original version [12] of the modulation technique with registering the longitudinal response of the spin magnetization [13]. The method can work at the saturation levels as low as 10⁻⁴ and allows determination of the T₁ values as short as 10⁻⁹–10⁻¹⁰ s [12 and 14].

The samples under study were the "cubic" compounds, La_1−xCa_xMnO₃ with x=0.1, 0.2, 0.25, and 0.33 (ceramics) and the layered ones, La_2−2xSr_1+2xMn₂O₇, with x=0.4 and 0.5 (single crystals). The preparation of the samples were described elsewhere [15 and 16]. All ESR and relaxation measurements were performed in the X-band (₀/2=9.4 MHz), in the magnetic fields of a few kG. The "amplitude" version of the modulation technique of measuring T₁ [12] was used; the DPPH standard was employed for the T₁ calibration. More emphasis was placed on the measurements in the paramagnetic phase, especially in the vicinity of the critical temperature T_c. For details see Refs. [16 and 17].

The temperature dependencies of both T₁ and T₂ value (the latter being determined from the ESR linewidth) in the "cubic" La_1−xCa_xMnO₃ manganites are shown in Fig. 1 and Fig. 2. At higher doping levels (x>0.2), the paramagnet-to-ferromagnet (PM–FM) transitions are rather narrow and accompanied by the insulator-to-metal (I–M) ones [1 and 2]. At x=0.1, the transition to the FM state is much broader, and the sample remains insulating below T_c=152 K. As can be seen from the Figures, the behavior of the relaxation times is quite remarkable. Well above T_c, the equality T₁=T₂ is fulfilled in all the compounds, just as expected under conditions of the exchange narrowing with the correlation time _c<<₀⁻¹ [18]. When approaching T_c, however, this equality breaks down. The transverse relaxation rate (i.e., the ESR linewidth) initially decreases upon cooling, then passes through a minimum slightly above T_c, and finally demonstrates a steep raising that might be understood as an evidence for critical "speeding up" predicted by the theory [19]. Unlike this, no signs of such acceleration are seen in the longitudinal relaxation. To the contrary, the T₁ values show a steep increase when approaching T_c, so that the T₁/T₂ ratio amounts up to an order of magnitude.

(7K)

Fig. 1. Temperature dependencies of the transverse (open symbols) and longitudinal (filled symbols) relaxation rates in the La_1−xCa_xMnO₃ ceramics for x=0.2 (circles), x=0.25 (triangles), and x=0.33 (squares). The lines connect the experimental points. Arrows indicate the critical temperatures.

(6K)

Fig. 2. Temperature dependencies of T₁⁻¹ and T₂⁻¹ (filled and open squares, left scale) compared with that of (T)⁻¹ (crosses with curve, right scale) in the La_0.9Ca_0.1MnO₃ ceramics.

The most detailed investigation was carried out on the x=0.1 sample (Fig. 2). In this case, the transition is broad enough to allow much more accurate determination of the critical parameters. In the Fig. 2, the relaxation rates are compared with the Huber formula [4] derived for the non-critical temperature range, T>>T_c:

(1)

Similar investigations have been performed on the single crystals of the layered manganites La_1.2Sr_1.8Mn₂O₇ (x=0.4) and LaSr₂Mn₂O₇ (x=0.5). The x=0.4 compound undergoes simultaneous PM–FM and I–M transitions at T_c=126 K [10 and 11]; the x=0.5 one demonstrates a charge ordering state below T_CO=226 K and the antiferromagnetic (AF) transition at T_N=170 K [20]. In addition to the "conventional" nearly isotropic EPR line, strongly anisotropic spectra typical of ferromagnetic resonance were observed in the both compounds at all temperatures below 300 K. Likely, these features are caused by parasitic phases [10 and 11] and so not considered here. Besides, in the x=0.5 system, a slightly anisotropic ESR line have been observed, probably related to magnetic polarons ("ferrons") existing on the AF correlated background [2].

In the La_1.2Sr_1.8Mn₂O₇ system, a pronounced deviation from the Curie–Weiss law was found, suggesting strong ferromagnetic correlations typical of quasi-2D spin systems. The violation of the T₁=T₂ equality takes place in a wide temperature range 126–270 K; in all this region, the T₁ temperature dependence obeys the "noncritical Huber Law", Eq. (1), see Fig. 3.

(6K)

Fig. 3. Temperature dependencies of T₁⁻¹ (filled squares) and T₂⁻¹ (open squares) in the La_1.2Sr_1.8Mn₂O₇ single crystal (H in the ab plane). Crosses with curve show (T)⁻¹ (arb. units).

The observed absence of the critical "speeding up" of T₁ can be attributed to the influence of the external magnetic field which is known to suppress critical phenomena while approaching the ferromagnetic order [21 and 22]. The strength of this suppression depends on the FM correlation length in the PM phase. So our data suggest the effectiveness of such correlations in the CMR manganites, even well above T_c, and especially for the quasi-2D (layered) systems. This finding is in agreement with the data on the muon spin relaxation which demonstrate a total suppression of the critical speeding up in manganites by a moderate field of about 3 kG [23].

In regard to the T₂⁻¹ rate as measured from the ESR linewidth, we believe that the broadening observed in some manganites near T_c [3, 6, 7, 8 and 11] is in fact inhomogeneous and caused by non-uniform magnetization in the presence of strong ferromagnetic correlations. Such "critical" broadening disappears in good single crystals with perfect surface [5].

In conclusion, the longitudinal electron spin relaxation has been investigated in a series of the cubic and layered CMR manganites. No signs of critical acceleration of T₁ was observed; instead, the experimental data were found to be consistent with the "noncritical Huber Law" in a wide temperature range including close vicinity of the FM phase transition. The suppression of the critical speeding up can be caused by the external magnetic field and implies the existence of strong ferromagnetic correlations well above T_c.

The research was supported by the Russian Foundation for Basic Research (Grant No. 02-02-16219) and Swiss National Science Foundation (Grant No. 7GEPJ062429).

References

1. J.M.D. Coey, M. Viret and S. von Molnar. Adv. Phys. 48 (1999), p. 167. Abstract-Compendex | Abstract-INSPEC   | $Order Document | Full Text via CrossRef

2. E.L. Nagaev. Phys. Rep. 346 (2001), p. 387. SummaryPlus | Full Text + Links | PDF (1313 K)

3. M.T. Causa et al.. Phys. Rev. B 58 (1998), p. 3233. Abstract-INSPEC   | $Order Document | APS full text | Full Text via CrossRef

4. D.L. Huber et al.. Phys. Rev. B 60 (1999), p. 12155. Abstract-INSPEC   | $Order Document | APS full text | Full Text via CrossRef

5. F. Rivadulla et al.. Rev. B 60 (1999), p. 11922. Abstract-INSPEC   | $Order Document | APS full text | Full Text via CrossRef

6. V.A. Ivanshin et al.. Phys. Rev. B 61 (2000), p. 6213. Abstract-INSPEC   | $Order Document | APS full text | Full Text via CrossRef

7. A. Shengelaya, G.-M. Zhao, H. Keller, K.A. Muller and B.I. Kochelaev. Phys. Rev. B 61 (2000), p. 5888. Abstract-INSPEC   | $Order Document | APS full text | Full Text via CrossRef

8. S.L. Yuan et al.. Phys. Rev. B 62 (2000), p. 5313. Abstract-INSPEC   | $Order Document | Full Text via CrossRef

9. O. Chauvet, G. Goglio, P. Molinie, B. Corraze and L. Brohan. Phys. Rev. Lett. 81 (1998), p. 1102. Abstract-INSPEC   | $Order Document | APS full text | Full Text via CrossRef

10. S.M. Bhagat, S.E. Lofland and J.F. Mitchell. Phys. Lett. A 259 (1999), p. 326. SummaryPlus | Full Text + Links | PDF (64 K)

11. N.O. Moreno et al.. Phys. Rev. B 63 (2001), p. 174413. Full Text via CrossRef

12. V.A. Atsarkin, V.V. Demidov and G.A. Vasneva. Phys. Rev. B 52 (1995), p. 1290. Abstract-INSPEC   | $Order Document | Full Text via CrossRef

13. J. Herve and J. Pescia. C. R. Hebd. Seances Acad. Sci. (Paris) 251 (1960), p. 665.

14. V.A. Atsarkin, V.V. Demidov and G.A. Vasneva. Appl. Magn. Reson. 19 (2000), p. 329. Abstract-INSPEC   | $Order Document

15. Y. Moritomo, A. Asamitsu, H. Kuwahara and Y. Tokura. Nature (London) 380 (1996), p. 141. Full Text via CrossRef

16. V.A. Atsarkin, V.V. Demidov, G.A. Vasneva and K. Conder. Phys. Rev. B 63 (2001), p. 092405. Full Text via CrossRef

17. V.A. Atsarkin, V.V. Demidov, G.A. Vasneva and D.G. Gotovtsev. Appl. Magn. Reson. 21 (2001), p. 147. Abstract-INSPEC   | $Order Document

18. A. Abragam. The Principles of Nuclear Magnetism, Clarendon Press, Oxford (1961).

19. D.L. Huber. J. Phys. Chem. Solids 32 (1971), p. 2145. Abstract-INSPEC   | $Order Document

20. D.N. Argyriou et al.. Phys. Rev. B 61 (2000), p. 15269. Abstract-INSPEC   | $Order Document | APS full text | Full Text via CrossRef

21. M.S. Seehra and R.P. Gupta. Phys. Rev. B 9 (1974), p. 197. Abstract-INSPEC   | $Order Document | Full Text via CrossRef

22. V.N. Berzhanskii, V.I. Ivanov and A.V. Lazuta. Solid State Commun. 44 (1982), p. 771.

23. R.H. Heffner et al.. Physica B 230–232 (1997), p. 759. Abstract | Abstract + References | PDF (241 K)

Corresponding author. Tel.: +7-095-203-0156; fax: +7-095-203-8414