Novel aspects of charge and lattice dynamics in the hole-doped manganite La_0.67Sr_0.33MnO₃

Abstract

Previous infrared studies on the hole-doped manganite La 0.67 Sr 0.33 MnO 3 (LSMO) have analysed its charge dynamics in terms of one type of charge carrier despite evidence of both electron and hole Fermi surfaces. Here, we investigate the charge dynamics of an LSMO film with infrared and optical spectroscopy in order to provide a complete picture of metallic conduction. In the ferromagnetic metallic phase, the low-frequency optical conductivity is best explained by a two-carrier model comprising electrons and holes. The number densities, effective masses and relaxation response of the delocalized electrons and holes are quantified. We discover that only one-third of the doped charges are coherent and contribute to the dc transport. Metallic LSMO cannot be classified as a bad metal at low temperatures because the mean free path of the coherent, mobile charge carriers exceeds the Ioffe-Regel-Mott limit. The incoherent spectral response of the doped charges manifests itself as a broad mid-infrared feature. We also report the first observation of splitting of an infrared-active phonon due to local Jahn-Teller distortion in the vicinity of the thermally driven transition to the nonmetallic, paramagnetic phase in LSMO. This demonstrates that infrared spectroscopy is capable of detecting the presence of local lattice distortions in correlated electron systems.

Role of electron-electron interactions in the charge dynamics of rare-earth-dopedCaFe2As2

“…The coefficient of the quadratic term d = 0.008 86 K -2 cm -1 = 1.67×10 9 K -2 s -1 is much larger than that of a good metal like gold (~10 7 K -2 s -1 ) [44,45]. The temperature dependent quadratic term is likely electronic in origin and is similar to that seen in other Mott-correlated and Hund-correlated systems like V2O3, NdNiO3, La0.67Sr0.33MnO3, and CaRuO3 [45,46].…”

“…The coefficient of the quadratic term d = 0.008 86 K -2 cm -1 = 1.67×10 9 K -2 s -1 is much larger than that of a good metal like gold (~10 7 K -2 s -1 ) [44,45]. The temperature dependent quadratic term is likely electronic in origin and is similar to that seen in other Mott-correlated and Hund-correlated systems like V2O3, NdNiO3, La0.67Sr0.33MnO3, and CaRuO3 [45,46]. The CT phase transition in Pr-doped CaFe2As2 at ~ 70 K with a hysteresis of ~ 30 K [14] precludes the preceding quantitative analysis, but we note that the temperature dependence of the scattering rate above 70 K closely resembles the data for La-doped CaFe2As2.…”

Temperature dependent infrared nano-imaging of La0.67Sr0.33MnO3thin film

“…There are no topographic features near the center of the scan that can potentially affect the nearfield infrared signal. Therefore, the nearfield infrared amplitude contrast is due to differences in the local dielectric function: the lower amplitude is from regions that are more insulating compared to the surrounding material [23,[29][30][31]. Note that the 5 nm high island appears as a dip in the nearfield infrared signal.…”