Research Theme T1


(T1)  Magnetic, electrical, thermal and ferroelectric properties of the oxygen-isotope substituted half-doped manganites

CSULA Faculty Participants:  Guo-meng Zhao, Oscar Bernal, Jose Rodriguez, Radi Jishi

Penn State Faculty Participants:  Vincent Crespi, Venkatraman Gopalan, Long-Qing Chen; Roman Engel-Herbert, Susan Trolier-McKinstry

In achieving an efficient mutual control of electricity and magnetism, the use of multiferroics is one of the most promising approaches towards the development of low-power-consumption spintronics devices. Multiferroic materials simultaneously exhibit ferromagnetism, ferroelectricity, and possibly ferroelasticity. The close interplay among spin, charge, and lattice degrees of freedom may lead to electric-field control of magnetism (magnetoelectric effect). Perovskite oxides such as EuTiO3 [107, 108], GdFeO3 [109], BiFeO3 [110], and TbMnO3 [111] have been found to show multiferroic properties, but the magnetoelectric effect is rather small and only occurs at low temperatures. Therefore, to achieve practical applications, it is essential to find new materials that can exhibit strong magnetoelectric effects at room temperature.

Doped manganites exhibit colossal magnetoresistance (CMR), which would have an important application in spintronic devices, but attempts to utilize the CMR effect have been rather fruitless due to the fact that the magnitude of the required magnetic fields is too large to realize practical magnetic storage, for instance. On the other hand, there is hope to make progress taking a different route as there is a strong interplay among spin, charge, orbital, and lattice degrees of freedom in manganites featuring strong electron-phonon coupling. The phase transition from a charge-ordered (CO) antiferromagnetic (AF) phase to a ferromagnetic (FM) phase is often accompanied by an insulator-to-metal transition (IMT). Remarkably, the IMT in some manganites can be strongly modified by many different perturbations such as hydrostatic pressure [112], substrate strain [113], electric field [114], photon illumination [115], and even by oxygen-isotope substitution [116]. Of particular interest is the strong current or electric field effect on the resistivity or magnetization in some charge-ordered manganites [114]. The data appear to indicate that the pronounced magnetoelectric effect occurs at
temperatures in proximity to the charge-ordering temperature TCO. Therefore, it is desirable to study manganites whose TCO’s are close to or above room temperature.

The half-doped manganites Ln0.5Ba0.5MnO3 (Ln = rare-earth element) provide a good opportunity for these kinds of studies. The manganites in this chemical composition have two possible forms of the crystal structure depending on the synthetic condition [117]. One is the A-site disordered and the other is the A-site ordered perovskite. Only the A-site ordered half-doped manganites can have charge-ordering temperatures at or above room temperature when the A-site ionic radius is below about 1.27 Å [118]. When the A-site ionic radius is above 1.27 Å, the ground state is ferromagnetic and the Curie and IMT temperatures coincide [118]. Our prior work showed that the TCO’s of the 18O polycrystalline samples of La0.5Ca0.5MnO3 [119] and Nd0.5Sr0.5MnO3 [120] are significantly higher than those of the 16O samples. The giant oxygen-isotope shift of TCO has recently been confirmed in singlecrystalline samples of Nd0.5Sr0.5MnO3 (see Figure 1). The results suggest that we should be able to fine-tune TCO to maximize the magnetoelectric effect by partial oxygen-Figure 1. T-dependence of the magnetization and specific-heat of the oxygen-isotope exchanged single-crystalline Nd0.5Sr0.5MnO3 upon heating in a magnetic field of 5 T (the 18O sample contains about 85% 18O isotope).

isotope substitution. It is interesting that, upon cooling, the specific-heat anomaly becomes negligibly small and the TCO shifts down by about 10 K. This irreversible behavior is consistent with a first order phase transition, as demonstrated theoretically for a structural phase transition where the specific heat anomaly almost disappears upon cooling [121]. The first-order phase transition and large magneocaloric effect in this material may lead to an important application for magnetic refrigeration [122], an environmentally friendly technology that may minimize the use of ozone-depleting gases. Our results also suggest that the magnetocaloric effect increases rapidly with increasing the charge ordering temperature. Therefore, it is desirable to find a material with a higher TCO to achieve a larger magnetocaloric effect. We further showed that the extreme sensitivities of the physical properties to the various perturbations were closely related to the intrinsic electronic phase separation, which has been probed by various techniques such as thermal expansion [123], neutron diffraction [123], NMR [124], and muon spin rotation (muSR) [125]. It should be interesting to study the correlation between the magnetoelectric effect and the electronic phase separation. We propose to synthesize and study the oxygen-isotope (16O, 17O, or 18O) exchanged poly-crystals, single-crystals, and epitaxially grown thin-films of Ln0.5Ba0.5MnO3 with varying A-site ionic radius from 1.21 to 1.36 Å. The poly-crystals and single-crystals of the materials will be prepared at CSULA (Zhao’s group) while the films will be prepared at Penn State (Engel-Herbert). Zhao’s group will make oxygen-isotope substitution for all the samples. Magnetic, resistivity, Hall effect and dielectric constant measurements will be performed on all the samples. Specific heat, 55Mn and 17O NMR/NQR measurements will be performed on the polycrystalline and single-crystalline samples. Zhao’s group will perform magnetic, resistivity, and Hall effect measurements and Bernal’s group will measure specific heat and NMR/NQR while the dielectric constant measurement and other sample characterization, not available at CSULA (e.g., HRTEM), will be performed at Penn State (Trolier-McKinstry). The nature of the phase transition and the magnetocaloric effect can be well determined by the specific-heat and magnetization measurements in different magnetic fields. NMR/NQR measurements could, in addition to probing electronic phase separation, also specify the nature of the phase transition via the thermal hysteresis of the spectral linewidth near the transition. For more advanced measurements at the nanoscopic level, either Zhao or Bernal will travel to PSU to collaborate with Gopalan, whose group has been doing similar studies on 18O-doped VO2. They have also done optical illumination studies which would be interesting to compare with similar studies in the manganites. Their instrumentation arsenal would also allow us to perform in-situ spectroscopic ellipsometry of the IMT under various conditions.

Complementary muSR experiments [126] will be performed by Bernal’s group on all bulk samples at TRIUMF Canada) and Paul Scherrer Institute (PSI), Switzerland to get information on intrinsic electronic phase separation. In addition, the relatively new technique of low-energy muSR was developed at PSI precisely to study physical properties at surfaces/interfaces and thin films [127]. In this technique, muons are slowed down to energies low enough to penetrate the sample surface from about 1 to 100 nm and can map the magnetic environment as a function of depth. This technique has already been used to obtain the magnetic penetration depth of high-Tc superconductors, for instance [128]. We anticipate that this powerful technique is extremely valuable in probing variation of the magnetic environment as a function of distance from the interface of a multilayer thin-film heterostructure. The presence of 17O nuclei in the 17O-isotope exchanged films or multi-layer thinfilm heterostructures can also be probed by the muons via direct dipolar interactions between the muon and nuclear spins. We expect that the ground-work on the bulk materials will provide the direction for designing desirable multi-layer heterostructure profiles to be probed by low-energy muons. CSULA has a long and productive history of NMR and muSR research; see e.g., Refs [129, 130].

Elucidation of interfacial effects will also be part of the proposed research. It is known that switching of ferromagnetic domains in cobalt films can be achieved by the control of the magnetocrystalline anisotropy axis of antiferromagnetic ferroics via electric fields as explained in Ref. [131]. Similarly, history-dependent hysteretic behavior is observed at the ferromagnetic interface between BiFeO3 and lanthanum strontium manganite (La1- xSrxMnO3) [132]. Motivated by these reports, we will be searching for similar phenomena at the interface of BiFeO3 and Ln0.5Ba0.5MnO3 samples with different oxygen-isotope substitutions. The proposed isotope study should be able to confirm whether electronic conduction at such interfaces is limited by polaronic dynamics, for example, which is an idea that has been advanced recently [133]. Ideally, it will determine what fraction of the interfacial electron effective mass is due to the electron-phonon interaction. The proposed isotope study will also reveal the interplay between such polaronic effects and magnetism at the heterostructure interface, which also limits electron dynamics. Theoretical study of interfacial polaron formation will be carried out by Jishi's group while the theoretical study of interface magnetism will be carried out by Rodriguez’s group [134]. Both Jishi and Rodriguez will collaborate with PSU’s Chen, who has extensive expertise in modeling heterostructures and interfaces. The experimentation required for probing these effects and test the theoretical predictions will be partially carried out at Penn State facilities by CSULA faculty and students in collaboration with Gopalan and his IRG1 group, which has extensive experience in in-situ x-ray characterization (in collaboration with Haidan Wen at Argonne National Labs) to study transient phases during electrical and optical switching. On the other hand, low-energy muSR experiments to probe interfacial magnetism will be performed at the user facility PSI by CSULA faculty and students.

All the experimental part of the research project will be coupled directly with theoretical work at CSULA by building on previous and current accomplishments. For instance, Jishi’s group has performed first-principles calculations based on density-functional theory (DFT) to obtain the electron-phonon coupling contribution to the pairing interaction in Fe-based superconductors [135]. Similarly, we will use these techniques to extract information on electronic band and magnetic structures as well as spin-lattice interactions. By changing the Asite ionic radius of the half doped manganites, one expects changes in band parameters, electron-phonon and spin-lattice couplings and even magnetic interactions. All these quantities can be obtained using the DFT.

The experimental and theoretical components of the proposed research together will produce a comprehensive set of results to shed light on the basic physics of manganites (e.g., the microscopic mechanisms for CMR, ferromagnetism, spin and charge ordering), which still remains controversial. The proposed research could in addition also lead to a breakthrough in technological applications if a strong magnetoelectric effect were observed in our manganite samples at room temperature. The proposed research will also provide a good opportunity for UG and MS students to participate and be exposed to hands-on experiences. This has been clearly demonstrated in our prior research on manganites. Student John Mann worked in Zhao’s group on the oxygen-isotope effect in doped manganites in 2007 and 2008 and published a paper in Phys. Rev. B (Rapid Communications) [136]. He then went to UC-Riverside for the Ph.D. program in 2008. Victor Aguilar worked on the oxygen-isotope effects in single crystalline manganites. It had never been possible to obtain a high percentage of 18O substitution for a large single crystal. Victor successfully made the oxygen-isotope substitution for large (5mm x 5mm x 1mm) single crystals of manganites (reaching about 85% 18O substitution). He is a co-authored in three publications [137—139] and went to Michigan State University for a Ph.D. Carlos Sanchez who worked in Bernal’s group, did the specific heat measurements on the 16O and 18O crystals of Nd0.5Sr0.5MnO3. We found sharp specific heat anomalies at TCO when the measurements were taken upon warming. The anomaly was found to be negligibly small for polycrystalline samples [140], which may be attributed to inhomogeneity in those samples. We also found that the entropy change associated with the charge ordering increases rapidly with increasing TCO, which can be tuned by the oxygen-isotope substitution and magnetic field. More intriguingly, when data were taken upon cooling, the specific heat anomalies became negligibly small, which appears to be consistent with a first-order structural phase transition [121].