Aller au menu Aller au contenu
Physico chemistry of solids, thin films, biotechnologies
Applications for micro & nano- technologies, energy, health ...

> Research > MSTIC

Multiferroic heterostructures

Updated on November 30, 2009
A+Augmenter la taille du texteA-Réduire la taille du texteImprimer le documentEnvoyer cette page par mail cet article Facebook Twitter Linked In
Multiferroic heterostructures Students: I. Gélard, G. Huot, A. Prikockyte, J. Théry, Objectives Multiferroics research field has been blooming out in 2003/2004 in the international community. In 2006, two STREP projects dedicated to multiferroics were funded by Europe, which of one we are involved in. Our work has been focused on RMnO3 hexagonal manganites, whose properties as thin films were not well known (YMnO3) or even almost unknown, for rare-earths other than yttrium. The bulk compounds are ferroelectric (Tc ~ 900-1100 K) and antiferromagnetic (TN ~ 70-100 K). Main results RMnO3 thin films : The CVD growth of YMnO3, HoMnO3, ErMnO3, TbMnO3 and DyMnO3 thin films was optimized on two types of substrates : ZrO2(Y2O3) (111) for epitaxial growth and Pt/TiO2/SiO2/Si for electrical measurements and integration purposes [C. Dubourdieu et al., Phil. Mag.Lett. 87, 203 (2007)]. Hexagonal TbMnO3 and DyMnO3 films were obtained through epitaxial phase stabilization, while they normally crystallize in an orthorhombic structure. We evidenced, for all rare-earths, a strong effect of the stoichiometry on the structural properties, which had not been previously described in the literature and which explains the discrepancies found in the reported data for lattice parameters. The crystalline structure was further investigated by reciprocal space mapping (xray diffraction) and confirmed a two-layer model. A local study of the defects in these compounds was performed by TEM. Change in the Mn valence and in the local structure is being investigated by x-ray absorption spectroscopy at NSLS synchrotron. Finally, Raman spectroscopy was performed on both single crystals (provided by T. Palstra, Gröningen) and thin films. The results will be compared to ab initio calculation of the phonon spectra performed in the group of Ph. Ghosez at Liège University (co-tutelle Ph-D thesis). The magnetic properties of the films were studied by neutron diffraction. We showed that films with thickness down to 50 nm can be effectively measured. For the first time for such compounds, we clearly evidenced the antiferromagnetic transition in YMnO3, HoMnO3 and ErMnO3 films (Fig. 1.22) [I. Gélard et al., Appl. Phys. Lett. 92, 232506 (2008)]. We also showed that TN decreases with decreasing film thickness, which was correlated to the in-plane strain. We determined the size of the antiferromagnetic domains (20 to 50 nm), which was found to be similar to the one reported for BiFeO3 films. The thickness dependence of the domain size can be described by a power-law of LLK type (Landau, Lifshitz, Kittel), similarly to ferroelectric or ferromagnetic domain sizes. As far as magnetoelectric effect is concerned, no significant changes could be obtained for thin films as compared to bulk hexagonal RMnO3, for which the linear magnetoelectric effect is forbidden by symmetry (~ 0.1 %/T at room temperature). The permittivity of YMnO3 films was measured to be 20, similar to the bulk value. A typical leakage current density of 6x10-3 A/cm2 (at 1 V) was obtained for 10 nm films. We evidenced a strong effect of the rare-earth nature on the leakage current. Finally, ferroelectricity at room temperature for all considered RMnO3 films was demonstrated by second harmonic generation spectroscopy in the group of M. Fiebig at the university of Bonn [T. Kordel et al., Phys. Rev. B (2009) in press]. The local ferroelectric properties are currently investigated by piezo-force microscopy in the group of R. Ramesh at the University of Berkeley. In conclusion, we showed that RMnO3 films, even quite thick (500 nm), exhibit significantly different ferroelectric and magnetic properties than the corresponding single crystals, which has important consequences for applications purposes. New routes have been proposed to tentatively increase the magnetic ordering temperature towards room temperature such as chemical substitution or multilayers approach. 
Figure 1.22 - Neutron diffraction on an epitaxial YMnO3 film (450 nm) : (a) 010 Bragg peak measured at different temperature and gaussien fits. (b) Intensity of the maximum of the 010 peak measured as a function of temperature. Crosses represent the background level. Multilayers for properties enhancement : The growth of multilayers was optimized on both oxide and Pt substrates. Highly c-axis oriented superlattices were obtained on Pt-buffered Si substrates (Fig. 1.23). A giant magnetocapacitance effect, associated to a giant magnetoresistive effect, was obtained in a (YMnO3/HoMnO3)15 superlattice at a temperature close to room temperature (Fig. 1.23). We evidenced that the HoMnO3 films did not behave like a dielectric ; the magnetoimpedance effect is probably related to the conducting nature of these films and to interfaces.
 Figure 1.23 - HRTEM of a (YMnO3 5 u.c. / HoMnO3 12 u.c.)15 superlattice deposited on Pt (111)-buffered Si(001) - Capacity and resistance of this superlattice measured as a function of temperature for different magnetic fields.   Multilayer approach is also of interest in order to study the magnetism in such systems when ultrathin films are stacked. Indeed, it is probable that the magnetic moments of manganese are modified at the interfaces and that canting can induce ferromagnetism. We combined two compounds of different magnetic symmetry : YMnO3 (P63cm) and ErMnO3 (P63cm). Ultrathin layers of few monolayers were stacked in superlattices. The first results obtained by neutron diffraction on (YMnO3 20 nm / ErMnO3 20 nm)10 indicated an important result : the ordering temperature TN measured on the two compounds is not the one expected for single films of same thickness (~ 40 K) but is considerably higher (TN = 80 K). This result needs to be confirmed and the origin for such enhancement will be explored.  Collaborations CRISMAT Caen, EMAT Anvers, ICMAB Barcelone, Institut Néel Grenoble, LLB Saclay, LTM, Grenoble, New Jersey Institute of Technology USA, SAFC Hitech, UK, University of Berkeley, University of Bonn Allemagne, University of Geneva, University of Liège. Funding STREP MaCoMuFi, NoE FAME, GDR Matériaux multiferroïques
A+Augmenter la taille du texteA-Réduire la taille du texteImprimer le documentEnvoyer cette page par mail cet article Facebook Twitter Linked In

Date of update November 30, 2009

Université Grenoble Alpes