Soutenance de thèse de Marine Verseils

Marine Verseils, doctorante dans l'équipe DEMARE, soutiendra sa thèse le jeudi 19 octobre 2017 à 14 h.

IMPMC - UPMC - 4 place Jussieu, 75005 Paris, tour 23, 4e étage, couloir 22-23, salle 401

Abstract

In the present Thesis, we address the open issue of the unusually large ferroelectricity induced by magnetism recently reported in (AMn3)Mn4O12 quadruple perovskites. To do so, we focus our study on the isostructural and isoelectronic compounds (LaMn3)Mn4O12 (LMO) and (YMn3)Mn4O12 (YMO), for they display single-valent Mn³⁺ properties and a commensurate C-type antiferromagnetic structure of the B-site Mn³⁺ ions. These simple features offer an ideal playground to elucidate the controversial contribution of the exchange striction and of the antisymmetric exchange to the electric polarization. YMO and LMO are both metastable phases stabilized under high-pressure; YMO is a new phase, not reported before, where the comparatively small Y³⁺ ion exerts a large chemical pressure, which is expected to enhance the exchange interaction and, thus, the spontaneous polarization as compared to LMO.

According to this expectation, in YMO we find an ordering temperature, TN,B = 108 K, of the B-sites, 30 K higher than in LMO. On the other hand, we surprisingly find identical values of the spontaneous polarization, P = 0.54 μC cm^-2, in both compounds. To the best of our knowledge, this is a record value for magnetic ferroelectrics, two times larger than the previous record value reported in (CaMn3)Mn4O12. This result is very promising for multiferroic applications considering that it is obtained in polycrystalline samples and 6-10 times larger values are expected in single crystals of epitaxial films.

In spite of the similar structural and electronic properties and the identical polarization found, the properties of magnetic ferroelectricity appear to be very different in the two compounds. Namely, in LMO, ferroelectricity occurs at the magnetic transition at TN,B=78 K, although no indication of inversion symmetry breaking is detected either by x-ray diffraction or by a combined Raman and IR study. We argue that this puzzling observation is consistent with Levanyuk and Sannikov’ prediction of domain structure corresponding to degenerate solutions of the multi-component order parameter, which is characteristic of improper ferrolectrics. On the other hand, in YMO, the observation of ferroelectricity at T*=70 K is consistent with a polar structural modulation below Ts=200 K, however T* does not correspond to any long-range magnetic order. Indeed, the ordering of the B-sites occurs well above, at TN,B=108 K, while T* marks a magnetic anomaly consistent with short-range magnetic correlations and suggesting a latent magnetic phase. We put forward the hypothesis that the above polar distortion of the crystal structure occurring may force the alignment of polar domains that would be otherwise randomly oriented because of the random orientation of the magnetic domains.

The above results prompt us to carry out further work that may: (i) lead to the achievement of even larger polarizations well above 1 μC cm^-2 in single crystals or epitaxial films, which is a prerequisite of multiferroic applications; (ii) provide further hints as to the role of polar and magnetic domains and of short-range magnetic correlations in the stabilization of such large polarizations in improper ferroelectrics.

Résumé

En dépit d’une structure et de propriétés électroniques similaires, ainsi qu’une même valeur de P, les propriétés de ferroélectricité magnétique semblent très différentes dans les deux composés. Plus précisément, dans LMO, la ferroéléctricité apparaît à la transition magnétique des sites B, à TN,B = 78 K bien qu’aucune indication de brisure du centre d’inversion n’ait pu être reportée par diffraction de rayons X ou par spectroscopie Raman et infrarouge. Nous montrons que ce résultat inattendu est en accord avec la théorie de Levanyuk et Sannikov, qui prévoie l’apparition d’une structure de domaines polarisés du fait de la dégénérescence des solutions pour le paramètre d’ordre dans le cas des ferroélectriques impropres. D’un autre côté, l’observation de la ferroélectricité à T* = 70 K dans YMO est cohérente avec la modulation structurale apparaissant à Ts = 200 K. Cependant, T*  ne correspond pas à l’apparition d’un ordre magnétique à longue portée. En effet, l’orientation des sites B a lieu bien plus haut, à TN,B = 108 K, et à T*  il apparait seulement une anomalie magnétique qui peut être expliquée par des corrélations magnétiques ce qui suggère l’existence d’un ordre magnétique latent. Nous émettons l’hypothèse que la distorsion polaire de la structure qui a lieu à Ts est responsable de l’alignement des domaines ferroélectriques qui seraient sans cela aléatoirement orientés du fait de l’absence d’orientation à longue portée des domaines magnétiques.

Jury

• Isabelle Mirebeau (Research Director) - LLB (CEA-CNRS) - Reviewer
• Maxim Mostovoy(Professor) - University of Groningen - Reviewer
  
• Agnès Barthélémy(Professor) - CNRS/Thales - Examiner
• Valia Voliotis(Professor) - INSP - Examiner
• Gianluca Calestani(Professor) - Chemistry Department (Univ. Parma) - Examiner
• Andrea Gauzzi(Professor) - IMPMC/UPMC - Director
• Edi Gilioli(Researcher) - IMEM-CNR (Parma) - co-Director