Multiferroics turned upside down

Experiments demonstrating the inversion of entire domain patterns in multiferroic crystals highlight just how versatile this class of materials is, and indicate a route to exploring novel functionalities.

Multiferroics have something for everyone. These materials exhibit both magnetic and ferroelectric order, and as the two forms of order are coupled in these systems — meaning that, for instance, magnetisation can be controlled using electrical fields — a tremendous range of potential applications seem possible, from switches to memory devices. At least in principle, that is. In practice, however, controlling these materials and tapping into this wealth of possibilities is everything but straightforward. The existence of several order parameters may well provide a knob for novel functionality, but at the same time this complexity makes multiferroics notoriously difficult to understand. Writing this week in Nature [1], an international team led by Manfred Fiebig at ETH Zurich and including Naëmi Leo, Jonathan White and Michel Kenzelmann in the NUM division at PSI, present an innovative route to taming this complexity: they demonstrate how combining different order parameters in these materials enables functionality that is not practicably possible in materials with only ferromagnetic or ferroelectric properties, and they introduce a general framework that might guide further research.

General framework explaining magnetoelectric inversion of domain patterns: one parameter contains the information about the domain pattern (bottom layer), and a second parameter (middle layer) can be switched. That switching leads then to an inversion of the entire pattern (top layer). (Image: Naëmi Leo)

Specifically, Leo (a former PhD student in Fiebig’s group and one of the two lead authors of the paper) and her colleagues found a novel strategy for inverting ferromagnetic and ferroelectric domain patterns. Inverting the polarity in each of such domains without changing their shape and distribution is something hard to achieve in conventional ferroic materials. In a ferromagnet, for example, applying a global magnetic field creates a single domain, and flipping spin by spin is barely practical. However, using the coupling between magnetization and electric polarization in the material Co3TeO6, Leo et al. showed now that they can use a global external magnetic field oriented perpendicular to the inherent magnetisation of the crystal to efficiently invert the sign in each domain, without changing the overall pattern.

New levels of functionality

Taking this concept further, the researchers also combined order parameters in the multiferroic material Mn2GeO4, in such a manner that by changing the strength of an applied magnetic field, the sign of polarization in ferroelectric domains could be changed. Also in this case the overall domain pattern remained almost unchanged. Mn2GeO4 is a system with a long history of study at PSI, with Kenzelmann and White having been involved in several of the original studies establishing the material as a ferromagnetic ferroelectric [2] with unconventional magnetoelectric couplings [3]. In those works, Mn2GeO4 was shown to be characterised by a total of four order parameters. The demonstration now that these can be combined to achieve domain-pattern inversion points to new levels of functionality.

Leo et al. explain their concept of magnetoelectric inversion of domain patterns within a general framework (see the figure): one parameter contains the information about the domain pattern (bottom layer), and a second parameter (middle layer) can be switched. Based on free-energy considerations, that switching leads then to an inversion of the entire pattern (top layer). Exploring this framework and applying it to other multiferroics could help discovering and ultimately exploiting a broad range of effects and functionalities that currently remain hidden in the complexity of these systems.

This work is a collaboration between researchers at ETH Zurich, Switzerland; the Paul Scherrer Institute, Villigen, Switzerland; the Helmholtz-Institut für Strahlen- und Kernphysik, Bonn, Germany; Stockholm University, Sweden; the Université de Picardie, Amiens, France; the High Energy Accelerator Research Organization (KEK), Tsukuba, Japan; the University of Tokyo, Japan; the Karpov Institute of Physical Chemistry, Moscow, Russia; TU Wien, Vienna, Austria; and the Norwegian University of Science and Technology (NTNU), Trondheim, Norway.