Electrical characterization of magnetoelectrical materials
- PDF / 495,244 Bytes
- 10 Pages / 585 x 783 pts Page_size
- 70 Downloads / 228 Views
Electrical characterization of magnetoelectrical materials J.F. Scotta) Centre for Ferroics, Earth Sciences Department, University of Cambridge, Cambridge CB2 3EQ, United Kingdom (Received 9 November 2006; accepted 19 January 2007)
A brief review is given of electrical properties of magnetoelectric, multiferroic materials, with emphasis on magnetocapacitance effects, nanostructures, integration into real random access memories, and critical phenomena, including defect dynamics near phase transitions.
I. INTRODUCTION
Materials that are simultaneously ferroelectric and ferromagnetic were originally termed “magnetoelectric” but more recently have been called “multiferroic.” In addition to being of considerable academic interest because of the ways in which electric polarization P can couple with magnetization M, they are also of potential interest for use in engineering devices as random-access memory (RAM) elements; in this application, they can be read magnetically (nondestructive readout with no reset operation required), and they can be erased and rewritten electrically (faster and less power-consuming than magnetic rewrite). They also have potential device application as weak-magnetic-field sensors; these would be similar in performance to superconducting quantum interference devices but would be operational at ambient temperatures; hence, much cheaper. The search for materials that would be good magnetoelectric materials began in earnest in the late 1950s in Leningrad in the group of Prof. Smolenskii at the Ioffe Institute.1 The materials studied included BiFeO3 and other perovskite oxides, generally doped (most recently with Mn) to make them better insulators and less semiconducting. Transition metal oxides are rarely good insulators because of oxygen vacancies and/or multivalent metal ions.
II. PITFALLS AND ARTIFACTS
Electrical measurements of ferroelectrics in the earlier years (1920–1980) were generally made on bulk insulaa)
Address all correspondence to this author. e-mail: [email protected] This paper is based on tutorial notes used at the MRS meeting in Boston, November 2006. DOI: 10.1557/JMR.2007.0260 J. Mater. Res., Vol. 22, No. 8, Aug 2007
tors. Under these conditions, the measurements of hysteresis were usually performed with a Sawyer–Tower circuit, which actually measured the switched charge Q, typically at a frequency (f) of 50 or 60 Hz. This is important; nature does not provide us with a simple way of directly measuring polarization P. For an ideal insulator, the charge switched is determined by the displacement current dD/dt in the system, where for most ferroelectrics the displacement vector D from Maxwell’s equations is nearly the same as polarization P. In a parallel plate capacitor, the switched charge when polarization is reversed in a ferroelectric is Q ⳱ 2APr, where A is the area of the electrode and Pr is the remanent polarization. If the material is not a perfect insulator, in addition to the displacement current dP/dt there will be a real conduction with current j ⳱ EA where
Data Loading...
