Probing the Dead Layer in Barium Strontium Titanate Capacitors Made by Pulsed Laser Deposition

  • PDF / 179,470 Bytes
  • 6 Pages / 612 x 792 pts (letter) Page_size
  • 97 Downloads / 253 Views

DOWNLOAD

REPORT


Probing the Dead Layer in Barium Strontium Titanate Capacitors Made by Pulsed Laser Deposition L. J. Sinnamon, R. M. Bowman and J. M. Gregg Department of Pure and Applied Physics The Queen’s University of Belfast Belfast BT7 1NN U. K. ABSTRACT

Thin film capacitors with structures Au/Ba0.5Sr0.5TiO3/SrRuO3 were fabricated by pulsed laser deposition onto {001}-oriented MgO single crystal substrates. As part of an overall programme to investigate the so-called ‘dead-layer’ effect, polarisation loops were taken from the capacitors and analysed as a function of BST film thickness. Analysis was done within a Landau-Ginzburg-Devonshire (LGD) framework, and showed that for film thickness greater than ~100nm, changes in the dielectric behaviour of the BST were dominated more by epitaxial mismatch strain, than by the conventional notion of the dead-layer. INTRODUCTION

Since thin film devices using high-dielectric constant materials are of great current interest, the observation that the dielectric constant collapses with decreasing thickness is becoming an issue of fundamental importance. Barium Strontium Titanate (BST), for example, has a maximum bulk relative dielectric constant of 15,000 which falls to just 150 in a 24nm thick device [1]. The decrease in dielectric constant can be effectively modelled by assuming the existence of ‘dead layers’ with severely depressed dielectric constants at the electrodeferroelectric interfaces. These interfacial dead layers act as parasitic capacitors in series with the ‘bulk-like’ ferroelectric. Hence the decrease in dielectric constant is said to follow the ‘series capacitor model.’ Despite the wide acceptance of the dead layer concept, an experimentally consistent explanation for the nature of such layers is not yet evident. Various models have been proposed, but none is definitive nor free from contradictory evidence. Further, while direct imaging has occasionally shown distinction between bulk and interface, usually microscopy shows no evidence of interfacial layers (see [2] and references therein). In an attempt to shed further light on the problem, the authors and their collaborators have embarked upon a three-pronged experimental approach, involving the analysis of the functional properties of Au/BST/SrRuO3/MgO capacitors made by pulsed laser deposition. The approach involves: (i) Detailed mapping of the low-field dielectric constant as a function of thickness to see at what point the series capacitor model breaks down (at what point the BST is completely composed of low-dielectric constant, ‘interfacial’ material). This has been done down to a BST thickness of 7.5nm, with the conclusion that the total ‘dead-layer’ thickness must be less than 7.5nm [2]. (ii) Determination of the transport properties of the capacitors, with particular emphasis on thicknesses below ~100nm (initial work suggests that field-penetration into the SrRuO3 may produce the interfacial capacitance [3]). C13.6.1

(iii) Analysis of the polarisation-field behaviour in the context of Landau-Ginzburg-Devonshire