Maybe now we can use a dislocation as a memory-bit.

By the application of an electric field at room temperature, Krzysztof Szot and colleagues reveal that even undoped SrTiO₃ single crystal could show bistable switching of the conductance between non-metallic and metallic behaviour originating from electrochemically reversible transport of oxygen up or down the dislocation [ 1 ]. The high density dislocations within the skin region [ 2 ] can act as short-circuit paths for oxygen transport and drive the material into a macroscopically detectable metallic state [ 3 ].

Note a dislocation that initially displays metallic conductivity, application of an electric field above a critical threshold results in the dislocation becoming resistive. When a field of opposite sign is applied, the dislocation returns to a metallic state (Fig.1). This is different from the past findings that the bistable switching is a bulk property of SrTiO₃ and only observed in the chromium-doped materials [ 4a, 4b ]. The researchers describe their results in the May, 2006 issue of Nature Materials.

Edge dislocations, as quasi-one-dimensional defects within a crystal (Table 1), introduce nanoscale inhomogeneities and destroy the long-range order of crystal lattice. Edge dislocations can be visualised as being formed by adding an extra half-plane of atoms to a perfect crystal, so that a defect is created in the regular crystal structure along the line where the extra half-plane ends (dislocation line). Defects may be desirable or undesirable. In silicon microelectronics, silicon transistors relay on controlled "doping"  which inevitably introduce large amounts of defects. 

Figure 1 Resistance switching of single dislocations.

Table 1. Crystalline defects.

 Figure 2 Typical a[100] edge dislocation of SrTiO3

Dislocation core followed the simulation by Zhang ZL and coworkers [ 6(a) ]

It is known that the dislocations have an important influence on the measured macroscopic properties due to the related local lattice distortion. Recent studies on the transition-metal oxides with perovskite structure, such as titanates, zirconates or manganites, have indicated that they have great variability in their electrical properties induced by external control. It has been shown that a deviation in stoichiometry, especially by simply modifying the concentration of oxygen at defects through external stimuli, and in geometry (lattice distortion) could induce additional electrical properties such as conductivity (form electrically conducting paths along the dislocation lines)  [ 3 , 7 ] and ferroelectricity [ 8 ]. The dislocation lines serve as the highway for an enforced diffusion of the dopant elements (so-called pipe diffusion), and result in the formation of conducting nanowires in an otherwise insulating bulk crystal [ 9 ]. 

Angus Kingon [ 10 ] feels that the individual switching of dislocations in SrTiO₃ may offer just the right alternative for high-density memory devices, possibly even suggesting a configuration for the 'ultimate' non-volatile memory for high-storage density applications. However, the write speed is somewhat low because the switching is based on the oxygen diffusion along the dislocation. Also there is problem for the reliability due to the stability of the dislocations. Perhaps the most substantial to be considered, as Kingon said, is the need to control the density and precise location of the dislocations. Unfortunately, it seems impossible now. Even we can do it someday; the industry requisition of the cost-effectively parallel fabrication could feel that the memory device based on the porous alumina film may be a competitor [ 11 ].

Of course, the demonstrated work of Szot et al. is extremely elegant in physics. The potential for tuning the dislocations in nanoscale regime provides new access to functional materials beyond their natural phases. □

* Lin PU is in the Physics Department of Nanjing University, Nanjing 210093, CHINA

1 Institut für Festkörperforschung, Forschungszentrum Jülich, 52425 Jülich, Germany
2 Institute of Physics, University of Silesia, 40-007 Katowice, Poland
3 Institut für Werkstoffe der Elektrotechnik, RWTH Aachen, 52056 Aachen, Germany
Correspondence to: Krzysztoof Szot | Rainer Waser   

References