|
THIS YEAR A NEW research program into the nature of electrical
conduction in quantum semiconductor transistors has been initiated. These
transistors, which form the basis of many high-speed electronic systems
from computers to mobile telephones, rely on the ability to conduct
electricity, controllably and reproducibly, such
that they can switch between metallic (ON) and insulating (OFF) states.
However, the fundamental question of whether the highly mobile two dimensional
(2D) sheet of charge carriers in a quantum semiconductor transistor behaves like
a metal or an insulator and, what drives it between these two states,
remains unanswered.
The discovery of a metallic state of matter in exceptionally high quality
2D silicon transistors, and more recently in high quality GaAs hole transistors,
has sparked intense international interest into the method of conduction
in semiconductor systems. Early, well-established theoretical works,
supported by experimental evidence, predicted that there could be no metallic state in
2D systems. Its unexpected discovery therefore challenges our
basic understanding of conduction in highly pure quantum transistors. At
present there is no consensus as to the nature of the metallic state.
|
It is important to be clear about the difference between a metal and
an insulator. The distinction is only properly made at the absolute zero
of temperature, since thermal excitations can permit an insulator to carry
a current. If the resistance is finite at T=0 then the system is metallic, otherwise
it is insulating.
The importance of phase coherent effects had not previously
been recognised because the mean free path in these high quality systems is so
large that weak localisation is only observable at very low, often
inaccessible, temperatures. We directly measured weak localisation effects in the
metallic state and found that they moved to lower temperatures as we went
further into the metallic regime. These
results suggest that the metallic behaviour is a finite temperature effect and that there
is no true quantum phase transition. However, much work remains to
be done before we can answer the apparently simple question is
it possible for a two-dimensional system to be a metal?
Alex Hamilton &
Michelle Simmons
|
|
Figure 2: A phase diagram for the metallic state for
two different electric fields, where increasing the electric
field has the same effect as increasing the temperature.
[A.R. Hamilton, M.Y. Simmons et al., Phys. Rev.
Letters 82, 1542 (1999).]
|
