While there is a lot of discussion on Nanodot and links
therefrom about the Hewlitt-Packard patents on wiring
up molecular electronics at least for the next few
more years its good old Silicon & Germanium that will
The Science paper cited below suggests that current
techniques may be extended. As I've wrote a while ago
one of the problems facing the industry is a need for
an insulator with a higher dielectric constant than
SiO2. It looks like they may get it with things "crystalline
oxide semiconductors", that are a combination of
Ba-SrO and SrTiO3 on Silicon or BaTiO3 on Germanium.
Of major significance is the possibility that they may
be able to construct field-effect transistors on Germanium
(something previously impossible because of the difficulty
of getting an insulator made out of germanium oxide).
That would be significant because electrons have a much
higher velocity in germanium compared with silicon.
If IBM incorporates this technology into their Si-Ge
process it will likely provide a way for them to increase
the amount of Ge used and probably push the transistors
past the 50-100 GHz currently available.
R. A. McKee et al "Physical Structure and Inversion Charge
at a Semiconductor Interface with a Crystalline Oxide"
Science 293(5529):468-71 (20 Jul 2001)
Silicon:germanium-based mixed-signal technology for
optimization of wired and wireless telecommunications
B. S. Meyerson, IBM J. R&D 44(3) (Dec 1999)
HIGH DIELECTRIC CONSTANT GATE OXIDES FOR FUTURE CMOS
This archive was generated by hypermail 2b30 : Fri Oct 12 2001 - 14:39:54 MDT