Manufacturing quantum computers

-----BEGIN PGP SIGNED MESSAGE-----

Hi all,

Regarding the quantum cryptography thread: Some people expressed the
view that although the idea may be workable, the manufacturing technology
was a long way of being able to produce quantum computers.
The quantum dot design proposed by Eckert et al for a quantum "factorisation
engine" requires 100000 quantum dots (to factorise RSA-129 or similar)
which are each about 10nm across to be fairly densly packed onto a 
chip. I'm not so sure about this being 'a long way off' in terms of materials
technology, and I found the following article on WWW which describes some
relevant research into this kind of manufacturing at a US army research
centre. I found this on the FedWorld (US Govt, http://www.fedworld.gov/ )
pages....

============================================================
Part of the "unique ARL (Army Research Lab) facilities" page
http://info.arl.army.mil/UAFD/uafd.html
============================================================
Ultralithography Center.

This facility provides the Army
and DoD with a capability that significantly
impacts the development of novel ultra-submicron
electronic device technology required by
next-generation and future military systems. A
multi-million-dollar Leica (Philips) EBPG-5HR
electron-beam lithography system, which represents
the state of the art in ultra-high-resolution
device patterning, is housed in a newly
constructed clean room. The Leica EBPG-5HR
accomplishes lithography by scanning a finely
focused beam of high-energy electrons over an
electron-sensitive polymer coating applied to the
substrate surface. With an electron beam spot size
approaching 10 nm, the Leica EBPG-5HR provides an
effective solution to the most demanding
lithographic applications, including ultra-small
(250 angstrom [= 25nm]) electronic and photonic devices.
These devices will provide an enabling foundation
for next-generation Army electronic and
optoelectronic systems. Devices such as sub-100-nm
gate length millimeter-wave high-electron-
mobility transistors (HEMTs), novel quantum-effect
and mesoscopic (phase-preserving) devices, and
quantum-well infrared photo-detectors are
patterned with this system. The instrument's high
acceleration voltage (100kV) provides a unique
capability to pattern closely spaced
nanometer-scale device features without feature or
sample- related distortion. Automated calibration
permits the writing of undistorted patterns over
the large sample areas required to fabricate many
photonic devices and optoelectronic circuits.

=======================================================

Well there you go. Perhaps the technology is nearer than we
think. Of course this still doesn't answer the
noise problem that critics of the proposed technique think will
render it all but useless. However I wanted to make the point
that we shouldn't be complacent about the materials technology
side of things.

Tim May makes the point that he is not selling his shares in intel,
and that conventional chip technology is not about to be supplanted.
The problem with this IMHO is that these are not two completely
different technologies we are talking about. The constantly improving
techniques being used to cram more and more onto conventional chips 
are directly applicable to the manufacture of these proposed quantum
dot devices.

Just my 2 cents

Sherry

Sheridan C. Mayo         | WWW pages include caving, sci fi and   
RSC, ANU, Australia      | crypto pages. NEW: X-files has its own
scmayo@rschp2.anu.edu.au | page with gifs/sounds/fanfic etc.
Finger for PGP key       | http://rschp2.anu.edu.au:8080/local.html




-----BEGIN PGP SIGNATURE-----
Version: 2.6

iQCVAgUBLo+G9uFu4n6w1qeBAQGVMgP+Kechf44WUe11qnQG5cD3Ybf+NuNc9jjr
ajI7ZXYmZgQb1xdhS7ruy+UOo39zBCPxgKOaCahAniKV9vlNOmHB2pqAr8aYoMWt
olhDdZdEWSGrLPAvfh4gVa/T8GI9C2NPc7kusIZujlVHnemBbSSz6FW+dJedR/FE
oRKvzzW0IHs=
=Ub7K
-----END PGP SIGNATURE-----

Sherry Mayo writes
> Regarding the quantum cryptography thread: Some people expressed the
> view that although the idea may be workable, the manufacturing technology
> was a long way of being able to produce quantum computers.
> The quantum dot design proposed by Eckert et al for a quantum "factorisation
> engine" requires 100000 quantum dots (to factorise RSA-129 or similar)
> which are each about 10nm across to be fairly densly packed onto a 
> chip. I'm not so sure about this being 'a long way off' in terms of materials
> technology,


Current art is fairly close to making components whose interaction
requires a full quantum description.

To make a quantum computer from such components requires that
that the components be orders of magnitude faster and more
reversible.

Presently known quantum algorithms cannot tolerate the loss of 
a single quantum of energy, as this will introduce
vacuum noise into the data.

The longer the algorithm takes, the less energy there
is in a single quantum of energy, thus the components
for any long quantum algorithm, such as factoring a 1024 bit
number, must be very fast indeed (near infrared frequencies)
and extraordinarily efficient (fully reversible classical,
non quantum computation.)

Although quantum computers are interesting and important,
they have no immediate practical relevance to cryptography.

Error tolerant algorithms could change the picture substantially,
but they would still require components far beyond current
art.




-- 
 ---------------------------------------------------------------------
We have the right to defend ourselves and our
property, because of the kind of animals that we              James A. Donald
are.  True law derives from this right, not from
the arbitrary power of the omnipotent state.                jamesd@netcom.com