Picotech? Femtotech?
Hugo de Garis
Dear Nanists,
On one of Drexler's recent trips to Japan, my boss and I
invited him to dinner and talked about nano. Towards the end of the meal,
I brought up the possibility of the existence of a smaller scale technology,
which would be to nanotech what nanotech is to microtech (or current
technology). This email challenges readers to consider the possibility of
a picotech, or even a femtotech. Presumably a picotech would be of the
scale of a nucleus (a hydrogen atom is about 0.1 nmeter, a proton is about
0.1 pmeter), and a femtotech would be into quark territory. Ask your
nuclear and particle physics friends, if it is feasable in principle to
have an engineering at such scales. Drexler was sceptical. If you feel such
scales are too small for engineering, lets hear why, so that others can
use your arguments as a target for refutation.
Cheers,
Hugo de Garis.
Dr. Hugo de Garis
Brain Builder Group,
Evolutionary Systems Department,
ATR Human Information Processing Research Laboratories,
2-2 Hikari-dai, Seika-cho, Soraku-gun,
Kansai Science City,
Kyoto, 619-02, Japan.
tel. + 81 7749 5 1079
fax. + 81 7749 5 1408
email. deg...@hip.atr.co.jp
[Pico-tech would presumably be to nanotech as the biology of R. Forward's
"Dragon's Egg" neutron-star critters to ours... and that's probably where
it'd have to take place. Does Dr. Forward read this list?
--JoSH]
Jack Boyce
>
> Presumably a picotech would be of the
>scale of a nucleus (a hydrogen atom is about 0.1 nmeter, a proton is about
>0.1 pmeter), and a femtotech would be into quark territory.
>
For the record this is incorrect. A proton is roughly 100 times smaller
than this (around 1 femtometers (= 1 fermi)), and a typical nucleus is
a few times larger.
I don't see a fundamental reason why it should be impossible. However,
consider the following:
1) Matter is fully quantum-mechanical at this level. To be more
precise, the deBroglie wavelength of a proton (with a reasonable
energy) is comparable to the distance between particles in the
system. At the atomic level the atomic deBroglie wavelength is
very small, so we can think of atoms as classical particles to
an extent. Not so with nuclear matter. At some point it really
becomes meaningless to think of a chunk of nuclear matter as
composed of baryons, each one a little sack containing 3 quarks.
Indeed, it's thought that for very heavy nuclei the protons and
neutrons will in effect blend together in a "quark-gluon plasma",
still unobserved.
2) The physics is much more complicated for protons than for
electrons, since the former have structure (and feel the strong
nuclear force) whereas the latter are point particles as near as
we can tell. It ought to be possible, from first principles, to
predict nuclear properties given the standard model (QCD, etc.),
but this task has proven to be very difficult computationally.
The bottom line is, to do engineering you need predictability, but
so far the problem is too difficult. People use supercomputers to
do even the simplest QCD calculations.
3) Stability. Because the strong force has a very short range, in
large structures containing protons the Coulomb repulsion (long
range) wins out over the short-range strong force. This is why
we can't make too much beyond element 108 (or whatever); the nuclei
get too big, and Coulomb wins out. Ok, so make everything out of
neutrons, you say. But then we run into:
4) Practical problems. Sticking protons together isn't easy, since
they are of like charge. If we use neutrons, the problem is to
find a way of grabbing onto them (can't use EM fields!).
Femtotech would require a complete abandonment of classical notions. For
atomic systems we can draw pictures of structures (as in Drexler's book),
and be sure that our view of things isn't too far off (the atomic
coordinates behave more or less classically). With nuclear matter, we
just can't do that. It becomes a hideous calculation with little
intuition involved, not to mention the problems in building a device.
Jack Boyce
jbo...@physics.berkeley.edu
Stephen Fulton
femtotech like nanotech ie dealing with nuclei and quarks as opposed to atoms.
My reply is probably not :
The problem with nanotechnology the size : we can't see,feel or otherwise
directly sense with what we are dealing with. But,given the right tools we can
still use these tools to manipulate nano-sized objects. This problem also
exists with picotech and femtotech. However,another problem exists with the
latter two technologies : quantum uncertainty. Whereas a group of several dozen
atoms (or even one atom) has a fairly well defined velocity and position
(accurate enough to use them anyway),a proton,electron,quark etc. does not.
Heisenberg states in his uncertainty principle that there is no way,even in
theory,for an observer to know both the position and velocity of a particle to
more than a certain accuracy. Because the numbers used (ie Planck's constant)
are very small this restriction is irrelevant for everyday life,makes things
difficult but possible for nanotech and makes things impossible for picotech.
This is why we cannot say where exactly individual electrons are,just the
probabilities of where they could be.(an electron cloud as opposed to six
electrons)
That's the reason I think it impossible. The other question is,ignoring the
uncertainty principle (which you can't) the particles you would be
manipulating in femtotech are quarks (and electrons and others) which,
according to latest theory cannot be seperated to give one quark at a time.
They can only exist in groups of two or three ie in protons,neutrons et al.
(The strong nuclear force prohibits it.) The same restriction will presumably
apply to combining quarks into groups of more than three. Goodbye femtotech.
Which leaves picotech. Neutrons and protons can exist on their own,neutrons
decaying after approx. 15 minutes,which could cause a problem. However,the
same strong nuclear force that governs quarks also governs nuclei; and states
that above a certain limit (above lead and bismuth in the periodic table) a
group of protons and neutrons ie a nuclei must decay (radioactivity) : the
strong nuclear force is not strong enough to hold them together. Also these
groups of pico sized particles must not include electrons and protons coming
together : they change to give neutrons. Too many restrictions to make it
feasible so Goodbye picotech.
Well,almost. Nature isn't bothered with everyday situations and designs things
like neutron stars. (mentioned by R. Forward) The only reason that they are
possible,and so anything that happens on them is possible,is due to gravity.
Neutron stars are so massive that gravity helps bind particles together and
they stay together (and don't decay). We can't unfortunately use these for any
useful purpose. They must be this massive to survive ie we can't build one and
then bring it into the solar system or the solar system and us die. To survive it
must,by definition,have too much gravity to be able to move it anywhere useful.
We can't go to one and use it or we (and any tools we might have) get torn apart
by gravity,sorted into neutrons,and deposited on the neutron star with no
further possible use. And if we put anything on them we couldn't it (in practice)
back off again due to gravity and strong nuclear force. This apply to working with
any group of protons and neutrons directly : put any tool in direct contact
with them and you can't remove the tool : the strong nuclear force will
attract the particles of the tool to the group.
Sorry for going on at such length,but I do think it is in theory (Heisenberg
Uncertainty Principle) and in practice (see above) impossible.
Stevephen Fulton.
My opinions are not necessarily Edinburgh University's opinions.
Hugo de Garis
Dear Nanists,
Concerning the possibility or otherwise of a picotech and a
femtotech, I would like to make the following simple point -
One of Drexler's arguments in favor of the possibility of a nanotech is the
existence proof of molecular biology. We know that molecular scale assemblers
are possible because nature has already made them in the form of ribosomes.
We know that molecular scale computers and molecular scale software are
possible because nature has already made them in the form of DNA and its
accoutrements. By analogy, one could use the same reasoning to plea for the
possibility of a picotech and a femtotech, in saying that one can build with
elementary particles because nature has already built nuclei.
However, the analogy may not be a good one, if the way in which nature built
its picotech and femtotech is so constrained, that there may only be very
few options open to 21st century picoengineers. For example, if one wants to
build with neutrons and if the only way to do that is to place them on the
surface of a neutron star....
What frightens me is that if a picotech (and hence a femtotech) is impossible,
then late into the 21st century, engineers will be feeling a definite
barrier, a sort of "hemmed in" or "edge of the world" mentality. Right now
nanotech is seen as a wonderful opportunity, a doorway into the arbitrary
control of matter. But if a picotech is impossible, then nanotech is the
end of the line. Nanotech will be seen as the last frontier.
Relativity tells us that it is impossible for matter to travel faster than
light speed, so many people argued the distant stars will forever be out
of our reach. Yet recently, astrophysicists have shown how in theory it
is possible to design worm holes which will allow a rocket to pass thru
without being crushed. Thus the speed of light limit can be circumvented.
If picotech is thought to be impossible for reason X, then maybe there is
a "way round" the problem.
Sooner or later, the question "is picotech possible" will become fundamental.
Most of future technology will depend upon a positive answer to this question.
John Hagerman
jbo...@physics2.berkeley.edu (Jack Boyce) writes:
>
> From a different angle, you would have to wonder: why pursue femtotech?
> Nanotech gives you so much capability in such a small volume it's hard
> to forsee a reason why somebody wouldn't be content with it. Then again,
> I remember thinking that the average person would never need >64K for
> their PC...
Some may like thinking about femtotech for the same reason they like
thinking about nanotech: because it is technologically interesting.
However, Drexler points out what may be a more important reason for
thinking about nanotech: social impact. Whether nanotech is achieved
in 5 years or 50, it is important to think NOW about the consequences.
I don't think the same immediacy applies to femtotech.
- John
Jack Boyce
>
> However,another problem exists with the
>latter two technologies : quantum uncertainty. Whereas a group of several dozen
>atoms (or even one atom) has a fairly well defined velocity and position
>(accurate enough to use them anyway),a proton,electron,quark etc. does not.
>
The uncertainty principle does not make computing impossible, it simply
forces you to revamp your conceptions of what a computer is. The topic
of "quantum computation" is a hot one these days. Instead of envisioning
a classical Turing-machine-like process, you think of setting up an
initial wavefunction which evolves according to some Hamiltonian you have
suppied (this is the program). You then make a measurement of the
wavefunction at some later time to get your result. In reality there
will be a range of possible results you can get at the end, of
course, but with ensemble averaging you can remove this type of quantum
randomness to arbitrary precision.
The interesting theoretical question is: Does this type of computer allow
one to calculate anything more efficiently than a classical machine? Is
there some NP-complete problem out there which can be solved in polynomial
time by a quantum computer? The jury's still out, as far as I know.
From a different angle, you would have to wonder: why pursue femtotech?
Nanotech gives you so much capability in such a small volume it's hard
to forsee a reason why somebody wouldn't be content with it. Then again,
I remember thinking that the average person would never need >64K for
their PC...
Jack Boyce
jbo...@physics.berkeley.edu