Microreactors for processing native materials
Nick Szabo
from Earth, is crucial to future efforts at large-scale space
industrialization and colonization. At that time we will be
using technologies far in advance of today's, but even now
we can see the technology developing for use here on earth.
There are a myriad of materials we would like to process,
including dirty organic-laden ice on comets and some asteroids,
subsurface ice and the atmosphere of Mars, platinum-rich
unoxidized nickel-iron metal regoliths on asteroids, etc.
There are an even wider array of materials we would like to
make. The first and most important is propellant, but
eventually we want a wide array of manufacturing and
construction inputs, including complex polymers like Kevlar
and graphite epoxies for strong tethers.
The advantages of native propellant can be seen in two
recent mission proposals. In several Mars mission proposals
[1], H2 from Earth or Martian water is chemically processed
with CO2 from the Martian atmosphere, making CH4 and O2
propellants for operations on Mars and the return trip to Earth.
Even bringing H2 from Earth, this scheme can reduce the propellant
mass to be launched from Earth by over 75%. Similarly, I
have described a system that converts cometary or asteroidal
ice into a cylindrical, zero-tank-mass thermal rocket.
This can be used to transport large interplanetary payloads,
including the valuable organic and volatile ices themselves
into high Earth and Martian orbits.
Earthside chemical plants are usually far too heavy to launch
on rockets into deep space. An important benchmarks for plants
in space is the thruput mass/equipment mass, or mass thruput
ratio (MTR). At first glance, it would seem that almost any system
with MTR>1 would be worthwhile, but in real projects risk must be
reduced through redundancy, time cost of money must be accounted for,
equipment launched from earth must be affordable in the first
place (typically <$5 billion) and must be amortized, and
propellant burned must be accounted for. For deep-space
missions, system MTRs typically need to be in the 100-10,000
per year range to be economical.
A special consideration is the operation of chemical reactors
in microgravity. So far all chemical reactors used in
space -- mostly rocket engines, and various kinds of life
support equipment in space stations -- have been designed
for microgravity. However, Earthside chemical plants incorporate
many processes that use gravity, and must be redesigned.
Microgravity may be advantageous for some kinds of reactions;
this is an active area of research. On moons or other plants,
we are confronted with various fixed low levels of gravity
that may be difficult to design for. With a spinning tethered
satellite in free space, we can get the best of all worlds:
microgravity, Earth gravity, or even hypergravity where desired.
A bigger challenge is developing chemical reactors that
are small enough to launch on rockets, have high enough
thruput to be affordable, and are flexible enough to
produce the wide variety of products needed for space
industry. A long-range ideal strategy is K. Eric
Drexler's nanotechnology [2]. In this scenario small
"techno-ribosomes", designed and built molecule by molecule,
would use organic material in space to reproduce themselves
and produce useful product. An intermediate technology, under
experimental research today, uses lithography techniques
on the nanometer scale to produce designer catalysts and
microreactors.
Lithography, the technique which has made possible the rapid
improvement in computers since 1970, has moved into the deep
submicron scale in the laboratory, and will soon be moving
there commercially. Lab research is also applying lithography
to the chemical industry, where it might enable breakthroughs to
rival those it produced in electronics.
Tim May has described nanolithography that uses linear arrays of
1e4-1e5 AFM's that would scan a chip and fill in detail to 10 nm
resolution or better. Elsewhere I have described a class
of self-organizing molecules called _nanoresists_, which make
possible the use of e-beams down to the 1 nm scale. Nanoresists
range from ablatable films, to polymers, to biological
structures. A wide variety of other nanolithography techniques
are described in [4,5,6].
Small-scale lithography not only improves the feature density of
existing devices, it also makes possible a wide variety of new
devices that take advantage of quantum effects: glowing nanopore
silicon, quantum dots ("designer atoms" with programmable
electronic and optical properties), tunneling magnets, squeezed
lasers, etc. Most important for our purposes, they make possible
to mass production of tiny chemical reactors and designer catalysts.
Lithography has been used to fabricate a series of catalytic
towers on a chip [3]. The towers consist of alternating
layers of SiO2 4.1 nm thick and Ni 2-10 nm thick. The deposition
process achieves nearly one atom thickness control for both SiO2 and Ni.
Previously it was thought that positioning in three dimensions
was required for good catalysis, but this catalyst's nanoscale 1-d
surface force reagants into the proper binding pattern. It
achieved six times the reaction rate of traditional cluster catalysts
on the hydrogenolysis of ethane to methane, C2H6 + H2 --> 2CH4.
The thickness of the nickel and silicon dioxide layers can be varied
to match the size of molecules to be reacted.
Catalysts need to have structures precisely designed
to trap certain kinds of molecules, let others flow through,
and keep still others out, all without getting clogged or
poisoned. Currently these catalysts are built by growing
crystals of the right spacing in bulk. Sometimes catalysts
come from biotech, for example the bacteria used to grow
the corn syrup in soda pop. Within this millenium (only 7.1
years left!) we will start to see catalysts built by new
techniques of nanolithography, including AFM machining,
AFM arrays and nanoresists Catalysts are critical to the oil industry,
the chemical industry and to pollution control -- the worldwide
market is in the $100's of billions per year and growing rapidly.
There is a also big market for micron-size chemical reactors.
We may one day see the flexible chemical plant, with hundreds of
nanoscale reactors on a chip, the channels between them
reprogrammed via switchable valves, much as the circuits
on a chip can be reprogrammed via transitors. Even a
more modest, large version of such a plant could have a
wide variety of uses.
Their first use may be in artificial organs to produce
various biological molecules. For example, they might replace or
augment the functionality of the kidneys, pancreas, liver, thyroid
gland, etc. They might produce psychoactive chemicals inside the
blood-brain barrier, for example dopamine to reverse Parkinson's
disease. Biological and mechanical chemical reactors might
work together, the first produced via metaboic engineering[7],
the second via nanolithography.
After microreactors, metabolic engineering, and nanoscale catalysts
have been developed for use on Earth, they will spin off for use in
space. Microplants in space could manufacture propellant, a wide variety
of industrial inputs and perform life support functions more efficiently.
Over 95% of the mass we now launch into space could be replaced by these
materials produced from comets, asteroids, Mars, etc. Even if Drexler's
self-replicating assemblers are a long time in coming, nanolithographed
tiny chemical reactors could open up the solar system.
====================
ref:
[1] _Case for Mars_ conference proceedings, Zubrin et. al.
papers on "Mars Direct"
[2] K. Eric Drexler, _Nanosystems_, John Wiley & Sons 1992
[3] Science 20 Nov. 1992, pg. 1337.
[4] Ferry et. al. eds., _Granular Nanoelectronics_, Plenum Press 1991
[5] Geis & Angus, "Diamond Film Semiconductors", Sci. Am. 10/92
[6] ???, "Quantum Dots", Sci. Am. 1/93
[7] Science 21 June 1991, pgs. 1668, 1675.
--
Nick Szabo sz...@techboook.com