NIAC: Aerospace products
From Wise Nano
Aerospace Products from Nanoscale Modules
This NIAC project will investigate the design, fabrication, and performance of human-scale aerospace products constructed from aggregates of nanoscale functional units. The goal is to demonstrate that such units can be combined to form useful products, and that such products can be designed and constructed with approaches familiar to current engineering practice. Phase 1 would investigate issues such as large-scale structures, curved sliding interfaces, fractal networks, and automated construction of awkwardly-shaped products.
This web page is our collaborative tool for project work (and will end up as the 3-month report).
Our initial poster for the October NIAC conference is on a separate page.
Problems in Space Flight
- Risk
- Weight
- Difficulty of new designs
Mitigated by general-purpose manufacturing systems.
- Build what you need, when you need it
- Bring along a little feedstock instead of a lot of different replacement parts
- Less redundancy/reliability needed on repairable systems
- Build improved designs not available at start of mission
- Inexpensive rapid prototyping on the ground will improve R&D
Nanoscale manufacturing and products are high performance
- Scaling laws: Small things have higher frequency of operation
- A billion-atom mechanosynthesis robot (~200 nm) might duplicate itself in hours
- With mechanosynthesis, you get extremly high precision
- Small feature size
- Improved materials
- So products are also extremely high performance
- High-bond-density, anisotropic materials can have strength approaching bond strength (not limited by defects)
- Motor power density inversely proportional to size: electrostatic motors 10^9 W/cm^3
- Computer density varies as cube of size: nm features implies Earth Simulator in a mm^3
- Computer efficiency also very high (assuming straightforward use of reversible logic)
Convergent assembly: kilogram-scale products from nanoscale components
Nanoscale fabricators can build nanoscale products efficiently. How to combine those into a large system?
One approach: general-purpose assembly
- Requires lots of motion planning
- and per-product design
- and complicated manipulators
- and awkwardly-shaped parts
Our chosen approach: simple assembly of simple parts
- Cubes with gripping surfaces: just push them together
- Functional interfaces between cubes (power, signal, etc) outlined in Nanofactory 3.2.2: Functional joints
- The smallest 200-nm cube has room for 8 million nanoscale features: enough to build a functional module in a single cube
- 8 cubes make a bigger cube, recursively... 17 levels goes from sub-micron to several cm.
Products to build
Propulsion Engine (jet, rocket, or mass driver)
Jet engine
May be useful for getting out of the atmosphere efficiently. Turbines can accelerate to ramjet speeds, then scramjet; meanwhile the body can be changing shape for optimum performance at each speed.
Rocket engine
A pumped liquid-fueled rocket engine (the most efficient kind, at least if you have efficient pumps) does several things in sequence.
- Take propellants (fuel and oxidizer) from pipes
- Pressurize the propellants
- Inject the propellants into a combustion chamber
- Ignite the propellants
- Expel fast-moving exhaust gases through a carefully designed nozzle
Many fuels and oxidizers are viciously reactive. This means that the pipes and pumps have to be designed to be very inert. Fluorinated polymers can handle liquid oxygen, but not at elevated temperatures. Basically, anything except ceramics will burn in hot high-pressure oxygen. Alumina (sapphire) and silica will be inert, and also will not let oxygen diffuse through them, but are brittle and may pose thermal expansion problems. This NASA publication provides some insights into just how reactive oxygen can be. Section 303 has information on materials. More research will be needed to see if fluorinated diamond can withstand oxygen in the context of rocket engines.
Fuel can be a light hydrocarbon, not especially hard to handle.
Positive-displacement pumps (e.g. piston pumps) will probably be preferable to turbopumps. They scale well at small sizes and are competitive with turbopumps (according to Jordin Kare as seen in various space newsgroups); this implies that, given high reliability, many small pumps can be ganged together. They can be more efficient than turbopumps for two reasons: first, electromechanical drive with electricity generated by fuel cell has higher efficiency than turbine drive which is limited to Carnot efficiency. Second, electromechanical drive can have extremely high power density per mass and volume, even counting the fuel cell (see "Power generation").
The propellant injectors can be standard design. Ignition can use a smaller rocket engine (big fleas have lesser fleas...) as VCOR proposes; the smallest engines can be ignited by spark.
The combustion chamber and rocket nozzle can be cooled by propellant. Diamond has excellent thermal conductivity. Nanostructured diamond should not be suceptible to the same thermal-cycle and thermal-shock damage mechanisms as metals.
Reliability and safety can be greatly enhanced by embedded sensors and redundant components. Every cubic millimeter can be instrumented. Because cost-per-manufactured-feature is essentially zero, sensors and controls can be placed to do the most good without affecting manufacturing cost, and components can be chosen according to the optimum size for scaling laws and reliability. We could imagine one large rocket engine, fed by a thousand small piston pumps, which are powered by a million fuel cell elements.
With at least one order of magnitude strength:weight improvement over aluminum, diamondoid construction would allow extremely lightweight construction. Thus it would be easier to include multiple engines fine-tuned for different conditions of atmospheric pressure, or alternatively multiple fuels. Reconfigurable nozzles, chambers, and bells are not beyond the realm of possibility.
Mass driver
Power source: Solar cells; maybe chemical fuels (Fuel cell can be more efficient than Carnot; does this mean that a chemical-powered mass driver can be more efficient than a rocket engine?)
Passive structure (wing, strut, aeroshell)
Life support equipment
Surface transporters for lunar and planetary missions
Electrical Generator
From fuel: For efficiency, use fuel cells (better than heat engines). The mechanics of a proton exchange membrane (tiny pores) could certainly be constructed by mechanosynthesis, and probably several other nanotechnologies. Question: What kind of surface/electrical properties does a PEM need? Can we do it in just diamondoid? Freitas says in Nanomedicine 6.4.3.5 that the channels should be "lined with atoms of oxygen, fluorine or nitrogen, creating negatively charged channels with high proton affinity." He calculates that a cubic-nanometer channel could pass protons at almost a terawatt per cubic meter.
From sunlight: In outer space where sunlight is never diffused by atmosphere, thermionics may be the simplest approach; CVD diamond thermionics have been demonstrated at 50% efficiency IIRC. Thermionics need a radiator surface. Buckytubes are excellent heat conductors, and amorphous carbon is almost a perfect black-body radiator.
From mechanical motion: An electrostatic motor (Nanosystems 11.7) makes a great generator.
How to combine these into macroscale systems?
Fuel cells could use fractal plumbing; this is mere structure, not hard to do. Should be scalable to kilowatt modules of flexible form factor. How to deal with impurities? With sufficiently thin membranes and large surface area, it may be acceptable to let the fuel cell be the filter. For example, suppose a pore weighs 1000 AMU and can be blocked by a 20 AMU molecule. Then a 1-kg fuel cell will be 50% blocked by 10 grams of contaminant. But if the fuel is clean to 1 PPM, the fuel cell can process 10 tons of fuel before that happens.
Solar panels could be very thin and somewhat modular. You'd want to focus sunlight (using a holographic lens) onto a small spot, which would shoot electrons onto a target that would have to be cooled by thermal connection to a radiator. How small could this be made? Need calculations to see what the temperature of a flat blackbody radiator would be, if warmed by earth-orbit solar flux. If that's too high for efficiency, then you need a bigger structure--a large fin, or small modules spaced apart. But if a flat surface is OK for cooling, then just string flat-back modules together to make a panel.
Since electrostatic motor/generators can convert power at a petawatt per cubic meter, there's usually little need for efficient use of space. Fractal gear trees would be efficient. But simply putting a layer of traction-drive motors around a shaft would normally be enough. (Note to self: verify this with calculations.) Questions of environmental seals around rotating shafts are deferred to another part of the project.
Actuators
Computers and networking
A 200-nm nanoblock has room for maybe 20 kB of RAM or 20,000 logic gates, enough for an 8086-class CPU. That's not much. But there's room on a 200x200 nm face for thousands of pincushion rod-logic connections. And the switching time is about 100 ps, for a clock rate of 1 GHz. So for simple tasks, a high-speed microcontroller could be put in a single block. For more complex tasks, coprocessors could be added, or modular processor architectures developed.
Signals can be sent micron distances by mechanical motion. For longer distances, charge pulses could be generated and detected mechanically, and sent along coax.
With such small (and efficient) computers, reliability can be achieved by running several computers in parallel and voting on the answer.
Sensors
Most effects boil down to mechanics or electronics.
- Electric fields can be detected by the force they exert on an electric charge (electret).
- Chemicals can be detected by whether they fill a compatible binding pocket.
- Magnetic fields may be detected by their effect on an atomic magnet, though electronic detection may be easier.
- Photons, even in the visible region, may be rectified with a tuned antenna; in theory, this could allow high collection efficiency.
- At the nanoscale, sound is very slow mechanical vibration, to be detected with accelerometers or strain gauges.
- Temperature can be detected by differential expansion.
For more information on sensors, see Nanomedicine chapter 4.
Questions:
- How can complex-shaped products be made from cubical output?
- Unfolding
- Inflating
- Sparse cubes
- Working-face construction
- How complex should the cubes be?
- A few logic gates, or a CPU?
- A bearing, or a whole motor or robot?
- We need to balance:
- Manufacturing efficiency
- Product performance
- Ability to self-manufacture?
- How general a solution can we find?
- Cube size and complexity
- Manufacturing process and method
- How can larger functional units ("smart" "materials," large mechanisms) be designed?
- The design has to translate itself automatically into assembly instructions
- What range of products will we be able to make with this approach?
- Entire spacecraft?
These questions will be answered on another slide.
Concept Description
1) General Purpose Manufacturing
A general purpose technology is one that is useful in many contexts, like electricity or computers. A general purpose manufacturing system will be able to make a wide range of products for many different functions. A computer-controlled system performing many simple reliable operations in a programmable sequence would satisfy this requirement. Such an automated system would minimize the labor cost of manufacturing, and if the operations were small enough (as in the systems contemplated here) the system could be quite compact.
2) Exponential Manufacturing
Several manufacturing systems have been proposed that include most or all of their structure in the set of their products. Such an “autoproductive” system could be used to produce exponential numbers of manufacturing systems, or exponentially growing integrated systems. A fully automated autoproductive system implies the easy availability of as much manufacturing capacity as desired, with the system cost comparable to the cost of its inputs.
3) Molecular Manufacturing
Several different families of chemistry are candidates for producing reliable, programmable, nanoscale features by selective covalent bonding. Drexler has proposed protein engineering, optionally with modified peptides; a simliar proposal is currently being investigated in a Phase II NIAC project by Dinos Mavroidis [1]. Drexler has more recently proposed that scanning probes can be used to build solid carbon-backbone lattices (“diamondoid”) by deposition reactions in vacuum (“mechanosynthesis”). Nucleic acids fold into well-defined shapes. Graphene molecules may be built in solution, and fullerenes may also be organically synthesized.
Any of these technologies may be the basis for building nanometer-scale machine components. Any of several methods, including self-assembly and direct robotic manipulation, could combine these components into machines. Because covalent chemistry is inherently digital, molecular manufacturing appears to offer easy error checking and high reliability, suitable for automated and autoproductive systems.
4) Simple Assembly
Larger products will have to be assembled from nanoscale machines and functional blocks. This assembly must be automated. One way to do this is with complex motion-planned robotics. However, it appears that very simple assembly operations will be sufficient to produce complex products. There are several ways to do this, all involving the attachment of modular components with simple shapes. Convergent assembly, working-face assembly, and self-assembly are three possibilities.
5) High Performance Products
Diamond, graphite, and fullerenes are extremely high performance materials. A molecular manufacturing technology based on them would have strength perhaps two orders of magnitude better than today's metallurgy or polymers. Small cheap features should allow additional material savings through the use of e.g. fractal trusses and “smart” materials. Nanoscale features also provide access to favorable scaling laws for power density, computation density, and speed of operation (frequency). Atomic precision should allow extremely efficient and reliable operation. Calculated performance indicates power density of 10^15 W/m^3, power efficiency >99% for many operations including some chemomechanical conversions, computation density of 10^26 gates/m^3, and computation efficiency of 10^16 instructions/second/watt. Manufacturing throughput is similarly impressive, with a self-contained macro-scale manufacturing system estimated to output its own mass of product in a matter of hours or less. The orders-of-magnitude improvement in functional density would allow massive overdesign, allowing improved reliability and less exacting design restrictions.
6) Rapid Prototyping
A molecular manufacturing system as outlined above should be able to make products that are at least competitive with a wide range of today's products. The system would be programmable, able to take direction from a blueprint file without retooling or other delay. This implies on-demand manufacturing, in which there would be no difference—including cost—between a prototype and a production unit. Design iterations would be very fast and very cheap. And a single system would be able to make any product that could be designed for it.
Aerospace hardware construction costs could drop by many orders of magnitude. This implies that testing of concepts would be much less sensitive to failure (at least for unmanned flights), allowing more aggressive research and development.
Rapid prototyping of flight hardware would allow in-flight repair without a store of spare parts—just one manufacturing system and a small amount of feedstock. (Even the manufacturing system, being autoproductive, could be rebuilt in-flight.) This would allow less overdesign for long missions.
Statement of Work
The nanofactory depends on the design of it's basic component: The nanoblock. Hence most of our work will involve two tasks:
- Nanofactory Design: How do these nanoblocks cooperate to form useful aerospace devices?
- Specify nanoblocks: What makes up these nanoblocks? What can they do, and how do they interface with each other?
Methods of Assembly
How can complex-shaped products be made from cubical output?
Given very strong materials, a product may require very little structural mass. Think of how small an air mattress or a tent folds up; now improve that by two orders of magnitude. The folded/condensed form of inflated and unfolded products could be made with convergent assembly; a more or less cubical block could contain thousands of folded layers or expanding trusses.
Inflating
Picture a product as a thin shell, where the shell is composed of two layers of nanoblocks. The shell is stiffened by holding high-pressure fluid in small cells within the shell. High-pressure water could be used safely, and would also improve fire resistance for organic constructions. Manufacture the product in its "folded" state, and then inject water to unfold it. (The water would not fill the product, but just stiffen the hollow shell.) For comparison, a 200-lb inflatable plane was built in the 1950's.
Creases and curves might be a problem. How can rectilinear cubes form a curved surface? How can a sharply folded surface unfold without breaking the nanoblocks? (Even 200-nm thick diamond won't be that flexible.) Tom Craver's suggestion (on CRN's blog) of running strings between wedge-shaped blocks may be useful here. (Blocks can be built to deform after manufacture.)
Watertight or airtight seals between the blocks might also be an issue. Graphite flaps, or bumpers as suggested in the Vasculoid paper, will probably do the trick, but this should be verified.
Unfolding
Huzita's origami axioms and Hoberman spheres may be relevant here.
Sparse cubes
Just put together cubes with pieces missing. As long as everything is well supported (which is not trivial) then the assembly will work. Problems with this method include floppy components that don't match reliably during assembly, and large assembly volume required to hold the complete product form factor.
Working-face construction
Combines the best of sparse cubes and convergent assembly. As nanoblocks are made, extrude them out a working face, placing them where they need to go in the final product. Scaling laws say that a 200-nm cube could be placed in less than 10 microseconds. At that rate, placing a meter of nanoblocks would require only 50 seconds--less than a minute!
Advantages: Build full-sized products without unfolding or inflating. The working face holds the parts, so long floppy assemblies aren't a problem.
Disadvantages: Keeping the working face sealed from the outside environment might be a problem. It might be better to do the whole thing in a clean bag. Also, mechanisms for gripping the working face while adding tightly packed cubes to it need to be designed.
Self-assembly
The trial-and-error process of self-assembly is incompatible with irreversible (strong) bonds. This is a serious limitation, which can be only partially corrected by a subsequent bonding step. Also, large heterogeneous structures may take too long to self-assemble (the parts won't have time to find their right place). Given how simple the other methods will be once we have even basic nanoscale robotics, it looks like self-assembly need not be considered further.
Types of Nanoblock Modules: How complex should they be?
How complex should the blocks be? We agree they should all be equal-sized cubes (or maybe sections of cubes) with standardized manipulation and joining interfaces.
Tihamer's Simple Cubes
- Purely structural (precise, inert diamondoid blocks with holes for the pin drive, pure diamond being an insulator)
- [Tee--the pin drive is specialized for internal robotics, not manipulating cubes; just use a gripper.]
- Mostly structural, with a variety of conducting paths embedded.
- Encapsulated logic gates (NAND/NOR, INV).
- Pin driven nanoblocks (one side only - power and control come in from other sides.
- Twisting and/or bending GPE (giant piezoelectric effect) or IPMC (Ionic Polymer Metal Composites) bending nanoblocks (Lead zinc niobate-lead titanate (PZN-PT) single crystals are able to sustain very large piezoelectric strains; IPMCs are even better).
- Light/electricity transducers (LEDs and photodiodes) for communicating with the outside world.
Chris's Complex Cubes
200 nm cubes contain about a billion carbon atoms. That leaves room for a lot of features, but a block is small enough to be made in a few hours by a single mechanosynthetic manipulator/fabricator. Each cube can be covered with ridge joints for easy assembly.
These blocks can be parameterized. In other words, each list item identifies a family of blocks that may be tweaked internally ("subclassed," in software terms) according to the needs of the design or component.
- Purely structural
- CPU (8086-class) Nanofactory 6.1
- Memory (20 kB)
- Energy and signal transmission, switchable
- Programmable optics (hologram/diffraction or semaphore)
- Mechanosynthetic manipulator/fabricator
- Actuator, moving surface (rotating or translating; useful for e.g. traction drives of large shafts)
References
(Included from the NIAC References page)
Chris Phoenix, "Design of a Primitive Nanofactory", Journal of Evolution and Technology,Vol. 13,October 2003, http://jetpress.org/volume13/Nanofactory.htm
Chris Phoenix and Eric Drexler, "Safe Exponential Manufacturing", Nanotechnology 15 (August 2004) 869-872. Nanotechnology © Copyright 2004 IOP Publishing Ltd., http://www.crnano.org/IOP%20-%20Safe%20Exp%20Mfg.pdf
Robert A. Freitas Jr., Christopher J. Phoenix, Vasculoid: A Personal Nanomedical Appliance to Replace Human Blood, Journal of Evolution and Technology, Vol. 11 - April 2002, http://jetpress.org/volume11/vasculoid.html
Tihamer Toth-Fejel, "Modeling Kinematic Cellular Automata: An Approach to Self–Replication.” Final Report, http://www.niac.usra.edu/files/studies/final_report/pdf/883Toth-Fejel.pdf
Tihamer Toth-Fejel, LEGO(TM)s to the Stars: Active MesoStructures, Kinetic Cellular Automata, and Parallel Nanomachines for Space Applications, International Space Development Conference, New York City, May 1996, The Assembler, Volume 4, Number 3 Third Quarter, 1996, http://www.islandone.org/MMSG/9609lego.htm
