The effects of scaling on performance
These are first order "back of the envelope" calculations about the effects of making things small. For reasons which will become apparent as you read this I doubt that true nano scale weapons will ever exist. What could possibly be built are micro scale robotic devices of a non self replicating type which could possibly be used as weapons. Let us find out how practical they might be. Let us start by examining the effects of scaling on things. We'll start with my Nissan Maxima and reduce it in size by a factor of ten. Instead of being about 17 feet long the scaled car will be about 1.7 feet long. Instead of weighing about 3000 lbs it will weigh about 3 lbs. Why is that? The answer is that the mass of a scaled object is proportional to its volume - which goes as the cube of the dimensional ratio. Ten times as long, ten times as wide, ten times as high has 1000 times the volume. The scaled engine would be 3 cc in displacement instead of 3000 cc. Instead of 222 Hp it would produce .222 Hp. Fuel consumption at this level would be one thousandth of that of the full size engine. Since the fuel tank is also one thousandth of the size of the full size vehicle one might be tempted to think that the distance between fill ups would be the same. However, the fuel consumption of the smaller vehicle is proportionally greater. Why? The smaller vehicle is one thousandth the weight but the frontal area of the vehicle - the size of which determines the drag - is one hundredth of that of the larger vehicle. Thus at the same speed the drag of the smaller vehicle is proportionally ten times as great as the larger vehicle. The optimal speed of the smaller vehicle is lower than that of the larger vehicle. Because drag goes as the square of the velocity, one thousandth of the fuel consumption will drive the smaller vehicle at a speed which is about 32% of the speed of the larger car and its range will also be about 32% of the full sized car's range. If we tried to make a car scaled down by a factor of 100 its speed and range would both be only one tenth (square root of a scale factor of 100) that of a full size car. We are forced to conclude that the product of speed and range of any vehicle with an internal fuel supply will scale directly with the scale factor. For example reducing the size of a jet plane by a factor of 100 makes it fly at one tenth the speed and one tenth as far. By the time we scale to nano sizes we have objects which won't go very far or very fast. A nano device is an exceptionally poor weapon delivery system compared to a full sized device; it can only move slowly, and it can't go very far. However there are other things which occur which would effect our attempt to simply scale an engine down in size. The first of these is the change in heat loss. In simplest terms the rate of heat production is proportional to the volume of a heat source, which means that heat production scales with the cube of the scale factor, but heat loss is proportional to the surface area of the object which scales as the square of the scaling factor. A smaller engine requires much less of a cooling system than a large engine does, if the engine is small enough it doesn't require a cooling system at all - it will lose heat naturally fast enough without one. Because of the square - cube relationship for heat loss there is a minimum size flame which is possible. A small ball of flame loses heat faster than a large one. If a ball of flame is too small it can't produce enough heat from internal combustion to maintain its temperature above the ignition point, and the flame can't exist. This means that if we try to scale our engine far enough it will refuse to run, it will lose heat too fast for the fuel to burn. Even making the engine out of heat resistive materials like ceramics only works to a certain size; eventually the heat loss will keep things from burning. This is part of the reason that biological cells use chemical reactions instead of combustion processes to produce energy. For these reasons we are forced to realize that if we are to produce any sort of self propelled small device we have to use similar chemical reactions for an energy source since combustion won't work. If we use chemical reactions the temperature of these reactions has to be lower than a flame - otherwise we could have combustion instead. The efficiency of any heat engine is directly proportional to the difference in temperature of the heat source and the temperature of the waste discharge; when the heat source is lower in temperature the efficiency of the heat engine declines. (Any biochemical system is a heat engine, the laws of thermodynamics apply to them just as much as they apply to a simple heat engine like an internal combustion engine.) Thus any nano chemical engine will be less efficient than an internal combustion engine because of the lower hot side temperature of the reaction. As an example: if we create a device which is 5 microns long instead of a car of 5 meters in length, the scale factor is one million. This implies a range one thousandth of the full size car, and a speed one thousandth as great. The 5 micron device will crawl about .4 miles in about 6.66 hours and then run out of fuel, if it is as efficient as a car - which it won't be. Thus our micro weapon could out run a snail, but anything else could simply walk away from it. Clearly a chemical power source for a nano device is insufficient. How about nuclear power? The ratio in energy between nuclear and chemical is about one million to one so the potential to achieve long range and high speed could possibly be raised to useful weapon delivery levels with nuclear power. If we investigate nuclear power for small use a few problems begin to occur. A controlled nuclear reaction such as fission or fusion won't work at nano scales. Both fission and fusion require physically large objects. In fission the mean free path of a neutron before a fission occurs is on the order of fractions of an inch. The only way to decrease that distance is to increase the density of the fissionable material. However the highest pressures that humans can create causes a fissionable material to compress very little; imagine trying to squeeze a solid steel ball to one tenth its radius to see the problems. To get fission at nanoscale would require densities approaching that of a white dwarf star - which is wildly beyond the capability of anyone to create. Fusion is even more ridiculously out of reach - we are unable to achieve controlled fusion at a net energy gain even on a large scale, let alone at the nano level. This leaves us with uncontrolled radioactive decay as the only nuclear energy source available. If we are to restore the speed and range of our full size car in a nano device we would need a radioactive isotope with a half life on the order of 7 hours. We would also have to fuel the weapons just before using them. Since the radio isotopes would start decaying as soon as they were created, any delay in use after fueling just wastes energy. Both Gamma emitters and Beta emitters are impractical for small use; the rate of energy deposition in the device would be minuscule with a device size of 5 microns. The radioactive emission which stands the best chance of being used is an alpha particle. Alpha particles with an energy of 5 million electron volts have a range in air of several cm. According to Lapp and Andrews text book "Nuclear Radiation Physics" page 119 equation 6-7 the same alpha particle would be stopped by about 20 microns of aluminum. If we use gold instead of aluminum we could absorb the alpha in about 7 microns which is in the ball park of the size of our device. There is no hope of making such a device a self replicator, since once the radioactive fuel is expended the weapon dies. In order to create an active micro (it is too large to call it a nano) weapon with such a nuclear fuel we would need to expose our micro weapons to a flux of activating neutrons from a reactor or particle accelerator. We would wind up with a device with about the performance of a standard automobile (That is, if we were able to get all of the energy of the atomic decay into useful propulsive energy, which simply isn't going to happen.) In order to use these weapons they have to get within about 400 miles of their target before they are activated, assuming their target is in the open air like a person, and also assuming that there is a road available which appears smooth to the micro device (an example of such a road would be optically flat glass plate polished to within 1/10 of a wavelength of light. If such a flat road is not available and the micro weapon has to deal with nature then tiny, barely visible, clods of dirt will look to it like a mountain with no roads. One burns rather a lot of energy and time in getting over just one of these). If the target is hardened the micro weapons have to burrow their way through the concrete and steel at the target. Their range and speed in hardened material would be greatly diminished compared to that in air. I would like to point out that we can forget about flying micro weapons: even nuclear power is barely theoretically able to produce enough energy for an automobile style performance. Underwater micro weapons are even more grossly limited in performance; water drag is far greater than air drag. Even nuclear powered micro weapons have extremely low range and low speed. Unlike a nano device - a micro weapon can't gain easy access to a submarine by moving a few atoms aside, it has to bore its way in. Sorry, that is not going to happen. Even the tiny amount of nuclear power a micro weapon could carry is insufficient to do that. On land, stopping an attack of micro weapons is fairly simple: because the weapons are so small they are very vulnerable to heat. The square - cube heat relationship means that an external heat source raises the temperature of the micro device very quickly. For example attacking micro weapons could be destroyed with something as simple as Thermite or Napalm. Thermal expansion from the external heat would distort, bind up and thus ruin any moving parts in the nano device. Bottom line: in the real world, and not the world of science fiction even ridiculously exotic nuclear powered micro weapons just aren't very good weapons. If they were sent to attack conventional nuclear weapons, the nukes could be used long before the attack could be completed. There is one possibility of an energy supply for a micro weapon: solar power. Solar power for micro devices has one gigantic advantage over just about any other potential energy source for these applications; it scales properly. Solar cells are surface area dependent instead of volumetric dependent. Because of this they are not subject to the square - cube law tyranny, and maintain the absolute performance of the device as it is decreased in size. There are limitations to solar power for powering micro weapons: 1. It works only in bright light. 2. There are size related electrical effects. Let us examine the first point. Any attempt to store energy from the solar cell internally in the micro device is subject to the cube - square law problem of any internal fuel supply. The range of the stored energy in darkness is very small and the speed available is poor. Now let us talk about the electrical effects of scaling. The very best insulating materials which exist have a maximum insulating capability of about 10 million volts per meter, above this electrical field level the insulator breaks down and shorts out - becoming useless. At the one micron level the maximum voltage possible is 10 volts. In order to withstand the voltage produced by a solar cell (.6 volts) the minimum theoretical thickness of the insulation of the wires to the cell would have to be 60 manometers. Given practical engineering constraints the actual insulation thickness would be on the order of .15 micron. Note that this insulation thickness has to be maintained inside the electrical motor also, to keep the motor from arcing over. Since the insulation is all around the wire the diameter of the wire simply from insulation would be .3 microns. If your car had to have insulation to that scale (5 micron device size) the wires would be .3 meters in diameter, including the wires inside of your starter motor! Clearly building an electric motor which could fit inside of a car with such wires is a fool's errand, as is building a motor for a scaled down 5 micron weapon. The bottom line is that 5 microns is too small for a solar powered device. A more possible size might be on the order of 150 microns - which would give the equivalent wire sizes in a full size car of about one cm. The motor with such wires would be big and bulky, but it could be built. The energy source, speed, and range problems I have outlined at the micro level become much worse at the nano level, where neither nuclear power or solar power can work. The only conclusion I can reach is that Nano weapons simply won't be practical because of those problems. There are other power problems which occur at the nano scale level but I see little point in beating a dead horse.