"Perpetual Force" Air Molecule Motor Part I

Can the movement of air molecules be exploited?

Fig 1. Specification drawing for foil with tapered holes
Alt 2 has foil thickness 0.05 micron, and hole dia 0.01 micron at upper surface; 0.002 micron at lower surface.
Alt 3 has foil thickness 0.5mm, and hole dia 0.1mm at upper surface; 0.02mm at lower surface.

I'll now say something about the foil shown and specified as Alt 2 in the above specification drawing, which I first posted on 17 September. I wanted to see whether a foil like this would experience a net force just from being immersed in air at normal temperature and pressure.

Operating principle

The proposed operating principle was similar to the Casimir-force foil, except that instead of zero-point energy fluctuations, it would be air molecule impacts delivering energy/momentum to the foil — but again to a higher degree on the lower surface. Air molecules arriving at the upper surface would generally enter the holes, and would be reflected from the sides of the holes, exerting only small forces resolved vertically, before (probably) exiting at the lower surface. Air molecules arriving at the lower surface would be more likely to hit that surface, because of its decreased hole diameter, exerting large forces resolved vertically.

No supplier

As already stated, no-one was able to supply any of the foils described in the specification drawing; not even Alt 3, which was a much larger version of Alt 2, and hence would have been much easier to make. Alt 3 was also intended for air molecule impacts, but in a fairly high vacuum, where the mean free path length of the molecules would be a lot higher.

Silux model

Since I was unable to test this idea in any physical experiment, I decided, later on, to see what could be done with a silux model. (For anyone unfamiliar with silux, see my posts of 7 April and 13 April 2014 on this blog).

The file folder named "models" that comes with the silux program includes a model named "Model and Simulation of an Almost Ideal Gas", which is also described further in the free PDF document named "Samples 2D" (page 49). This model demonstrates how the impacts of individual gas molecules, when aggregated together, produce an upward force that balances a weighted piston within a cylinder. (It also demonstrates visually how molecule velocities vary naturally over time — presumably a correct Maxwellian velocity distribution is eventually achieved, although it would take a lot of work to verify that).

I adapted this model to check the effect of air molecule impacts on a small, non-optimised portion of a foil, as shown in Fig 2 below. 

Fig 2. Silux model of a small portion of a foil with tapered holes,
in a container with 200 air molecules

Model details

I started with the 100 molecules in the original model, duplicating them again for a total of 200. Each molecule has a radius of 2mm, a mass of 0.261 gram, and is originally started at a velocity of 1.414m/s, as originally created by silux. The seven triangles are made into a single foil of mass 1kg, constrained by a macro against sideways or rotational movement. Movement up or down is permitted. All molecular impacts, to other molecules, or the foil, or the fixed container, are always 100% elastic. Gravity is inactive. The container "volume" in this 2D model is 20cm × 10cm = 200cm².

The simulation is started with the model configured as above. This is after a few seconds of prior running with the foil also held fixed, to ensure that the air molecule velocities are "randomised" before the foil is released.

Fig 3. Silux model at end of simulation

Results:

After 3.981 seconds of simulation time the foil is about to hit the upper boundary of the container, as shown. By then it has moved 0.03883m upwards, and has an upwards velocity of 0.02612m/s.

This looked reasonably promising, so I decided to look at some more cases, which I'll discuss next time.