I propose you the following conjecture: can a treatment of surface modify pressure?
The electronic industry uses a great amount of single-crystal silicon. This matter is also employed for the creation of miniaturized mechanical devices. The tools of this new industry are the same used in electronic industry: etching, doping, epitaxial deposition, thermomigration and more. For making micro-mechanical structures, etching is the most versatile tool. Chemical etchants for single-crystal silicon are numerous. Some of these etchants are anisotropic: the speed of etching is dependent of crystallographic orientation of exposed silicon. For example, KOH in water generates on a silicon area with a crystallographic orientation named <100> a pyramidal pit. The crystallographic orientation of the planes of the pit is in this case <111>. If E is the side of square entry of the pit, its depth P is calculated by the formula: P = E / 1.414. With the other tools of electronic industry, photolithography and masking, this sort of pyramidal pit could be generated in great number and in small dimensions on single-crystal silicon.
The possibility to make these pits in great number and in very small dimensions incited me to test the hypothesis: what could be the comportment of a gas in a pit when its dimensions are smaller than the mean free path of gas molecules?
For this purpose, I have made a software with Visual Basic (Windows version). This software named Pyramid simulates the comportment of gas molecules in a pyramidal pit. This simulation is very simple: the trajectory of each gas molecule is determined one after one. The simulation is realistic only in the case when the dimensions of the pit are smaller than the mean free path of gas molecules (for example, the mean free path of atmosphere at sea level is in order of 70 nanometres). With this condition, a gas molecule that enters in the cavity could have a shock or more on the planes without contact another gas molecule. The shock of gas molecules on solid planes are supposed elastic. After each contact of molecule on solid, the software calculates the new trajectory of molecule and the exchanged impulses until the gas molecule leave the pit.
The errors of the software are avoided by two processes. The geometrical process verifies the absence of abnormal trajectories. The physical process survey the stability of the speed modulus of gas molecule between entry and exit of the pit (the shocks are elastic so the speed modulus of molecules must be constant).
The result of the simulation is surprising: the shocks of gas molecules in the pit generate an overpressure. This overpressure represents 10% of the normal pressure when the pit is absent. Simultaneously, the flight movement of gas molecules when they leave the pit is enhanced.
Has this phenomenon any chance to be real?
For your judgement, the software Pyramid can be found here:
http://simulation.servehttp.comIt contains a detailed presentation of simulation (in French).
Dau.mic