If we remove some molecules from an enclosed container initially at 760 Torr, what happens to number density, mfp, and particle flux? The easiest quantity to understand is number density. If we remove half of the molecules from the container, the number density goes from 2.7 × 1019 cm-3 to 1.35 × 1019 cm-3. If we remove 99 percent of the original molecules, the number density is 2.7 × 1017 cm-3, still a huge number. The table shows the relationship between pressure, number density, mean free path, flux, and the time taken to completely cover a clean surface with a monolayer of air at room temperature.
With respect to the monolayer coverage, it depends on particle flux, molecular diameter, and the sticking coefficient of the gas molecules on the surface. The numbers given are for air which has an average molecular diameter of 3.7 Å and the sticking coefficient is ~1 on a clean, unheated surface.
Fractions of Atm
|Time for One
|1/1,000||0.76||2.7 x 1016||0.0065||2.9 x 1020||3 x 10-6|
|1/10,000||7.6 x 10-2||2.7 x 1015||0.065||2.9 x 1019||3 x 10-5|
|1/100,000||7.6 x 10-3||2.7 x 1014||0.65||2.9 x 1018||3 x 10-4|
|1/1,000,000||7.6 x 10-4||2.7 x 1013||6.5||2.9 x 1017||3 x 10-3|
|1/10,000,000||7.6 x 10-5||2.7 x 1012||65||2.9 x 1016||3 x 10-2|
|1/100,000,000||7.6 x 10-6||2.7 x 1011||650||2.9 x 1015||3 x 10-1|
When a chamber has no leaks, has no gas deliberately flowing into it, and has been pumped for several days, the pressure reaches an equilibrium value called the base pressure. In truth, because the pressure approaches equilibrium asymptotically and the outgassing rate undergoes exponential decay, even after a long time under vacuum, the chamber—theoretically—will never quite reach a stable pressure. But variations in vacuum gauge calibration, room temperature, pumping speed, backstreaming from the pump, etc., mask or counter any real pressure reduction and the chamber appears to have reached a steady state.
Often what happens is: the operator pumps the chamber for a few hours, grows tired of waiting, and claims the chamber is at base pressure.
This is not necessarily wrong. After all, if the pressure falls from 5 × 10-7 Torr to 4 × 10-7 Torr by waiting another ten hours, is all that much gained? Perhaps it doesn't conform to formal definition, but in a sense the base pressure is reached whenever the operator says it is and starts using the chamber.
The term “base pressure” defines conditions where no gas is deliberately flowing into the system. But sometimes the chamber is first pumped to its base pressure (to check for leaks or remove contamination) and then back-filled with a gas to an intermediate pressure. This is how processes such as sputter deposition, plasma etching, and CVD are done. This intermediate pressure is called the working pressure. To establish and maintain a working pressure, it is rarely sufficient to just close the pumping port, back-fill with gas, and walk away. Most back-fill applications require a flow of fresh gas to sweep away contaminants desorbing from the chamber walls. Often the back-fill pressure is stabilized with a feedback control system.
Ultimate Vacuum or Pressure
Vacuum-pump manufacturers gives two specifications: pumping speed and ultimate pressure (also called ultimate vacuum). The ultimate pressure is measured by capping the pump's inlet and finding the equilibrium pressure after operating the pump for many hours. Because it is measured under “ideal” circumstances, it is crucial to remember that a chamber connected to this pump will never reach the quoted ultimate pressure!
Perhaps worse, pump manufacturers measure the ultimate pressure of mechanical pumps using a McLeod gauge that cannot measure vapors such as pump oil and water. Consequently, the so-called ultimate (partial) pressure of a rotary vane pump may be quoted in the 10-5 Torr range, causing much confusion when the practical ultimate pressure (using a gauge that responds to oil and water vapor) is two decades higher.