Plasma physics on the factory floor: things the textbooks never worry about

Summary form only given, as follows. In theory, it should all be so easy. Decide on the plasma chemistry you will need, maybe do some simple experiments to decide on the plasma characteristics that will get you there, and away you go. The reality is much more difficult. Here's why:Cost - The cost of producing a plasma and its output, always a concern in the nonsemiconductor world, is now of primary concern even in semiconductor manufacturing. Technical elegance is no longer enough - cost concerns are paramount. The cost of the power, the cost of the matching, and the cost of the plasma device itself. Reliability - In all industries, the requirements for factory uptime are continually increasing. In some applications the plasma may be in operation nearly 100% of the time. Control - The subtle things can matter. Unstable tuning regions, hysteresis that comes and goes, process results that are installation-dependent are all pitfalls that need to be avoided. Size - Smaller is better, and often less expensive in the end also. This talk will review some of the underlying physics and chemistry issues driving the design of plasma processing equipment, with a particular focus on the real-world factoryfloor issues.