In the course of developing a model of a simple vacuum-deposition process, it is interesting to look at some of the sensitivities of the deposition process. Anything that has an effect on the process will ideally be controlled and wherever possible minimized.
In metallizing, I realized that for the sake of simplicity I had thought of the heat load onto the substrate being sort of linearly applied. This is because of thinking of the coating going from zero to full thickness between the start and end of the deposition zone. This is entirely wrong as the vapour cloud from the evaporation boats has a flux that is some power of a cosine law.
This means that as the substrate first sees the deposition flux the deposition rate is lower than it will be at the midpoint, directly above the center of the deposition source and similarly the rate falls again as the substrate approaches the end of the deposition zone.
When I looked at the rate of temperature rise, it became apparent that this was not the most significant contribution to the substrate temperature rise. The fastest substrate temperature rise occurs as the substrate enters the deposition zone. The radiant heat load hits the substrate and there is an immediate rapid temperature rise. A short time after the rate of temperature rise changes slope to a lower rate of temperature rise. The reason for this change in slope is the aluminum coating, which is reflective, and so as the coating thickness builds up more of the infrared is reflected back towards the sources and less is absorbed by the substrate. The continued rise in temperature is as a result of the heat of condensation of the depositing coating.
If this initial rise in temperature due to radiant heat load can be minimized, the substrate peak temperature will be lowered or the allowable coating thickness could be increased.
It is typical that during the length of the deposition run the system temperature rises. This would suggest that the heat load to the substrate is potentially higher towards the end of the run than initially. This would also suggest that the rolls will be more prone to wrinkling towards the end of the run than at the beginning of the run. It also suggests that there is a benefit from cooling deposition shields, particularly those surrounding the entry side of the deposition zone.
What might be interesting to evaluate would be to offset the deposition boats, moving the center of the boat slightly towards the entry side of the deposition zone. This would increase the deposition rate in the first half of the deposition zone which ought to result in a reduced heat load for the same coating thickness. The negative side of this would be that there would be reduced material collection efficiency.