It is essential to know the starting point so that when any changes are made to the materials or process any improvements can be quantified. Simplistically the barrier performance is dependent on defects in the coating and getting a measurement of the defects is a worthwhile exercise. If we consider what defects are likely to be present in the coated film the largest number and the largest in size are usually pinholes. Pinholes are produced by particles on the film surface that are coated and then later moved away leaving behind a hole in the coating.
This means there are usually two opportunities for measuring the defects, one is to count and size the particles present on the surface of the film. The second is the easier measurement to make as the metallised coating means that the contrast between the metallic coating and the pinhole can be enhanced by using a light behind the coating film. Here too the number and size of pinholes should be counted.
The target is then to make changes that reduce the size and number of pinholes.
The barrier performance measured may not immediately show any great improvement. If the area of the pinholes in any given area is more than ~5% then the coating will produce little or no barrier improvement to the film. This means that if the number of defects is reduced and the area is reduced but the total present is still close to 5% then it will be very hard to see any improvement. However it should be relatively easy to make further reductions in number and size of pinholes.
As the number and size of the pinholes is reduced to below 1% it will become harder and harder to continue making improvements. Initially it is easy to assume that all pinholes are due to particulates on the film surface but in reality they are the largest contributor to pinholes but not the only source of pinholes. Oligomer on the film surface and poor adhesion of the coating because it is on this oligomer means that it is easy to pick off the coating by any filler protruding on the back surface of the film that is pressed hard against the metal coating. Then there are the pinholes due to oligomer that is of a low enough melting point that it evaporates as the film passes through the coating zone. There will also be pinholes from the evaporation source where particles are spit from the boats and result in pinholes. To continue to make improvements all these also need to be addressed.
If the number and size of the pinholes is large then it is often not worth trying to make improvements on the minor sources of pinholes but it is far more productive to initially concentrate on the largest source of defects which is from the particulates on the film substrate.
I was recently asked what alternative materials were there to aluminium that could be used to make better barrier coatings. The answer for most metallizers is none. Changing the coating material but doing nothing to remove the cause of the coating defects will do nothing to change the barrier performance. Of course once you have eliminated all the particulate contamination on the substrate, all the oligomer present on the film surface, and filler that might be protruding from the film surface as well as using higher purity aluminium and improving the processing stability to eliminate spitting from the boats, then the barrier performance of the coating can be influence by the metal being deposited. If any of the above sources of defects has not been addressed and minimised then there is little point on changing to a different, and usually more expensive, material as the improvement possible will be lost in the much larger variation produced by the other defects.
Hence my advice is to measure the film surface quality with particular reference to particulate and oligomer contamination. Either by direct detailed microscopy of the incoming substrate or by the easier route of taking metallised film and using back illumination counting and sizing the pinholes in the coating. Once you have a measure of the current coating quality then to make improvements, one at a time, and measure any changes. Do not rely on barrier measurements alone as until the coating is of high enough quality the improvements may be hard to detect against the variation in the measurement technique.