The diffusion of O2 in or other material out of GaN is very hush hush at LD manufacturing level. I will check again with my sources, and see what I can find out. I have heard of the degrading, but maybe the coatings are sufficient.
As far as the green sealant, it's even used on our $2 red diodes. it's green to indicate it's RoHS Compliant. I'm sure there is a very limited number of manufactures of the sealer, same with other diode components. Sold to the big LD makers by other suppliers of the individual components. One of our red diode suppliers was doing about 6 million of their labeled LD's per month back in 2007/08. Production wnet way down from there, but even then, they made about another 5 million parts for 2 well know LD maker's of Japan. They concentrate on wafer production and diode assembly, a cool automated process done in small, plastic covered robotic cells about 5X6 feet. I only got to enter 2 LD assembly clean rooms: one doing 3MM/ month, and the other 1MM/ mo. Even at these volumes, the LD components are supplied from outside.
my 0.02
