10nm gives the best results I've been able to get. Silica particles are much more convenient to buy, though we did end up making a huge batch of titania since the shipping from Japan was being slow. And yeah, the coatings are applied LbL with what's basically an expensive fancy spinning version of a slide stainer. I can't back it up with tons of evidence yet, but the spinning action seems to make it much easier to get a nice even coating.
Speaking as an MIT student working on this project, yes it lets more light through- a lot. Uncoated silica reflects about 8% of incident light, as was posted elsewhere. With our coating, this drops below one percent through most of the visible spectrum, and below.2% at a peak wavelength dependent on the number of coating layers (around 550nm for a 14-bilayer coating). It's a pretty nice improvement- you can place a half-coated slide against white paper and the untreated side looks dirty by comparison. I can try to dig up the spectrophotometer measurements I took a few weeks ago, if anyone cares that much.
Slashdot Mirror
10nm gives the best results I've been able to get. Silica particles are much more convenient to buy, though we did end up making a huge batch of titania since the shipping from Japan was being slow. And yeah, the coatings are applied LbL with what's basically an expensive fancy spinning version of a slide stainer. I can't back it up with tons of evidence yet, but the spinning action seems to make it much easier to get a nice even coating.
Speaking as an MIT student working on this project, yes it lets more light through- a lot. Uncoated silica reflects about 8% of incident light, as was posted elsewhere. With our coating, this drops below one percent through most of the visible spectrum, and below .2% at a peak wavelength dependent on the number of coating layers (around 550nm for a 14-bilayer coating). It's a pretty nice improvement- you can place a half-coated slide against white paper and the untreated side looks dirty by comparison. I can try to dig up the spectrophotometer measurements I took a few weeks ago, if anyone cares that much.
/.ed. Party in lab today!
Also: Whoa, Rubner got