As a material scientist myself, I was curious about their study. So I pulled up their paper in JAP.

My first thought before reading was that the high intensity laser pulses were causing local heating such that any impurities in the solid (ie carbon) diffused onto the surface. Nope, they did an EDS scan and platinum was clearly the dominant element. There were traces of carbon but I think this was due to impurities in the microscope.

So basically what happens (my educated guess) is that the laser pulses break up the surface structure such that different phonon (http://en.wikipedia.org/wiki/Phonon) modes are excited when certain wavelengths of light are shined on the surface. Just so happens that wavelengths in the visible range are absorbed so the sample appears black. To create different colors, I guess they just found the correct amount of power and wavelength laser pulse to create the phonon mode that they need.

I don't believe they oversold their technique. Scratching won't really cause this phenomenon. Corrosion maybe, but I doubt they could achieve the same optical properties. Fragility is a legitimate concern, but I'm guessing that if they are interested in industrial applications they will do mechanical testing of the structures.

EDIT: I'd also like to add that I think it's perfectly OK for a scientist to upsell their work. We live and die by the grant system, and funding agencies are all looking for "broader impacts." I'm rather impressed that their group was able to drum up this much press.

Perhaps a plastic coating to preserve the microstructures... Still valuable, the colors apparently can be dialed up to order, unlike paint etc which depends upon mixing and batch etc.

A polymer coating would indeed help. In the jargon, we call that stuff "paint". :)

I think the GP meant a varnish, not a paint.

There's absolutely no difference; those are both completely informal terms. A "varnish" (or whatever, there are thousands of names for these thing) is nothing more than a transparent paint base without pigment added. The point was that the claim was made in the article that this is more durable than paint, when AFAICT the only way to make it so it to put a paint(-like coating) on top of the surface anyway. So how is it more durable? Scratch the coating and you scratch off the fancy micro-texture.

It could for instance be more durable because it doesn't fade due to ultra-violet light breaking down the pigment.

The same is true of ionic pigments, though. Only the organics fade in sunlight. Again, hardly an earth-shattering innovation. Might make it useful in toys where toxic pigments can't be used...

How about a micro-coating. Not applied like paint, deposited on the surface like an integrated circuit layer, microns thick. SUre you can still call it paint, you can call the layers of an integrated circuit paint but its deposited differently, is a differnt industrial process, mechanically different, different thermal characteristics, lots of things. More significant in my original comment was this: the color can be "dialed", kind of like "digital color". Paint is imperfectly mixed material with wide color/reflectivity variation over a surface. This new method of coloring a surface is uniform. Imagine the efficiency of a color filter, absorbtive surface, radiative surface etc if the optical properties could be tightly controlled over a wide area.