http://www.physorg.com/news90681726.html

Very nice information!

1. At that pressure (kilopascals, too, please?), the membrane (unless it was something like 0.1 mm in diameter or so) must have been supported by a structure with fine-pitched holes. That doesn"t detract from the development"s value, though; just significant info. not given in the article.

2. To quote, "At more than 4,000 times thinner than a human hair, the new barely-there membrane is thousands of times thinner than similar filters in use today."

That "times thinner than" phrase is simply ridiculous nonsense, although pathetically common. If the size of a typical human hair is maybe 0.1 mm, then 4000 times that is 400 mm. What you"re saying is that the membrane is roughly 400 mm smaller than a human hair. The details of quantum behavior defy "common sense", but, so does this! After all, PhysOrg is supposed to be about straight thinking, isn"t it?

The article in nature says the 15psi was measured on a 200x200 micron membrane (0.2x0.2mm) without any additional support. It looks like these could be made into large arrays to increase the effective surface area since they're built on a silicon wafer platform.

I'm not sure I understand your issue about more than 4000x thinner than a human hair. It's 15nm thick as stated in the article and this news piece. I'm under the impression that most people think that means smaller than 1/4000th the thickness of a hair - using your statement that a human hair is about 100 microns, the math works out.

Impressive work, and the company they mention at the bottom has a webpage, it's just www.simpore.com

The problem with filters is that they need regular cleaning which is not easy on that scale.