> Of course it doesn't actually work, but figuring out why makes an interesting exercise.

I've always found that kind of interesting. The US lets you patent just about anything.

But in my home country, Australia, one of the patent requirements is that it actually appears to function in the way you are claiming it does. ("The application must be for something patentable, like a practical adaption, not for an idea or principle.")

I wonder if that has any impact on patent trolling, or if it's just as pervasive.

Theoretically it's a requirement in the U.S. as well. To get a patent, the invention has to be "useful", and presumably to be useful it has to work. But in practice this is never enforced.

Is it because quantum entangled particles propagate their changes at the speed of light? Making the invention work, just not any faster than normal methods?

No, it's a little more subtle than that. If you really want to know, the answer is in this paper:

http://www.flownet.com/ron/QM.pdf

Or if you prefer, this video:

https://www.youtube.com/watch?v=dEaecUuEqfc

That looks like a patent application, not a grant. You can't claim you have a patent on this yet.

I'm not sure why the application page rather than the grant page is the one turning up in Google's results, but the patent was granted. It is US Patent number 7,126.691, issued October 24, 2006.

Ah. Here's the link: https://www.google.com/patents/US7126691