Hank Green: You'll get no arguments from me, the Hubble Space Telescope is fantastic, but it's not fantastic at everything. In fact, it and nearly every other telescope is crap at detecting most of the infrared wavelengths of light. The Hubble can detect the near-infrared, the shortest wavelengths of IR that are closest to visible light, but the real juicy goodness is in the mid-infrared wavelengths that are impossible to capture from most places on Earth and still hard as heck to detect in space. Now if you tried to scale up the Hubble's mirror to work in that range, it would be (1) larger in diameter than the biggest rockets that currently exist, in order collect enough light, (2) maddeningly difficult if not impossible to accurately polish, (3) deformed beyond usability at the extremely low temperatures necessary to see IR light, (4) too heavy to launch, and (5) worst of all, still unable to reflect light at the infrared range. So yeah, there's a reason people said that building a high-resolution IR telescope was impossible -- actually, five reasons that we just discussed. But it is possible, because we've built this mirror. [intro music] We've taken each of those challenges on and, remarkably, we're finished. Now it's just a matter of waiting for the rest of the James Webb Space Telescope to be completed and sent into deep space to start taking pretty pictures. But let's take a look at those five big problems, and, with the help of Northrop Grumman's Scott Villager, determine how we overcame each of them. First, the vehicle used to launch the Webb into space, the Ariane 5 rocket, has a diameter of five and a half meters [(18 feet)], and the Webb's telescope meter has a diameter of about six and a half meters [(21 feet)], and that's not even mentioning that it would be impossible to polish a mirror that big, so yeah, problematic. Solution: build your mirror in 18 foldable pieces.

Scott Willoughby (vice president, JWST program manager): The reason we built those in 18 segments instead of one, I was mentioned, there's two reasons. One is we couldn't polish a mirror that big, and the other thing is we had it segmented so we could put three and three on a hinged one.

As the telescope travels to its destination, these mirrors will fold out. In fact, the whole thing will transform out of its little cocoon into the beautiful butterfly that it was made to be. Hot stuff! Problem number one and number two solved. Now, number three. How do you make a mirror on Earth that will work in deep space? First, you use beryllium.

Scott: Beryllium was chosen because at these cold temperatures it's... it holds its shape, you know, basically precisely. You can't afford to have this be moving around.

But it's more than that. They have to anticipate how the mirrors will deform at 40° above absolute zero. They actually only finished the prescription of the mirror, just like the prescription in my glasses, after bringing the mirror down to -387° Fahrenheit [(-233° Celsius)] to see what its shape will be.

Scott: It would be blurry at room temperature, so when it goes down to 40° above absolute zero, it crisps out.

Still, beryllium, though fantastic in cold temperatures, is way too heavy to launch, so now we must solve problem number four -- how to be lighter per square meter than Hubble's mirror. Well, hollow the sucker out.

Scott: I think the block is about 400 pounds [(181.4 kilograms)], and it ends up being about 40 pounds. It's just a very thin face sheet, and if you looked at the back of it it's basically a grid of ribs. One piece of metal, uh, that gets basically hogged out by a machine.

And finally, how do you get it to reflect infrared when beryllium is crap at even being a regular mirror? Well, you find the one metal that reflects infrared so well that it's the only metal that actually appears reddish to the human eye: gold.

Scott: Gold is better at reflecting infrared wavelengths. It goes on in ångströms. It's done in New Jersey, which is really important. That's the most important contribution to this telescope is the, uh, the bedazzling, which you would think would come from New Jersey.

All of that gold spread out across 25 square meters! [(269 square feet.)] You'd think it would be a ton of gold, but it literally goes on atoms at a time. The amount of gold in these mirrors is roughly equivalent to the amount of gold in ten wedding rings -- no more than 50 grams. Dang. And now we have done it. We have a lightweight mirror that reflects infrared light very well, is perfectly polished to operate at very low temperatures, and can fold up into the payload of an Ariane 5 rocket. If you got a problem, these are the folks to fix it for you, as long as you've got a couple billion dollars lying around. Thank you for watching this episode of SciShow, and thank you to the people at NASA in Northrop Grumman not only for building this amazing piece of machinery but for letting me come see it and ask them questions about it. If you have any questions or comments or ideas, we're on Facebook and Twitter and in the comments below, and if you wanna keep getting smarter with is here at SciShow, you can go to youtube.com/scishow and subscribe. [outro music]