How do you build a mirror for one of the world’s biggest telescopes?


Buddy Martin and Dae Wook Kim from the University of Arizona give a detailed explanation of how telescope optics work.

When astronomers point their telescopes up at the sky to see distant supernovae or quasars, they’re collecting light that’s traveled millions or even billions of light-years through space. Even huge and powerful energy sources in the cosmos are unimaginably tiny and faint when we view them from such a distance. In order to learn about galaxies as they were forming soon after the Big Bang, and about nearby but much smaller and fainter objects, astronomers need more powerful telescopes.

Perhaps the poster child for programs that require extraordinary sensitivity and the sharpest possible images is the search for planets around other stars, where the body we’re trying to detect is extremely close to its star and roughly a billion times fainter. Finding earth-like planets is one of the most exciting prospects for the next generation of telescopes, and could eventually lead to discovering extraterrestrial signatures of life.

Size comparison of optical telescopes’ primary mirrors. Cmglee, CC BY-SA

Detectors in research telescopes are already so sensitive that they capture almost every incoming photon, so there’s only one way to detect fainter objects and resolve structure on finer scales: build a bigger telescope. A large telescope doesn’t just capture more photons, it can also produce sharper images. That’s because the wave nature of light sets a limit to the telescope’s resolution, known as the diffraction limit; the sharpness of the image depends on the wavelength of the light and the telescope’s diameter.

As optical scientists, our contribution to the next generation of telescopes is figuring out how to craft the gargantuan mirrors they rely on to collect light from far away. Here’s how we’re perfecting the technology that will enable tomorrow’s astrophysical discoveries.

Multiple mirrors

The question is how to build something substantially bigger than the current generation of telescopes, which have effective diameters of 8 to 12 meters (26 to 40 feet). One of the biggest challenges is making a bigger mirror to collect the light.

Optical diagram of the Giant Magellan Telescope. Giant Magellan Telescope – GMTO Corporation, CC BY-ND

First, it helps to know the basic optical layout of a telescope, illustrated here by the Giant Magellan Telescope (GMT) that is being built in Chile. A large primary mirror collects incoming light and reflects it to a focus. The light is reflected a second time by the smaller secondary mirror, to form an image on an instrument located at a safe, accessible place below the primary mirror, where the image is recorded.

A mirror much larger than eight meters, made of a single piece of glass, would be too expensive and too hard to handle. Everyone involved in building giant telescopes agrees that the solution is to make the primary mirror out of multiple smaller mirrors. Multiple pieces of glass are shaped and aligned to form one gigantic mirror, called a segmented mirror. Gaps between the segments are acceptable as long as the segments’ surfaces lie on a continuous nearly parabolic surface, called the parent surface.

The three extremely large telescope (ELT) projects now in development have made very different decisions about the design of this segmented primary mirror. Two of the ELTs, the European ELT and the Thirty Meter Telescope, have adopted the approach pioneered by the 10-meter Keck Observatory telescopes in Hawaii – they’ll make a giant mirror out of hundreds of 1.5-meter segments.

The third project, the Giant Magellan Telescope, takes a different tack. Its 25-meter primary mirror will have only seven segments. They’re the largest single mirrors that can be made, the 8.4-meter (28-foot) honeycomb mirrors we produce here at the Richard F. Caris Mirror Lab at the University of Arizona. The GMT’s 3-meter secondary mirror also has seven segments, each paired with one of the primary mirror segments.

Artist’s representation of the seven giant mirrors installed in the Giant Magellan Telescope. Giant Magellan Telescope – GMTO Corporation

Large, stiff and light

Big mirror segments guarantee a smooth surface over their entire large areas. The more segments there are in the primary mirror, the more its accuracy depends on their precise alignment to keep them on the parent surface. Because of the pairing of primary and secondary mirror segments in the GMT, the fine control needed to form sharp images can be done by moving the small, agile segments of the secondary mirror rather than the 8.4-meter primary segments. A second advantage of the 8.4-meter honeycomb mirrors is their strong legacy, including use in what is currently the world’s largest telescope, the Large Binocular Telescope here in Arizona.

One of the challenges of using a large mirror is that it tends to bend under its own weight and the force of wind. The mirror is exposed to wind like a sail on a yacht, but it can only bend by about 100 nanometers before its images become too blurry. The best way to overcome this problem is to make the mirror as stiff as is practical, while also limiting its weight.

We accomplish this feat by casting the mirror into a lightweight honeycomb structure. Each mirror has a continuous glass facesheet on top and an almost continuous backsheet, each about one inch thick. Holding the two sheets together is a honeycomb structure consisting of half-inch-thick ribs in a hexagonal pattern. Our honeycomb mirrors are 70 centimeters thick, making them stiff enough to withstand the forces of gravity and wind. But they’re 80 percent hollow and weigh about 16 tons each, light enough that they don’t bend significantly under their own weight.

Mold for casting an 8.4-meter honeycomb mirror for the GMT. The glass will melt around the hexagonal boxes to form the honeycomb.
Ray Bertram, Steward Observatory