Eyes In the Sky

One of America’s least known National Historic Landmarks may also be its ugliest. It’s kept hidden inside Building 32 on the grounds of the Johnson Space Center in Houston and is identified simply as Chamber A. The “landmark” resembles nothing so much as a bank vault, albeit one with a 40-ton, 40-ft.-wide door.

When the door is shut, however, and the right machinery is turned on, Chamber A becomes, effectively, a giant pocket of outer space. Pumps create a vacuum, and a liquid helium and nitrogen cooling system drives the temperature down to –440°F, not far from absolute zero, the thermal floor at which most molecular motion stops.

The chamber was built in 1965 and earned its landmark status both for its innovative design and for its work stress-testing the Apollo lunar spacecraft. Now, it’s preparing to inflict its punishment on the next great space machine to come its way: the James Webb Space Telescope.

On a recent afternoon, the main mirror and instrument package of the Webb–named after the NASA administrator who ran the agency in the early part of the Apollo era–sat in the filtered-air clean room outside the chamber, being prepped for a 93-day stay in simulated space. That test, which will begin in July, will be a very high-stakes exercise. The mirror is the heart of the telescope, measuring 21.3 ft. across. It’s made of 18 smaller hexagonal mirrors arranged in a honeycomb configuration. Altogether, the assembly has seven times more light-collecting space than the main mirror of the celebrated but aging Hubble Space Telescope. So big an eye will give the Webb the power to look much farther into space–and much further back in time–than Hubble can. That might reveal something spectacular–possibly the very moment in cosmic history when the first stars switched on.

For a limited time, TIME is giving all readers special access to subscriber-only stories. For complete access, we encourage you to become a subscriber. Click here.

“We will be watching the universe light up after the Big Bang,” says NASA’s Eric Smith, Webb’s program director.

The Webb has been in development for more than 20 years at a cost of $8.7 billion and is at last set to launch in October 2018. In addition to witnessing first light, it may also see the first primal galaxies taking shape, the first planetary systems forming around stars, even signs of early biology–if it exists–emerging on alien worlds. Though Webb is the biggest news in the telescope community, it’s not the only news. NASA is betting big on cosmic observatories. Even before Webb flies, the space agency will launch the Transiting Exoplanet Survey Satellite (TESS), which will conduct a study of the entire 360-degree bowl of the sky, looking for planets orbiting the half-million brightest, closest stars in the galaxy.

After that may come the Wide-Field Infrared Survey Telescope (WFIRST), which, among other things, will study dark energy–the still-mysterious force that is forever pulling the universe outward. At least two more spacecraft observatories are also being developed–to study the universe in the X-ray wavelengths and to look more closely at habitable planets. In all, NASA has earmarked about $9.2 billion for Webb and TESS alone. The other telescopes, which are still in early development, would cost what a NASA spokesperson estimates simply as “several billion dollars” each. But that may be a price worth paying.

“Humankind has always wondered about the universe, and now our telescope technology has caught up with our questions,” says Paul Hertz, NASA’s director of astrophysics. “This is a great time to be a scientist.”

For the Webb telescope, surviving in space may be easy compared to the fight it faced to survive here on Earth–a fight it almost lost. The telescope was proposed in the mid-1990s at a cost of $500 million and was projected to be ready to fly in 2007. But inventing new technology has a way of defying deadlines and confounding cost projections. By 2011, Webb had already burned through $6.2 billion, with no firm launch date in sight.

Congress responded the way Congress often does in these situations, which was to threaten to cancel the whole project. If throwing away billions in sunk costs seemed hard to justify, there was at least some precedent. Familiar with the work of the great American particle accelerator in Waxahachie, Texas? No, you’re not, because it’s nothing but a giant, unused tunnel, one that cost more than $2 billion before Congress lost patience with the similarly behind-schedule, overbudget project and shut it down in 1993.

For the Webb, however, Washington agreed to hold its fire. When the mirror was finally delivered in 2012, the funding spigot was turned back on. “There was strong support from the science community for the mission,” says Smith, “though it was certainly a tense time.”

After that near-death experience, the Webb’s next big challenges will be the ones it will face when it at last gets to work. Unlike Hubble, which flies in Earth’s orbit at an altitude of just 353 miles, Webb will park itself in space about 1 million miles away. There, it will circle a spot known as L2, one of five so-called Lagrange points, where the gravities of Earth and the sun achieve a balance that can hold objects in more or less the same position. That’s a good, safe place for a ship like Webb.

The telescope will do much of its observing not in the optical wavelengths the human eye can see, but in the infrared. The primary source of infrared radiation is heat, and the wavelength can stream straight through the cosmic dust that prevents Hubble from seeing some of the oldest and most remote provinces of space. The problem is, that makes Webb extremely temperature-sensitive; stray heat on its mirror would be like stray light on Hubble’s, washing out images.

Webb will thus turn its back to the sun, Earth and moon, facing out to space with a solar shield protecting it. About the size of a tennis court and roughly diamond-shaped, the shield–which is too large even for Chamber A–is made of five layers of a foil-like material known as kapton. Each layer is as thin as a human hair and is separated from the layers on either side of it by up to 12 in. The temperature on the bottom layer–the most sunward side–will reach about 185°F, not far from the boiling point of water. Each successive layer will get colder and colder–with the vacuum gap between them acting as further insulation–ultimately reaching a low of -370°F on the side of the mirror.

“Five layers gives you enough cooling so that you don’t need an active refrigeration system,” says Smith.

The 18 segments of the mirror are made of beryllium, a metal whose molecular structure can be manipulated into one that functions like glass but that can be polished more predictably and consistently. A thin layer of gold is applied for reflectivity. The gold covers 269 sq. ft. of the mirror, but is so thin that if it were peeled off and tamped down, it would form a mass roughly the size of a golf ball. The beryllium surface, meanwhile, is polished so smoothly that if it were expanded to the size of the U.S., its biggest imperfection would be just 3 in. tall.

The fact that the mirror does not have to be protected from ambient starlight means that it doesn’t have to be enclosed in a cylindrical housing like Hubble’s. Instead, it sits directly atop the sun screen, completely exposed to space. That saves weight, but also exposes the mirror to intermittent micrometeoroid bombardment. “Hubble gets beat by stuff all the time,” says Doug McGuffey, a lead systems engineer for Webb.

What works in Webb’s favor is the micro part of micrometeoroid: even at high speed, the particles don’t have the mass to do catastrophic damage. And if mirror segments do get dinged over time, actuators–or tiny motors–behind them can adjust their position to refocus them. “Damage to one mirror,” says McGuffey, “can be compensated for by the others.”

Such flexibility will help the Webb avoid the kind of problem Hubble faced, when no sooner did it arrive in space in 1990 than NASA discovered that its primary mirror was warped, leaving it nearsighted. It took a servicing mission by space-shuttle astronauts to fix the problem–something that would not be possible at Webb’s million-mile distance.

All of that engineering care will pay off when Webb begins making its observations. An expanding universe like ours presents complexities a static universe wouldn’t. The most remote regions of space retreat the fastest, and the light that speeds toward us from those areas thus gets stretched like a Slinky, with its wavelength shifting toward the red end of the spectrum–the very end Webb is built to see.

The farthest infrared signatures are also the oldest in the approximately 13.8 billion-year-old universe. Webb will get very close to seeing back to the very beginning, picking up signals that have been traveling to us since just 200 million years after the Big Bang, and converting that information to pictures. An image it delivers of, say, a brand-new galaxy won’t be the galaxy as it looks today, but as it looked 13.6 billion years ago–the cosmic equivalent of live-streaming videos of your newborn across a network that takes, say, 80 years to complete the transmission. The baby in the video will be an octogenarian by the time your receiver watches the stream. That time-capsule quality will be true of all of the observations Webb makes of stars and nebulae and other structures at the most distant removes of space.

The sheer ambition of the Webb mission has caused a lot of people to overlook what the telescope’s little sister TESS will do. But that less expensive ($378 million) observatory could make news. TESS will actually get off the pad first, launching from Cape Canaveral in the early part of 2018, and will go into an ordinary Earth orbit, where it will spend two years conducting its whole-sky survey.

The goal is to study the half-million stars closest to Earth, looking for flickering in their light that suggest they are being orbited by planets. The Kepler Space Telescope, launched in 2009, has already led researchers to conclude that virtually every star in the sky has at least one planet, but Kepler trains its gaze up to 3,000 light-years into space. A planet so far away is hard to study, given that a single light-year is about 5.9 trillion miles. TESS will limit its search to 200 light-years or less.

“Kepler’s great achievement was that it gave us the exoplanet population,” says Hertz. “But the exoplanets that are the closest are obviously the ones best suited for follow-up studies.”

The WFIRST mission is not as far along as TESS, merely in preliminary development. The telescope will observe the cosmos in more or less the same wavelengths as Hubble does, but it will take in 100 times more sky in a single viewing–the difference between peering through a straw and peering through a window.

Even after all of these observatories take flight, NASA is roughing out plans for still more–pending budgetary buy-in. Particularly promising are LYNX–an X-ray-frequency telescope that would be especially good at studying black holes–and HabEx, which would analyze the atmospheres of exoplanets looking for signs of gases associated with life, such as methane and carbon dioxide.

It says something both odd and exceptional about our species that while we could rightly be preoccupied with the simple business of surviving on the one world we’ve got–keeping the people in our own small tribe fed and healthy and safe from the perceived menace of the tribes across the valley–we always have one eye trained outward. We can’t say exactly what we’re looking for–deliverance, company, answers to eternal questions–but we look out all the same.

Building the instruments that make that wondering gaze possible isn’t easy or cheap, and none of it pays the kinds of earthly dividends that pick-and-shovel programs like fixing roads or building airports do. But there are other kinds of dividends as well, and if uncovering the universe’s most ancient secrets doesn’t qualify, what would? Washington could certainly spend its money more frugally, but it’s hard to see how it could spend it more imaginatively.