Star gazers

In the mid-1700s, a young German musician, William Herschel, moved to England and took up astronomy as a hobby. He soon realised that his new-found interest was hampered; he couldn’t buy a telescope good enough to give him a sharp view of the moon.

So he made his own, learning how to cast and polish metal mirrors. Within a few years he was building bigger and better instruments, inventing a new alloy to provide even clearer images and spending his days making mirror moulds from horse dung, pouring the molten mixture of copper and tin, then polishing the curved metal by hand for hours on end.

It was with one of these telescopes that he discovered Uranus in 1781; with another, he formulated a theory of the birth of stars within nebulae; and with another, in 1800, he discovered infra-red light.

It is fitting, then, that the latest and largest telescope to be sent into space bears Herschel’s name. Due to be launched on 6 May together with another telescope called Planck, the Herschel space telescope is the European Space Agency’s contribution to the growing field of infrared (IR) astronomy. Its primary mirror is the largest ever sent into space, at 3.5m in diameter, and it will use a battery of new technology to capture and interpret data.

Looking far into the IR, its users hope to explain how stars form in our galaxy and how the first galaxies formed soon after the Big Bang.

Herschel is the first major space observatory to be launched since the Hubble telescope, the stunning images of which have become icons of astronomy. However, Herschel is very much of a new generation. Not only is its mirror larger than Hubble’s 2.4m diameter, it also has a very different range of capabilities. Hubble is mainly an optical telescope, looking a little way into the ultraviolet and UV on either side, but that means it can only investigate some science.

‘Telescopes are time machines,’ said John Thatcher, a specialist in space optics and instrumentation from EADS Astrium. ‘The further away things are, the further back in time we see them; but because they’re moving away from us, the light that comes from them is red-shifted.’ This is a property of energy -emitting objects moving away from the viewer — the faster they are moving, the longer the wavelength of the light appears when it reaches the viewer. The expansion of the universe is accelerating, so the light from distant objects is red-shifted deep into the IR spectrum. ‘Also, if you want to look at how stars and planets form from dust and gas clouds, you need IR, because the dust absorbs visible light,’ Thatcher added.

The IR region is barely explored, explained Professor Matt Griffin of Cardiff University’s school of physics and astronomy. ‘Most of it is blocked out by the Earth’s atmosphere, although there are some little window-regions in the spectrum where transmission is poor, but usable on the ground.’ So to get full access to the IR spectrum, you need to go above the atmosphere.

Earth’s orbit was fine for the Hubble; it lifted it above atmospheric haze and light pollution. But it’s no good for IR astronomy. ‘If we were close to the Earth, first of all the planet would fill half the sky,’ said Griffin. ‘But also, the Earth radiates energy. We need to get away from that warmth, because we’re using IR detectors and the radiating heat would just swamp them completely; we’d see nothing.’

Therefore, both Herschel and Planck are going to a quite different orbit. The destination for these spacecraft is the second Lagrange point of the Earth-Sun system, known as L2. At 1.5 million km from the Earth — four times the distance from the Earth to the Moon — it is a solar orbit, but one with special properties. At the five Lagrange points, the gravitational fields of the Earth and the Sun interact with the centrifugal forces acting on a small orbiting body, so it can maintain its position relative to the two larger bodies with little additional power.

L2 is on the opposite side of the Earth to the Sun — normally, the more distant the orbit, the slower the rotation: so Herschel should lag behind the Earth. But because of the peculiar balance of forces at Lagrange, the Earth effectively tows the satellite, so it always stays exactly between the telescope and the Sun. ‘It’s a good place to be pointing away from the Earth, Moon and Sun, which all radiate; plus there’s a good view all around,’ said Ralph Cordey, science and business development manager of Astrium UK.

The L2 orbit, then, ensures that Herschel is cold and shielded from radiation; it uses a sunshade to keep the telescope at around 40K. ‘It emits radiation at the same wavelengths we’re looking at, which is a problem; imagine being on Earth and using an optical telescope that was luminous,’ said Griffin. But the starting point for the observatory was that huge mirror.

Jean Dauphin, Astrium’s director of earth observation and satellites, headed the team in Toulouse that designed and built Herschel’s optics. ‘Because we’re looking in the far IR at distant objects, there are a relatively small number of photons coming in to the telescope. That means you need the wide collection-capacity. The scientists concluded that a 3.5m diameter was absolutely necessary to collect enough photons for the instruments to analyse: anything smaller than that and it would have no scientific interest at all. Moreover, 3.5m was the limit of what was possible for an Ariane launch — any bigger and it wouldn’t have fitted inside the fairing.’

Back in the mid-1990s, when Herschel was in its design phase, ESA studied all the materials that could be used to make a space telescope. ‘We looked at carbon-reinforced plastic, aluminium, titanium, a special kind of glass used for optics called Zerodur, even beryllium, but none of them were suitable,’ Dauphin said. ‘The plastic couldn’t be polished with enough accuracy. The Zerodur was fine for that, but the mirror alone would have weighed a tonne. You couldn’t put a tonne of mirror on top of an Ariane; it would have destroyed the satellite’s structure.’

The answer came in a material that Astrium had been investigating for space optics: silicon carbide. ‘It was originally used to make seals for pumps in the automotive and chemical industries, but its properties were very suitable for optics. It is lightweight, very durable, it can be polished quite easily and with the same accuracy as Zerodur.’ Moreover, it can be made into any shape, as it is formed by sintering: a fine SiC powder is pressed into a mould and heated in a furnace to form a solid.

Even though Astrium developed the sintering process to increase the size of components from 15-20cm to around 2m, there was no way to make the mirror in one piece, so Dauphin’s team had to develop a method for welding pieces together. ‘It’s a tricky process because the welds have to be perfect, as the whole mirror is then machined, ground and polished,’ Dauphin added. ‘We use another oven, with a specific patented process, to join together the 12 petals of the mirror into the single 3.5m disc.’ After polishing, the mirror’s surface has a roughness of less than 30nm.

The entire telescope assembly — primary and secondary mirrors, supporting struts and all the other harnesses and structures — is made of SiC and weighs 350kg. ‘This is not only the largest mirror to go into space, it is the largest SiC mirror ever made,’ Dauphin said.

With the L2 orbit so far away, there is no chance of the telescope being serviced once in operation, so the team had to be certain the mirror would give clear images. ‘It has to be demonstrated and qualified before launch, so we tested it when it was installed into the spacecraft and fully aligned and we’ve done environmental tests in thermal vacuum and under vibration. We’re very certain that it’ll perform. If there turns out to be a focus problem, we can do small alterations by controlling the temperature of the primary mirror, but that would be an exceptional event.’

The telescope sits on top of the 7.5m-high, 4.5m-wide satellite, immediately above the payload — the scientific instruments that will analyse the light it collects. While the mirror is cooled passively, by shielding it from the Sun, the instruments must be colder still, just 1.5K. Therefore, they sit inside a cryostat — a 2.5m-high vacuum flask —through which liquid helium will circulate throughout the mission.

There are three instruments inside the cryostat: the Photodetector Array Camera and Spectrometer (PACS); the Spectral and Photometric Imaging Receiver (SPIRE) and the Heterodyne Instrument for the Far Infrared (HIFI). Matt Griffin, the principal investigator for SPIRE, explains that all three have an important role in the observatory’s mission and were designed to work in concert with each other.

‘SPIRE and PACS both have a camera, to measure the brightness, luminosity and temperature of the objects they are observing, and a spectrometer, to look for patterns of absorption of IR that are characteristic of elements and molecules in the stars, gas and dust clouds. PACS covers wavelengths from 60µm to about 100µm, then SPIRE takes over up to 670µm. HIFI is like a radio receiver; it tunes into celestial wavelengths to look at the radiation emitted by the structures we’ll observe with very precise spectral resolution, to give us information on their composition.’

All the instruments are multi-part, with their detectors in the cryostat and their control electronics outside, nearer the Sun, where they can operate at a range of temperatures. But even the 1.5K helium bath isn’t cold enough for the SPIRE and PACS detectors. ‘We have internal coolers in the instruments that cool the detectors down to 0.3K, which gives us the best performance. For that, we use a different kind of helium, the He3 isotope, which has one less neutron and boils at a lower temperature.’

There are other subtle differences between PACS and SPIRE. For example, the spectrometer in SPIRE looks at the whole spectrum from a body in a single glance, using a mathematical technique called Fourier Transform to separate out all the absorption signals. PACS, which looks at a narrower temperature range, uses a grating spectrometer, which scans along wavelength by wavelength to look for absorption signals.

Griffin’s work will begin soon after launch. After about a week, the Herschel instruments will be switched on, then during the 60-day journey to L2, the instruments will be checked out. ‘Midway through that, there’s a very important event: the removal of the lid that is currently over the cryostat. It’s currently protecting the instruments inside, but once we’re in vacuum and it’s cooled down, we’ll take the lid off and the instruments will see the sky. There will be a certain amount of nervousness,’ he said. Once at L2, final checks will be carried out, then Herschel will have around three years to scan the skies. After that, the helium in the cryostat will have evaporated and the instruments will no longer work; the huge eye will be blinded.

But by that point, another, even larger, IR telescope should have joined Herschel at Lagrange. The James Webb Space Telescope, NASA’s successor to Hubble, is now beginning to take shape, scheduled for a 2013 launch. James Webb looks different from Herschel. Instead of a cylinder, it is a flat structure, with a tennis-court sized sunshield cooling its primary mirror. This mirror will be 6.5m-wide and made from 18 hexagonal segments rather than Herschel’s single disc. It is so large it will be folded for launch and deployed in flight, with three segments on either side unfolding to create the full array. Each segment can be moved independently through seven degrees of freedom to focus the array so that it acts like a single mirror.

The telescope isn’t entirely American. One of its two instruments, MIRI (Mid-IR Instrument) is a collaboration between the Jet Propulsion Laboratory and 21 institutions in 10 European countries, each funding its own efforts; John Thatcher is the project manager for the European consortium. Both a camera and a spectrometer, MIRI will look at wavelengths from 5-25µm, to help the telescope fulfil its mission: to look for ‘first light’, when the post-Big Bang universe had cooled to a few thousand Kelvin and radiation could flow freely, rather than being absorbed by ionised gases.

Like Herschel, James Webb needs to be cold, but not quite as cold. Its five-layer sunshield, made of thin leaves of aluminised film, reduces the temperature from a few hundred Kelvin on the sunward side to about 40K in the shade, and its instruments are cooled mechanically down to about 6K.

‘Nominally the mission is about 10 years, but because of the passive and mechanical cooling, rather than using cryogens that boil away, we are limited by things such as electronics failure and micrometeorite strikes; we could go on for longer,’ Thatcher said.

Astrium got involved because the principal investigator, Gillian Wright of the UK Astronomy Technology Centre in Edinburgh, assembled a team with experience of cryogenic IR astronomy, but mainly using Earth-based instruments. ‘We got the job of engineering management and project assurance,’ Thatcher explained.

Currently, the consortium members are starting to assemble the flight version of MIRI after building and testing several trial versions. ‘This is the exciting bit,’ Thatcher added.