Printable page E-mail this

ON THE face of it, a telescope with a liquid mirror sounds about as useful as a chocolate teapot. The enormous rigid mirrors that capture and focus light inside large telescopes weigh several tonnes, and take many years of painstaking effort to cast, grind and polish—a process that shapes their surfaces to within a few billionths of a metre. So how could a pool of liquid mercury do the same job? The answer is surprising. Pour any liquid into a cylindrical container, rotate the container at a constant speed, and the surface of the liquid will become a paraboloid, which just happens to be the perfect shape for a telescope’s mirror.

This idea is not new. Indeed, it may go back as far as Sir Isaac Newton, the inventor of the reflecting telescope. But liquid-mirror telescopes (LMTs) were, for a long time, seen as of theoretical interest only. Over the past decade, however, the idea has looked increasingly attractive, for both economic and technical reasons. A few LMTs were constructed during the 1990s, and the largest yet made, with a mirror six metres across, will soon capture its first light in the forest east of Vancouver, Canada. Advocates of LMTs believe the technology has now come of age—and that it could even spell the end of conventional telescopes.

Compared with making a rigid mirror, making a liquid mirror is a picnic. The main ingredient is a small amount of mercury, which is placed inside a strong yet light container made of a composite material such as Kevlar. This container is a paraboloid, but its shape does not need to be made as accurately as a rigid mirror. Once the container is spinning at the correct speed (typically a few revolutions per minute), the mercury spreads out to form a layer less than a millimetre thick, and in such a way that variations in the thickness of the mercury layer compensate for the imperfect shape of the container. The result is a liquid mirror as accurately shaped as a conventional rigid one—but at about 1% of the cost.

There is more to a telescope than its mirror, of course. But when the other components and housing are taken into account, an LMT still comes in at around 5% of the cost of a comparable conventional telescope. The Large Zenith Telescope—as the six-metre LMT that is being constructed by Paul Hickson and his team at the University of British Columbia, in Vancouver, is known—is being built on a shoestring, and is expected to cost less than $1m. In comparison, the construction cost for the two eight-metre Gemini telescopes in Hawaii is $184m, and the 4.2-metre Southern Astronomical Research telescope being built in Brazil will cost $28m.

The ability to construct large telescopes for a fraction of the usual cost could transform astronomy. It would mean that individual research groups could have their own telescopes, instead of having to book time on shared instruments months in advance, as they do now. Having sole use of a dedicated telescope would make new kinds of research possible, too. At the moment it is not feasible to dedicate all the observing time on a six- or eight-metre telescope to search for exploding stars known as supernovae in other galaxies. This is a pity since, in an ideal world, such searches would involve repeated observations of the same galaxies every few days. But with an LMT, using a large telescope for such specialised research would become possible.

There is, however, a catch: LMTs can only point straight upwards. Tilting the spinning container causes the mirror to lose its shape, so it is not possible to point an LMT at a particular spot in the sky—and even when an object of interest is directly overhead, it is not possible to track it by moving the telescope to compensate for the rotation of the earth.

This restriction is not as limiting as it sounds, though, at least for certain kinds of astronomy. Cosmologists investigating the structure of the universe do not really mind which direction a telescope is pointing, since the universe is isotropic and homogeneous—which is a fancy way of saying that it looks the same in all directions. Similarly, long-term sky surveys and supernova searches can be done with a zenith-pointing telescope: over the course of a year, the telescope can observe an entire band of the sky.

Another drawback of LMTs has been solved by the advent of digital sensors. Use a photographic plate to record an image from a zenith-pointing telescope, and you will end up with a streaky image, as a result of the earth’s rotation. Use a digital sensor, however, and it is possible to shuffle the image across the sensor in such a way that it is exactly in step with the earth’s rotation.

One of the leading proponents of the liquid-mirror approach is Ermanno Borra, a physicist at Laval University in Quebec. It was Dr Borra’s team that worked out how to coax mercury into forming a film less than a millimetre thick. (Normally, surface tension causes the metal to form much thicker blobs.) Dr Borra also carried out a series of key experiments in the 1990s to show that LMTs would actually work. His research prompted scientists at America’s space agency, NASA, to build a three-metre LMT. This is now the 17th-largest telescope in the world and is used to track space debris—another application where the inability to point the telescope does not really matter that much.

Dr Borra is eager to point out that the Large Zenith Telescope, and a proposed four-metre telescope called the International LMT, are being built to do real science, rather than merely to investigate the properties of LMTs. He is particularly excited by the prospect of being able to dedicate a large telescope to the search for supernovae, since these are used as intergalactic yardsticks by astronomers. The puzzling finding in 1999 that the universe’s expansion seems to be accelerating was based on observations of a few dozen supernovae. An LMT ought to be able to find several thousand supernovae over the course of a year, and confirm or disprove this result beyond doubt.

As well as spreading the word about the advantages of LMTs, Dr Borra is doing his best to overcome their inability to point. A liquid mirror spins once every few seconds, so provided the liquid is sufficiently viscous to stay put for half a revolution when the mirror is tilted, it will not lose its shape. Dr Borra worked out just how viscous the liquid would need to be, and found that it had to be the consistency of thick honey. The problem is that mercury is far less viscous than this; and honey does not reflect light, and so would not work as a mirror. The search thus began for a liquid that was both viscous and highly reflective.

As luck would have it, Dr Borra came across a suitable class of materials. They are called metal liquid-like films. Tiny particles of silver are coated with organic molecules so that they float, and are then added to silicone oil. The result is a thin reflective layer that floats on the surface of the oil.

So far, Dr Borra has achieved a reflectivity of 50% (that is, half of the light falling on the mirror is reflected), but he is confident that he will be able to equal the 80% reflectivity of mercury. At this point it will be feasible to build a tilting LMT. And he reckons that, if he can build a four-metre or larger LMT that can tilt by at least 30°, that will spell the end of the classical telescope.

This is a bold claim. But suppose it were possible to build a telescope that could do everything a conventional telescope can do for a fraction of the price—around 10%, by Dr Borra’s estimate. The impact of his crazy-sounding spinning mirrors on the economics of astronomy would be nothing less than, well, revolutionary.