See more: “Scenes from a universe gathering dust“
THE universe, it seems, has the same problem I have. Dust, everywhere. Looking round as I type this, I find myself wondering where on earth it all comes from. Just how does it accumulate without invitation on every surface, in every nook and cranny?
Increasingly, astronomers can be heard muttering something similar. Unlike my bothersome but insignificant pilings, cosmic dust is important stuff. Its wispy grains, mainly formed of amorphous carbon, carbonate and silicate, are just fractions of a micrometre across – about the size of a smoke particle. If it weren’t for them, though, the night sky would be much brighter, with thousands of extra stars on view. Not that stars could exist in such numbers without dust: its presence cools down clouds of interstellar gas and aids their collapse into stars. In addition, small molecules meet and bind on the grains’ surfaces, allowing more complex chemicals to form than would be possible through chance encounters in the cosmic outback. Cosmic dust is the starting point for building whole planets and more besides. Go back far enough in time, and dust is the stuff that made us.
Advertisement
Yet we have a problem with cosmic dust, one so big we can’t just sweep it under the carpet: we don’t know what made it.
Not so long ago, we thought we did. Long-lived stars in their final stages of existence were the dust factories. When a star like the sun ages, changes in its internal chemistry cause it to bloat and turn into a red giant many times its original size. Once the sun goes that way, it will be curtains for Mercury, Venus and possibly Earth, but a happy day for dust fans. Our swollen star’s tenuous outer atmosphere will provide a perfect environment for solid grains to condense from hot gas, like soot in a chimney flue.
In the Milky Way, this picture fits the facts rather well. The amount of dust we see lying about the place is about what we would expect from the number of red giants we estimate have formed and died in the galaxy’s 10-billion-year history. But one galaxy does not a cosmos make. “As soon as we started looking beyond the Milky Way, the problems became clear,” says , an astronomer from University College London.
The critical point came in the 1990s, with the advent of telescopes capable of seeing dust in very distant galaxies just a few hundred million years after the birth of the universe. Stars need at least a billion years before they enter the red giant phase, so by rights these galaxies should be pristine and dust-free. In fact, they are chock-a-block with dust. A typical example is the galaxy J1148+5251. Despite being observed only 900 million years after the big bang, it contains more than 10 times as much dust as the Milky Way.
“There is no doubt any more that the universe was dustier in the past than today,” says , an astrophysicist from University of Copenhagen in Denmark. Clearly, something other than red giants is making dust in these distant galaxies, and doing it very well, too. But what?
The observations suggest a culprit: the live-fast, die-young stars destined to explode as supernovae. Such stars pass through a brief red-giant phase before blowing themselves up, and there would have been plenty of time for them to have formed and died in galaxies such as J1148+5251. What’s more, with rates of star formation higher in the early universe, we would expect there to have been more supernovae back then, and so more dust, too.
The problem is that the violence of a supernova’s nuclear explosion should blast any dust formed to smithereens, returning it to its constituent atoms. Nevertheless, Matsuura and her colleagues followed a hunch and pointed the instruments of the European Space Agency’s Space Observatory at SN 1987A, the remains of a star that exploded 25 years ago in the Large Magellanic Cloud, a dwarf neighbour of the Milky Way. SN 1987A is the nearest supernova to occur since the invention of the telescope, and an essential test bed for theories about exploding stars.
Supernova shocker
Whenever a photon of visible light collides with a grain of dust, its energy is absorbed and heats the grain a little. This heat is then re-emitted as a tiny blip of infrared radiation. With a mirror 3.5 metres across, Herschel is the largest space telescope yet flown, and ideally placed to sweep up dust signals.
Matsuura and her team last year, and they were a shocker. By conventional thinking, the supernova should have produced, at most, about a tenth of the mass of the sun as dust. But the observations suggested between four and seven times that amount drifting around in the aftermath of the explosion – enough to make more than 200,000 Earths. Scale that up to a galaxy the size of the Milky Way, and you can forget about red giants: all the dust we see in it could have come just from supernovae ().
“Scale the result up to a galaxy the size of the Milky Way, and you can forget about red giants: all the dust we see could come just from supernovae”
So, it seems clear: dust comes from supernovae, not from red giants, and we need to adjust our theories of what happens when stars form and die to take account of that.
But not so fast.
“That’s an extreme amount of dust for a supernova,” Mattsson says. He questions some of the assumptions Matsuura had to make to convert the amount of light Herschel observed into the amount of dusty mass the supernova produced. Mattsson thinks that although the supernova is creating the ingredients of the dust, it is not moulding them into solid particles, but leaving them as molecules of gas.
Yet if it were gas, Matsuura says, the Herschel signal would show that. Carbon and oxygen are among the most abundant atoms produced by a supernova, and they combine readily to form carbon monoxide gas. Carbon monoxide emits radiation at a characteristic infrared frequency, so Herschel should pick up its signature. In as yet unpublished follow-up measurements, Matsuura says, she has found no trace of carbon monoxide.
Even if dust can somehow survive the initial force of a supernova explosion, it is still far from clear whether it would last long enough to enter dust reservoirs in the wide open spaces of a galaxy. Shock waves would reverberate around the supernova’s vicinity for centuries and probably shatter the grains. “Dust survival is an unknown quantity,” admits Matsuura. “We are hoping it is going to survive but we don’t know yet.”
The 1987 explosion is too recent for us to tell, but a glance at the sites of other nearby supernovae is not encouraging. The next youngest one is a 300-year-old remnant called Cassiopeia A, in our own galaxy. Its dust mass is lower, in line with Mattsson’s estimates, although that depends on measurements of infrared light which may be masked by other dust in our line of sight along the Milky Way.
If all that dust in faraway galaxies cannot come from red giants, because they didn’t exist in the very early universe, and cannot come from supernovae, because they destroy most of what they make, what then? For the past 20 years, of Princeton University has been proffering a third alternative: that dust is growing of its own accord in those relatively empty bits of galaxy in between stars.
In his view, that must be so, because supernova shock waves criss-crossing the galaxy would destroy not only almost all supernova dust, but also that produced by red giants. But slightly denser “molecular clouds” in interstellar space might provide a refuge from the worst shock-wave depredations. Dust could slowly accrete there, with those few particles to have survived being ejected from supernovae or red giants perhaps forming the nucleus for further dust formation. The dust clouds eventually become dense enough to collapse into new stars and planets.
Is that a credible alternative? Ultimately, answering that will mean capturing and analysing some of the dust itself. That’s not so difficult, even though the nearest supernovae and red giants are out of reach of our spacecraft. “We’re constantly ramming into these particles,” says , a mineralogist at the Natural History Museum in London. The solar system itself orbits the centre of the galaxy, ploughing through clouds of dust given out by all manner of processes.
Reach for the stardust
Since 2004, Kearsley has been working with colleagues at NASA to analyse the grains that the Stardust mission has bumped into. Launched in 1999, Stardust circled the sun and encountered the dust trails of two comets before parachuting its samples back to the Utah desert in 2006. Most of the dust it collected came from the comets, although quite a bit was shards that fell off the spacecraft itself. And thanks to a project called , in which 30,000 members of the public inspected the impact tracks made by the dust grains as they hit the spacecraft’s collectors, a few grains could be identified as following paths that began outside the solar system.
Analysing the composition of these grains should reveal their ultimate origin, as quite different cocktails of isotopes are concocted in the nuclear furnace of a supernova and the cooling outer layers of a red giant. The grains’ fluffiness, meanwhile, should tell us something about how long they took to accrete, and so what part, if any, molecular clouds played in their formative years. Kearsley expects the first results to be out next year. “It will finally give us some ground truth to work with,” he says.
If it doesn’t, we have other options. Already gathering measurements is the in Chile, a radio telescope that works at wavelengths where dust shines brightly. Japan is planning an infrared space telescope similar to Herschel, , for launch in 2018, and the same year should see the launch of the , the long-delayed, infrared-optimised successor to NASA’s Hubble Space Telescope.
Between them, these instruments should be able to study dust in older supernova remnants and galaxies even further away in the distant cosmos, and perhaps finally sweep aside the accumulating mysteries of the stuff that made us. As a shaft of sunlight briefly illuminates a swirling cloud in my office, it is good to know this is one problem where the dust is not going to settle.
