Astronomers discover dusty galaxies at ancient epoch, track build-up of star- and planet-forming material

Oct 09, 2012
A small portion of one of the CANDELS fields; small circles indicate galaxies included in the survey. Galaxies seen at various distances are circled in colors according to epoch. Galaxies at a redshift of 4, or 1.5 billion years after the Big Bang, are circled in magenta, while galaxies at 5, 6, 7 and 8 are circled in blue, green, yellow and red, respectively. The three insets show a zoomed in view of three galaxies; the upper-left panel is one at a redshift of 4, the lower-right is at a redshift of 6, and the lower-left is at a redshift of 7. This composite image is made up of an optical image (shown in blue) by Hubble Space Telescope as part of the GOODS survey, combined with an infrared Hubble image (red and green) taken in dedicated CANDELS observations. Credit: S. Finkelstein/CANDELS Collaboration/STScI/NASA

(Phys.org)—Dust is an annoyance in everyday life, but an important building block of stars and planets. As such, astronomers need to understand how cosmic dust forms over time—it's an integral step in figuring out the evolution of galaxies, and the stars and planets within them.

 To better understand , University of Texas at Austin assistant professor Steven Finkelstein and colleagues are pursuing one of the largest projects to date, studying in thousands of galaxies over a wide range of cosmic time. They published some early results in a paper lead by Finkelstein in a recent issue of The .

 "We don't yet understand how galaxies build up their dust reservoirs," Finkelstein said. "We know that dust builds up through time, but exactly when the formation of dust begins is unknown."

 Finkelstein is part of a large team of astronomers working to rectify that . They are studying nearly 3,000 galaxies seen 500 million to 1,500 million years after the Big Bang—only a moment after the initial event, when compared to the 13.7-billion-year . The project is called CANDELS: the Cosmic Assembly Near-infrared Deep Extragalactic Legacy Survey.

 Some of the galaxies came from the team's own ongoing Hubble observations: more than 900 orbits studying distant galaxies with the 3 (WFC3). They also include data from several other large Hubble galaxy surveys (including Great Observatories Origins Deep Survey (GOODS) and the Hubble Ultra-Deep Field survey).

 Finkelstein said that previous studies with smaller surveys, including some of his own, appeared to indicate that ancient galaxies were dust free. But CANDELS has found that even at a very early epoch, already contain a lot of dust, in the form of grains of carbon and silicon (in astronomical jargon, "").

The red circles from left to right represent the measured average color of galaxies at a redshift of 4, 5, 6 and 7, where the bottom axis shows the time since the Big Bang. The light blue bar running through the center of the diagram shows the color of the local galaxy NGC 1705, which contains no dust. When CANDELS astronomers saw that the redshift 7 galaxies in their sample have a similar color as NGC 1705, they derived that those are also dust free. The gradual reddening they observed at lower and lower redshifts reveals that the galaxies are getting dustier with time. Credit: S. Finkelstein/CANDELS collaboration

"We found something we wouldn't expect," Finkelstein said. "Although dust can form quickly, I don't think many people expected galaxies at only 800 million years after the Big Bang to have a lot of dust. These observations caused us to change our thinking."

 Studying such ancient, and thus faint, galaxies is tricky even for Hubble. Only a miniscule amount of information comes through in the tiny stream of photons they send our way, but their color can be determined. This was the team's quarry. Galaxy colors are a clue to the amount of dust a galaxy contains: The redder a galaxy appears, the more dust it contains. The bluer it appears, the less dust it contains.

 And finding a significant amount of "heavy metal" dust in these early massive galaxies means they must have been forming stars for a while, Finkelstein said. That's because heavy elements were not created in the Big Bang itself. They are built up over time inside stars, as they fuse lighter elements into heavier ones through nuclear fusion at their cores. When a massive star runs through all of this nuclear fuel, it explodes as a spectacular supernova, spewing these heavy elements into the galaxy. These heavy elements are the building blocks for the dust for which CANDELS was looking.

 "These results are very interesting because they tell us that dust does form at early times," he said. "This is important because the same elements that compose the dust grains are necessary for the formation of planets. Also, we think that dust is a key component in allowing hydrogen gas to form molecules, which is necessary for star formation."

 Additionally, the team found that in between 800 million and 1.5 billion years after the Big Bang, "all galaxies—not just massive ones—get dusty," Finkelstein said.

 This work has him excited about the future, Finkelstein said. "The presence of dust means that a previous generation of stars has lived and died. So, when we can peer back to even farther later this decade with JWST [the James Webb Space Telescope], there should be a lot for us to see!"

Explore further: An unprecedented view of two hundred galaxies of the local universe

Related Stories

Galaxies in the young cosmos

May 21, 2012

(Phys.org) -- The universe was born about 13.7 billion years ago in the big bang. The Sun and its system of planets formed about five billion years ago. What happened, then, during that long, intervening stretch ...

Mysterious red galaxies

Dec 12, 2011

(PhysOrg.com) -- Perhaps the most astonishing and revolutionary discovery in cosmology was Edwin Hubble's observation that galaxies are moving away from us with velocities that are proportional to their distances. ...

Backgrounds

Jun 28, 2011

You’ve probably heard of the cosmic microwave background, but it doesn’t stop there. The as-yet-undetectable cosmic neutrino background is out there waiting to give us a view into the first seconds ...

'Big baby' galaxy found in newborn Universe

Sep 28, 2005

The NASA/ESA Hubble Space Telescope and NASA’s Spitzer Space Telescope have teamed up to 'weigh' the stars in distant galaxies. One of these galaxies is not only one of the most distant ever seen, but it appears to be unusually ...

Recommended for you

The origins of local planetary orbits

17 hours ago

A plutino is an asteroid-sized body that orbits the Sun in a 2:3 resonance with Neptune. They are named after Pluto, which also orbits the Sun twice for every three orbits of Neptune. It is thought that Pluto ...

Wild ducks take flight in open cluster

18 hours ago

The Wide Field Imager on the MPG/ESO 2.2-meter telescope at ESO's La Silla Observatory in Chile has taken this beautiful image, dappled with blue stars, of one of the most star-rich open clusters currently ...

Image: The Pillars of Creation

19 hours ago

The Pillars of Creation (seen above) is an image of a portion of the Eagle nebula (M16) taken by Hubble Space Telescope in 1995. It soon became one of the most iconic space images of all time. The Eagle nebula ...

User comments : 10

Adjust slider to filter visible comments by rank

Display comments: newest first

GSwift7
5 / 5 (1) Oct 09, 2012
So, when we can peer back to even farther later this decade with JWST [the James Webb Space Telescope], there should be a lot for us to see!"


It would be such a disappointment if that thing doesn't work. If it fails, I doubt there would be a comparable replacement in my lifetime.

I think JWST will unravel several of the big questions remaining.
Tuxford
1 / 5 (6) Oct 09, 2012
Face it. The model should be called, the BBF, the 'Big Bang Fantasy'. Wave those hands. Say it ain't so!
Shinichi D_
5 / 5 (1) Oct 09, 2012
So, when we can peer back to even farther later this decade with JWST [the James Webb Space Telescope], there should be a lot for us to see!"


It would be such a disappointment if that thing doesn't work. If it fails, I doubt there would be a comparable replacement in my lifetime.

I think JWST will unravel several of the big questions remaining.


GMT, TMT, E-ELT are under construction or in the final design phase. They will come close to JWST, or even exceed it in some aspects. So don't worry. We are actually going to see the first stars of the universe by 2025-2030.
Shinobiwan Kenobi
3 / 5 (2) Oct 10, 2012
Face it. The model should be called, the BBF, the 'Big Bang Fantasy'. Wave those hands. Say it ain't so!


So there's another theory out there that lines up with all recorded observations thusfar?! Let's all join hands and herald the enlightenment Tuxford will bestow upon us, for surely, there must be a logical framework we have yet to understand that Tuxford will elucidate upon to warrant such a dismissal!
Torbjorn_Larsson_OM
5 / 5 (4) Oct 10, 2012
Face it. The model should be called, the BBF, the 'Big Bang Fantasy'. Wave those hands. Say it ain't so!


Even a cursory examination of Wikipedia's Big Bang article shows that the current inflationary standard cosmology is enormously predictive. In other words, it is precisely _not_ handwaving. It is also the first self-consistent cosmology, so there are no obvious errors that would make us think it can be replaced with anything radically different.

Take your fantasies of alternate cosmologies elsewhere, we know they do not longer belong on science sites.
Tuxford
1 / 5 (3) Oct 10, 2012
Shino, you can start here, as I did.

http://phys.org/n...ars.html

It takes courage to think different. Do have courage? Well...do ya???
cantdrive85
2 / 5 (4) Oct 10, 2012
Surprise, we'll add this to the heaping piling of observations that require these scientist to "rethink" or "adjust the models to fit the observation". Contrary to bjorn's claims of predictability, nearly every observation comes with surprise and unexpected results, just as this article points out.
Fleetfoot
5 / 5 (3) Oct 10, 2012
GMT, TMT, E-ELT are under construction or in the final design phase. They will come close to JWST, or even exceed it in some aspects. So don't worry. We are actually going to see the first stars of the universe by 2025-2030.


Current simulations allowing for DM condensation starting before normal matter decouples from the CMBR suggest that the first stars would have formed around z=65 (age = 32 million years) which is beyond even JWST's range. If Pop IIIa stars could end as PISN then they would dump huge amounts of metals back into the mix so seeing some dust as early as 50 million years is entirely possible.

Articles often use a "back to the drawing board" hook to raise interest but in reality progress is often more incremental in nature.
ValeriaT
2.3 / 5 (3) Oct 10, 2012
Although dust can form quickly, I don't think many people expected galaxies at only 800 million years after the Big Bang to have a lot of dust. These observations caused us to change our thinking
Of course, such an observation may serve as an evidence of steady-state cosmology, albeit still inconclusive. The formation of dust would require the formation of metallic elements (silicon and heavier elements) together with oxygen during explosions of stars in few stellar generations, which would indeed require some time under normal circumstances. When the interstellar gas is concentrated, it may take few hundreds of millions of years for stars to explode in supernova, but the primordial matter is supposed to be very diluted and homogeneous because of inflation.
Fleetfoot
4.2 / 5 (5) Oct 10, 2012
I don't think many people expected galaxies at only 800 million years after the Big Bang to have a lot of dust.

Of course, such an observation may serve as an evidence of steady-state cosmology ..


Troll alert! :-)

The graph above would be horizontal in steady state.

The formation of dust would require the formation of metallic elements (silicon and heavier elements) ..


Actually "metals" means lithium and heavier in astrophysics.

.. When the interstellar gas is concentrated, it may take few hundreds of millions of years for stars to explode in supernova, ..


The gas density is irrelevant, it is the stellar mass and metallicity that determine its lifetime. Low metallicity stars are very poor radiators so can grow much larger than in later populations, perhaps up to 300Msun. Pop III stars are therefore expected to have had lifetimes as short as ten million years. That's 77 generations before the universe was 800 million years old (starting at z=65).