We often think of outer space, the bit between stars, as a complete vacuum. The reality is that, while it is a better vacuum than any we can create on Earth, it is far from empty. The interstellar medium (ISM) fills the space between stars in a galaxy. Almost all of the ISM is hydrogen and helium that formed in the early universe. The rest of is small particles of cosmic dust.

Composite image of the central part of the Orion Nebula, near where molecular oxygen was found. Credit:{link url="http://www.eso.org/public/images/eso0104a"} ESO/M.McCaughrean et al.{/link}

The densest parts of the ISM are molecular clouds, which are dense enough that molecules can form within them. Molecular hydrogen, H2, is the most common inhabitant of molecular clouds, but a recent discovery in one particular cloud — part of the Orion molecular cloud complex — has revealed another constituent, as well as the answer to a longstanding question in astronomy: molecular oxygen does exist in space.

Oxygen is the third most abundant element in the universe. Astronomers have known that atomic oxygen exists in space for a while now, although the amount of atomic oxygen is less than expected. Searches for molecular oxygen — two atoms of oxygen stuck together, O2 — had been, so far, unsuccessful.

Now, molecular oxygen has been found near the Orion Nebula by Paul Goldsmith, from the Jet Propulsion Lab at Caltech, and an international team of collaborators. Their paper, which will be published in the Astrophysical Journal, explains how Goldsmith and his colleagues used the Herschel telescope to detect signals that gave away the existence of molecular oxygen.

They knew they had detected molecular oxygen when they saw three specific spectral lines using the HIFI far-infrared instrument on Herschel. The relative intensities of these lines enabled them to estimate the temperature of the source of the O2 to be between 65 and 120 Kelvin (-208 to -153C).

A series of reactions that culminate in molecular oxygen formation are triggered when a cosmic ray ionises molecular hydrogen, H2, turning it into H2+. The H2+ then reacts with more H2, resulting in H3+, which then reacts with atomic oxygen, O, to make OH+. When the OH+ bumps into hydrogen molecule, the reactions leads to H2O+ and H3O+, which in turn combine with electrons and make water (H2O) and OH. The OH can react with atomic oxygen to make O2. (Phew!)

Each step in this long chain of reactions has a reaction rate, many of which have been estimated using measurements in the lab. The rate of the final step, which produces the O2, has been a topic of debate, but even the lowest estimates should produce lots of molecular oxygen in all but the coldest clouds of gas and dust. So why wasn't any O2 spotted before?

Two missions that have looked for molecular oxygen and failed are Sweden's Odin mission, and NASA's Sub-millimetre Wave Astronomy Satellite (SWAS). When each of these missions released results showing that molecular oxygen abundance must be lower than expected, astronomers tried to explain why this might be the case.

One hypothesis put forward was that oxygen could be freezing onto dust grains in areas with low temperatures. On the surface of the dust grain it would be converted into water ice, making it invisible to missions looking for oxygen.

Astronomers realised that when the dust grains in a molecular cloud are warm enough, the dust grain's surface should release the water ice in which the oxygen is locked up. This meant that warm areas should contain more oxygen gas, with lots of it in the form of molecules.

Believing that warmth would be key to finding oxygen, Goldsmith and his colleagues pointed the telescope towards a region of Orion near which stars are forming. They hoped the star formation would heat up the surrounding gas, allowing oxygen molecules to exist on their own, not locked up on the surfaces of dust grains. And it must have, because Goldsmith and his colleagues found what they were looking for. By their estimation, in the region they looked at there was one oxygen molecule for every million molecules of hydrogen.

Goldsmith and his colleagues are sure that they have found molecular oxygen, but admit that there are still questions to be answered about the source of the emission. For starters, they didn't see a lot of oxygen, and they still do not fully understand why some regions have more oxygen than others. But it's a start.


Paul F. Goldsmith, Rene Liseau, Tom A. Bell, John H. Black, Jo-Hsin Chen, David Hollenbach, Michael J. Kaufman, Di Li, Dariusz C. Lis, Gary Melnick, David Neufeld, Laurent Pagani, Ronald Snell, Arnold O. Benz, Edwin Bergin, Simon Bruderer, Paola Caselli, Emmanuel Caux, Pierre Encrenaz, Edith Falgarone, Maryvonne Gerin, Javier R. Goicoechea, Ake Hjalmarson, Bengt Larsson, Jacques Le Bourlot, Franck Le Petit Massimo De Luca, Zsofia Nagy, Evelyne Roueff, Aage Sandqvist, Floris van der Tak, Ewine F. van Dishoeck, Charlotte Vastel, Serena Viti, & Umut Yildiz (2011). Herschel Measurements of Molecular Oxygen in Orion Astrophysical Journal arXiv: 1108.0441v1