Distantly detecting deuterium

John N. Bahcall

A new era of astronomy is heralded by the article¹ on page 599 of this issue. There, Songaila et al. present two results that could be of fundamental significance to cosmology: the first tentative detection of deuterium outside our Galaxy, at a large redshift (z = 3.3), and a meaningful upper limit on the temperature of the cosmic background radiation at a different large redshift (z = 2.9).

The first result extends the tests of the Big Bang picture for the synthesis of light elements in the early Universe to deuterium, which is an exquisitely sensitive measure of the total amount of baryonic matter (the protons and neutrons of `ordinary stuff') in the Universe. It is either a tentative measurement or a definitive upper limit of the abundance of deuterium, as explained below. Previously, among the light elements, only the abundances of ordinary hydrogen and of helium had been measured outside our own Galaxy. The second result tests the essence of the idea that the cosmic microwave background radiation is indeed what it seems to be, a universal remnant of the Big Bang that obeys the laws of general relativistic cosmology.

Astronomers are bumping into each other, dancing in the dark corridors of their observatories. Why are they so happy? They are thrilled because the techniques used by Songaila and her colleagues show that the first of the two giant 10-m Keck telescopes on Mauna Kea, Hawaii, is a great success, and that some of the questions that astronomers have sought to answer for decades may be solved in a night's observations with these new eyes. The authors recount how they and other astronomers have previously struggled unsuccessfully, using more modest telescopes (4-m and 5-m class), to measure the deuterium abundance and the temperature of the cosmic background radiation at large redshifts.

The key to the present success is the marvellous operation of the Keck telescope, which captures light from distant objects with an efficiency and a sensitivity that makes easy what was until recently impossible. For the sensitive cosmological tests that Songaila et al. attempt, it is necessary to observe faint objects far away with a high ratio of signal to noise. Obviously, in this case, big is better.

Songaila et al. analyse a nearly primordial cloud of absorbing hydrogen gas that is illuminated by the light of a distant quasar, first examined in detail by Chaffee and collaborators² using the Multiple Mirror Telescope of the University of Arizona. This cloud has many characteristics that permit an accurate determination of the ratio of deuterium to ordinary hydrogen, D/H. The ration that Songaila et al. measure, D/H = 2.5 × 10^-4, is much higher than the ratio determined—after significant corrections for local effects—in the nearby interstellar medium of our own Galaxy³. Moreover, the high deuterium abundance indicated by the new observations has the effect, within the theory of nucleosynthesis of light elements, of reducing the calculated total abundance of baryons to about the value that is estimated from observations of all of the luminous matter in the Universe⁴. This inference, if correct, would imply that the massive haloes of unseen material that are believed to surround ordinary galaxies must be made of nonbaryonic matter, a result of enormous significance for physics and astronomy.

In one of those remarkable coincidences that are surprisingly common the the history of science, R. Carswell⁵ and an international collaboration of scientists have obtained very similar results on the deuterium-to-hydrogen ratio in exactly the same hydrogen cloud. The two analyses are independent and use different data obtained with different telescopes and spectrographs (Carswell's team used the 4-m Kitt Peak National Observatory telescope in Arizona), so the similarity of their answers increases confidence in the results.

A word of caution is appropriate. Songaila et al. estimate that there is a three percent chance that the absorption line that they think is caused by deuterium might instead be caused by absorption from a different hydrogen cloud, one that just happens to be placed so as to be confused with deuterium from the cloud they are studying. Carswell et al. make a similar cautionary statement. Not to worry: Keck can make many such observations. We should know within a year or two whether this first cosmological observation of the deuterium-to-hydrogen ratio was confounded by bad luck.

According to the basic ideas of relativistic cosmology, the temperature of the cosmic background radiation must scale in proportion to (1 + z), where z is the redshift at which the radiation is measured. It was suggested a quarter of a century ago⁶ that this predicted increase of temperature with redshift could be tested by observing the populations of excited fine-structure lines in the absorption spectra of distant quasars. Earlier attempts to observe this effect were limited by the signal-to-noise ratios obtainable with smaller telescopes^7,8. Songaila and co-workers, with the first Keck observations searching for excited fine-structure lines, find a 2 upper limit of 13.5 K for the temperature of the cosmic microwave background in a cloud that contains⁹ singly ionized carbon at z = 2.9. This upper limit lies very close to the temperature, 10.7 K, expected on the basis of relativistic cosmology. Songaila and her colleagues appear confident that definitive measurements at large redshifts will be possible with Keck.

These two new observational tests of relativistic cosmology will be among the many important questions that can be investigated once the new generation of large optical telescopes becomes more readily available to astronomers. Basic ideas of homogeneity and isotropy that are inherent in the standard Big Bang hypothesis predict that the primordial deuterium-to-hydrogen abundance ratios measured in different directions and at different cosmological distances will all be the same. Similarly, if the ideas of relativistic cosmology are correct, the temperature of the cosmic microwave background must be the same in all directions at the same distance and must scale as expected with redshift. All of this will be tested experimentally in future.

Astronomers and physicists are confident of the outcome, nearly certain that the fundamental concepts of cosmology and of Big Bang nucleosynthesis will once again be found to be consistent with observations. But perhaps present-day scientists are slightly less confident than their intellectual predecessors. Michelson and Morley, who set out to measure the velocity of the Earth with respect to the aether.

1. Songaila, A., Cowie, L. L. Hogan, C. J. & Rugers, M. Nature 368, 599-604 (1994).
2. Chaffee, F. H., Foltz, C. B., Roser, H. J., Weymann R. J. & Latham, D. W. Astrophys. J. 292, 362 (1985).
3. Linsky, J. L. et al. Astrophys. J. 402, 694-709 (1993).
4. Walker, T. P., Steigman, G. Schramm, D. N., Olive, K. A. & Kang, H. S. Astrophys. J. 376, 51-69 (1991).
5. Carswell, R. F. et al. Mon. Not. R. astr. Soc. (in the press).
6. Bahcall, J. N. & Wolf, R. A. Astrophys. J. 152, 701-729 (1968).
7. Meyer, D. M., Black. J. H., Chaffee, F. H., Foltz, C. & York, D. G. Astrophys. J. 308, L37-L41 (1986).
8. Bahcall, J. N., Joss, P. C. & Lynds, R. Astrophys. J. 182, L95-L98 (1973).
9. Sargent, W. L. W., Steidel, C. C. & Boksenberg, A. Astrophys. J. 69, 703-761 (1989).

John N. Bahcall is at the Institute for Advanced Study, Olden Lane, Princeton, New Jersey 08540, USA.