What is Richard Muller's Nemesis Theory

Far away from all telescopes that help to explore the universe, on April 6, 1992 a small area of ​​the cosmic view of the world began to shake. Astronomers had come together for an international conference on double stars at the Callaway Gardens Inn, a hotel on Pine Mountain, which is just 250 meters high, 100 kilometers south of Atlanta (US state Georgia). The discipline does not actually expect any revolutionary upheavals from the experts in this research area, because the objects of investigation often take decades to circle each other. But this time the individual findings presented by several scientists - the results of laborious observations with tricky procedures and the most modern equipment - came together to form a very surprising but uniform picture: According to this, contrary to previous assumptions, even the youngest stars often have stellar companions.

To the layman it may seem entirely plausible that young stars appear in pairs at least as often as older ones, but this discovery came as a shock to astronomers. According to most theories about the formation of such systems, stellar partners should not be formed or captured by a star until long after a star is formed. Consequently the youngest stars would have to stand alone in space.

But such ideas suddenly turned out to be out of date on that day. But after the discussions at the Callaway Gardens Inn, at least one model of binary star formation remained that is consistent with recent observations and perhaps the only explanation why such paired groupings of large celestial bodies are so common.

A rare loner

Most stars as old and developed as the Sun are in multiple systems; but our day star has, as far as we know, no gravitationally coupled counterpart. Richard A. Muller of the Lawrence Berkeley Laboratory in California and some of his colleagues nevertheless hypothesized in 1984 that the sun was orbited by a distant companion star in an elliptical orbit with a period of about 30 million years. They thought that they could explain the catastrophic extinction of animal species that occurs approximately at these intervals: every time the companion approaches the sun, its gravitational effect could disturb the orbits of lumps of matter in the outer regions of the solar system in such a way that a shower of comets directed to the inner planets and also bomb the earth. Because of this ominous quality, Muller named the sun's hypothetical partner "Nemesis", after the goddess who in Greek mythology personified retribution, punishment and disapproval of human arrogance.

Most astronomers, however, reject Muller's interesting idea. The closest known stars (those of the triple system Alpha Centauri) are 4.2 light years far too far away to be gravitationally bound to the sun; And so far there is no indication that our central star is anything other than a single star whose largest companion - namely the planet Jupiter - has a mass 1000 times smaller than itself.

But apparently the fact that humanity lives on a planet orbiting a stellar loner distorts our idea of ​​the cosmos. Star systems with two, three or more sun-like components are not the exception, but rather the rule.

Multiple systems

Antoine Duquennoy and Michel Mayor of the Geneva Observatory completed an extensive survey of nearby binary stars in 1990. They took into account all those who are less than 72 light-years away from the sun and how these belong to the so-called G-dwarfs. The sample thus comprised 164 principal components that should be representative of stars in the galactic disk. The two researchers stated that only a third of these systems can be regarded as true single stars; the rest have companions that are more than a hundredth the mass of the Sun, which is about ten times the mass of Jupiter.

Binary star systems are very different. In some G-dwarfs, the components are almost touching, in others they are almost 0.3 light-years apart. Accordingly, the orbital times vary from a few hours to a few tens of millions of years. Duquennoy and Mayor showed that triple and quadruple G dwarf stars are much rarer than double: They counted 62 different double systems, but only seven triple and two quadruple systems.

Furthermore, they found that the mutual distances in the multiple systems are hierarchically graded: A relatively close pair has either (in a triple system) a common center of gravity with a distant single star or (in a quadruple system) with another close double star. In order for such a configuration to remain stable for a long time, the distance between the distant partners must be at least five times that of the close pair. There are also arrangements with smaller distances (which are called trapezoidal systems after a young four-fold star in the Orion Nebula), but their relative movements are unstable, so that they will eventually fly apart. For example, if the components of a triple system all approach each other, the one with the lowest mass will generally be thrown out, leaving a stable pair of stars behind.

Can such complex multiple systems have planets? Absolutely, because stable orbits are possible as long as the planet is close to one of the stellar components or at a great distance from all of them (Fig. 4).

The inhabitant of a world that orbits a close double star at a safe distance, for example, could perceive fantastic spectacles: the daytime sky would be illuminated by two closely spaced suns that swapped their positions within a few days; and it would be fascinating to watch how at sunrises and sunsets first one glowing ball appears above the horizon and then the other, or sinks behind it. Other strange celestial phenomena could also occur, for example if the two stars cover each other at periodic intervals and then only the light of a sun reaches the planet.

Star Formation Regions

The sun was formed about 4.6 billion years ago, about half the time it took to get its energy from burning hydrogen. In about five billion years, when it has reached the end of what is known as the main sequence stage, it will expand into a red giant encircling the inner planets.

Even at the beginning of its history, the sun's radius was far larger than it is today. At that time it resembled a class known as T-Tauri stars that can be observed in the current star-forming regions of the galaxy. During their T-Tauri phase, the radius was about four times the current value of almost 700,000 kilometers. Even earlier the protosun must have had an extension of about 1.5 billion kilometers, which corresponds to ten times the current distance between the earth and the sun, i.e. ten astronomical units.

From today's T-Tauri stars, astronomers can learn what the Sun looked like at an early stage of its evolution. The closest of these objects are in two regions of the sky, the Taurus and the Rho-Ophiuchi Molecular Cloud, which are each about 460 light-years away from Earth (Fig. 1). The fact that young stars are always embedded in such concentrations of gas and dust is a convincing indication of their origin: They arise through contraction and subsequent collapse of denser regions in hydrogen molecule clouds.

Because young stars are mostly surrounded by large, opaque amounts of dust, they remain hidden from the human eye, even with the most powerful telescopes. At infrared wavelengths, however, their dust covers, which have been heated up by the radiation from the stars, can be detected quite easily. The development of infrared detectors therefore contributed significantly to a better understanding of star formation. In 1992, during the conference on Pine Mountain, Georgia, several astronomers presented the first results of their infrared surveys with which they sought to find stellar companions of T-Tauri stars in the constellations Taurus and Ophiuchus.

Andrea M. Ghez, who now works at the University of California in Los Angeles, and her colleagues Gerry F. Neugebauer and Keith Matthews from the California Institute of Technology in Pasadena used a newly developed electronic camera on the five-meter Hale telescope on the Mount Palomar to photograph the vicinity of well-known T-Tauri stars in the near infrared at a wavelength of 2.2 micrometers. (Visible light has wavelengths between 0.4 and 0.7 micrometers.) To improve the resolution, they used what is known as speckle interferometry, a process in which many images are exposed in rapid succession for only a few milliseconds, so that on them atmospheric turbulence, which would smear the picture, are, as it were, frozen. As it turned out, almost half of the 70 T-Tauri stars examined have stellar companions at distances between about 10 and 400 astronomical units - thus binary systems seem to be twice as common among the youngest stars as among main sequence stars (Fig. 3).

Christoph Leinert from the Max Planck Institute for Astronomy in Heidelberg and his colleagues also used speckle images in the near infrared. Of the 106 T-Tauri stars they examined, 43 have close companions. This also indicates that double systems are far more common in such young objects than in the sun-like G-dwarfs.

Hans Zinnecker and Wolfgang Brandner from the University of Würzburg and Bo Reipurth from the European Southern Observatory in La Silla (Chile) used a high-resolution digital camera on the European New Technology Telescope to record 160 T-Tauri stars at an infrared wavelength of one micrometer (Picture 2). They discovered 28 companions with distances of 100 to 1500 astronomical units to the main component - about a third more than would be expected with corresponding older sun-like stars.

Michal J. Simon from New York State University in Stony Brook presented a novel method for finding young double stars that he had used with Weng Ping Chen - who is now at the National Central University of Taiwan in Chungli - and several colleagues: Because the moon orbiting the earth once within four weeks, it occasionally covers a background star on its apparent orbit in the celestial sphere. If this is actually a multiple system, the components disappear one after the other behind the sharp edge of the lunar disc. The presence of stellar companions can then be revealed by high-resolution observations of starlight, including those that are very close to the main component and can no longer be detected with infrared cameras. The measurements by Simon's team again showed that a large part of the T-Tauri stars are double systems.

The same conventional method used by Duquennoy and Mayor was used by Robert D. Mathieu of the University of Wisconsin at Madison. By spectroscopic measurement of the Doppler shift, which varies with the orbital period, he demonstrated that some T-Tauri stars have companions with distances of less than one astronomical unit. It is also true for such close binary star systems that their frequency is higher under young stars than under sun-like stars (Fig. 3).

Looking for a theory

Where do all these stellar companions come from? Why did they arise so often and at such an early stage of development? From the many observations presented at the Callaway Gardens Inn, it follows that binary stars must form long before their pre-main sequence phase (the T-Tauri stage), and that whatever type of mechanism, this mechanism is extremely efficient.

In principle, a binary star system could arise if two single stars pass so close to each other that they would force each other into a stable orbit. But for this a third object would have to be present, which is able to absorb the excess kinetic energy. Such three-body encounters are too rare to contribute significantly to the registered double star frequency.

Cathy J. Clarke and James E. Pringle from the University of Cambridge (England) investigated a configuration that is more common: the gravitational coupling between young stars that are still surrounded by flat disks of gas and dust. These circumstellar disks could - at least theoretically - absorb enough kinetic energy; as the closer analysis showed, however, they would be much more likely to be torn apart. Consequently, this beautiful idea does not seem to be able to explain the formation of binary star systems either.

Therefore the astronomers had to think again about more direct formation processes. The British physicist William Thomson (1824 to 1907) - later Lord Kelvin - had already speculated in 1883 that double stars were created by "rotational fission". Based on studies on the stability of rapidly rotating bodies, he pointed out that a star must rotate faster and faster when it contracts - similar to an ice skater doing a pirouette. Eventually it would break and a double star would emerge from the rubble. Astronomers now know that pre-main sequence stars contract considerably as they approach hydrogen burning - the main sequence stage; but T-Tauri stars don't rotate fast enough to become unstable. In addition, the mechanism proposed by Kelvin would come in too late to be able to explain the frequency of young binary star systems. And as Richard H. Durisen of Indiana University in Bloomington and his colleagues showed, such an educational process must also fail for theoretical reasons: Calculations showed that the matter ejected from the original mass would end up as a spiral gas flow and not as a separate star.

A model developed only a decade ago, fragmentation, seems to provide a more realistic explanation. According to this, double systems arise when a dense molecular cloud collapses under its own gravity and forms protostars; As soon as the initially existing envelope of gas and dust has largely dissolved, the newly created double star (of the T Tauri type) becomes recognizable. In contrast to older theories, the fragmentation is completely in line with the most recent observations of young stars.

Compared with the development phase of stars lasting several billion years, the protostellar collapse that enables fragmentation takes place relatively quickly: within a few hundred thousand years. This turbulent compaction of a diffuse cloud, which can be called almost spontaneous for astronomical conditions, offers a special opportunity for a concentration of matter to break down into several subunits. Apparently, as astrophysicists determined, temperature and angular momentum are the main factors: Very cold clouds can fragment directly into binary stars, whereas warmer ones with stronger rotation initially form thin disks and later, when they have accumulated more mass or have become increasingly flat, break up.

Nebulous ideas

The most significant objection to the theory of fragmentation concerned the distribution of matter in protostellar clouds. It used to be assumed that it could be described by a power law. According to this, the matter would essentially be concentrated in the vicinity of the cloud center, and its density would rapidly decrease towards the outside (Fig. 6). Under such conditions, however, as Elizabeth A. Myhill of the University of California in Los Angeles and I independently showed, multiple systems could hardly arise.

Current radio astronomical observations in the submillimeter range with high resolution seem to suggest a different form of matter distribution: Last year, a group led by Derek Ward-Thompson from the Royal Observatory in Edinburgh (Scotland) examined various protostellar clouds that have not yet collapsed and found them that the density can rather be described by a (bell-shaped) Gaussian distribution. In this case the matter at the beginning of the star formation process would be less concentrated near the center than with a power law; fragmentation could therefore set in much more easily.

That this is indeed the case can be shown by astrophysicists by solving equations which describe the particle motions and the radiation transport in a protostellar cloud. I had already started modeling dense clouds with Gaussian profiles in 1986, when a sufficiently powerful computer and reliable software were available.According to these calculations, the cloud should split up during the gravitational contraction if it rotates fast enough to give the emerging binary star system the required angular momentum, and if the matter is cold enough before the collapse that its thermal energy is less than half the gravitational energy ( which applies to temperatures below 10 Kelvin) - these conditions are, it seems, completely fulfilled in the gas and dust clouds of the star formation regions.

Whether a double, triple or quadruple star ultimately forms depends on many other details, such as the original shape of the cloud, how inhomogeneously the matter is distributed in it and the exact values ​​of the thermal and rotational energy. In general, prolate (shaped like an elongated egg) clouds tend to form bar-shaped structures that further fragment into binary stars (Fig. 7), while oblate (disc-shaped) clouds flatten into disks, which then disintegrate into several components.

The collapse presumably takes place in two phases. First of all, extensive masses with radii on the order of ten astronomical units are created. As a result, fragmentation in this phase can only generate binary systems with greater spacing. Only in a second step do these bodies collapse further, whereby the actual protostars with stellar dimensions are formed. However, as Ian A. Bonnell and Matthew R. Bate from the University of Cambridge have shown, fragmentation can occur again. Protostellar nuclei can emerge from this, the distances between which are comparable to those of the closest main sequence stars. This hierarchically graded fragmentation can explain all the distances between the components that are observed in young binary stars - from the narrowest to the furthest systems (Figs. 5 and 7).

Brown dwarfs and giant planets

Can the mentioned observation methods also detect companions with an even lower mass? Duquennoy and Mayor found evidence that up to ten percent of Sun-like stars could have so-called brown dwarfs as partners - objects that have only 0.01 to 0.08 solar masses (about 10 to 80 Jupiter masses) and are therefore too small to be they would ignite the hydrogen present in them, as is the case in stars. However, they could be massive enough to fuse deuterium shortly after they were formed. After this reaction had subsided, they would no longer generate any more energy and would slowly cool down. Once cooled, they would be extremely difficult to spot. Although the clues found by Duquennoy and Mayor give hope and the search for brown dwarfs is intense, there has been no confirmed report of success so far.

Planets - which have even smaller masses than brown dwarfs - have also been eagerly on the lookout for a long time. Within the next ten years, it should be possible to improve observation methods to such an extent that planets the size of Jupiter can be discovered (or their existence can be ruled out) in some stars near the Sun. Whether it makes sense to look for them in binary star systems or to limit oneself to individual main sequence stars cannot be answered unequivocally. Astronomers will likely target both groups in their quest to discover extrasolar planetary systems.

(Shortly before this issue went to press it became known that two brown dwarfs and an extrasolar planet had apparently been discovered. R. Rebolo, MR Zapatero Osorio and EL Martin from the Astrophysical Institute of the Canary Islands on Tenerife have found a Pleiades in the 400 light-years distant Pleiades open star clusters, an object with a weak near infrared radiation - called Teide 1 - discovered, the mass of which is about 20 to 30 masses of Jupiter. Due to the young age of the Pleiades of about 100 million years, Teide 1 is still in the contraction phase, so it is sufficient Converted gravitational energy into radiation energy in order to be able to detect it. A group led by Geoffrey Marcy from the University of California at Berkeley was able to detect lithium in the atmosphere of the object PPl15, also located in the Pleiades, with the 10-meter Keck telescope on Hawaii - and thus prove that it is also a brown dwarf whose mass is not enough enough to trigger fusion reactions. Finally, on October 6, Mayor and his doctoral student Didier Queloz presented observations at a conference in Florence that indicate that the star 51 Pegasi, which is about 40 light-years away from the Sun, is orbited by a planet that lies between 0.6 and 2 , Has 5 times the mass of Jupiter. A few days later, Marcy and his colleague Paul Butler succeeded in confirming this finding. The editorial office.)


- With lunar covers on the hunt for young double stars in the Taurus. By Michal Simon and Christoph Leinert in: Stars and Space, Volume 31, Issue 6, Pages 380 to 385, June 1992.

- Formation of Binary Stars. By Alan P. Boss in: The Realm of Interacting Binary Stars. Edited by J. Sahade, G. E. McCluskey Jr. and Y. Kondo. Kluwer Academic Publishers, 1993.

- Stellar Multiple Systems: Constraints on the Mechanism of Origin. By P. Bodenheimer, T. Ruzmaikina and R. D. Mathieu in: Protostars & Planets, Volume 3. Edited by E. H. Levy and J. I. Lunine. University of Arizona Press, 1993.

- A Lunar Occultation and Direct Imaging Survey of Multiplicity in the Ophiuchus and Taurus Star-Forming Regions. By M. Simon, AM Ghez, C. Leinert, L. Cassar, WP Chen, RR Howell, RF Jameson, K. Matthews, G. Neugebauer and A. Richichi in: Astrophysical Journal, Volume 443, Issue 2, Part 1, Pp. 625-637, April 20, 1995.

- Pre-Main-Sequence Binary Stars. By R. D. Mathieu in: Annual Review of Astronomy and Astrophysics, Volume 32, pages 465-530, 1995.

- Discovery of a Brown Dwarf in the Pleiades Star Cluster. By R. Rebolo, M. R. Zapatero Osorio and E. L. Martin in: Nature, Volume 377, pages 129-131, September 14, 1995.

From: Spektrum der Wissenschaft 12/1995, page 46
© Spektrum der Wissenschaft Verlagsgesellschaft mbH

This article is contained in Spectrum of Science 12/1995