Where did the planet Jupiter come from?

Collision in space

In July of this year, we had the unique chance to observe a process that, four and a half billion years ago, led to the formation of the earth in multiple repetitions: the collision of Jupiter and a comet that it had already cast under its spell twenty years ago . Since then, "P-Shoemaker-Levy 9" has been circling the planet on an elliptical orbit. The last encounter was fatal for the comet. At the Institute for Theoretical Astrophysics, Günther Wuchterl explored how celestial bodies the size of Jupiter are formed. The astronomical scientist, who is now doing research at the Institute for Astronomy at the University of Vienna, used powerful computers to draw a new picture of the planet's birth in the sky.

Anyone who has ever wondered why the sun rises in the east, where the days, months and years have their duration from, why Venus only shines in the evening and in the morning, while Jupiter can be visible all night, who has wondered that the moon has a "face" that he always shows us from the same side, or that he found it remarkable that the planets describe their orbit on such a narrow segment of the sky, and not least because of this a concept like that of the zodiac was able to establish itself in astrology, in short: Whoever has lost the matter of course with this or that person, who wants to know where all this comes from, actually asked about the origin of the planetary system.

It was in 1986 at the Vienna University Observatory - our investigations into star formation were just beginning to bear fruit - when the idea arose to try something different with the radiation hydrodynamic methods developed in this context: to find a new trace of the problem of planet formation get.

The formation of the planets is seen today as a side effect of the formation of stars, both are closely interwoven in the genesis of the solar system. It is also instructive to view the gas planets, such as Jupiter and Saturn, as a kind of failed star: they were able to collect large amounts of hydrogen and helium as they grew, but not enough to ignite nuclear reactions inside them. In contrast to the suns, they have therefore been cooling since the time they reached their current mass. Your temperature has already reached absolute zero. The earth-like planets in turn correspond to a kind of failed gas planet in this picture.

At the time when we started our work on planet formation, the astrophysicists already knew that planet nuclei growing in the solar nebula - which consist of that portion of matter which is condensable under the conditions of the solar nebula - can bind gas to themselves. Through this process the formation of protoplanets with up to twenty Earth masses could be explained. However, this value was far below the masses of Jupiter, with 317 earth masses, and Saturn, with 97 earth masses. The growth of protoplanets beyond the twenty masses of the earth up to today's masses of the heavy gas planets was not understood. It seemed as if a "critical" state in the evolutionary history had been reached at twenty earth masses, since the original model assumptions collapsed here. The astronomers suspected that beyond the "critical" mass, the gas envelope of a protoplanet would have to collapse under the force of gravity of its core. In the course of this collapse, shell gas - until then a few earth masses - would be compressed to a much smaller volume around the core. This would create the conditions for the flow of additional gas from the solar nebula, and the protoplanets could grow into a Jupiter or Saturn in a very short time.

It was precisely this hypothesis on the evolution of protoplanets beyond critical mass that aroused our interest. Thanks to the advanced work on star formation, we found ourselves in the fortunate position of already having ample experience in studying collapse currents. For us, it made sense to take a closer look at what is happening around the critical mass and to investigate the question of what the expected collapse in the case of the protoplanets might look like.

In December 1987 preparations had progressed so far that the first calculations could begin. For this we had to modify the procedure developed for the protostellar collapse for the solution of the radiation hydrodynamic equations in such a way that it was suitable for the investigation of the protoplanetary development. The critical state could be reached quickly with our calculations, the agreement with the results of previous work proved to be good: As expected, the protoplanet had at this point of its development twenty times the mass of the earth. But when we finally ventured beyond the critical moment, the surprise came: the protoplanet did not collapse, it began to pulsate! With each oscillation it lost some mass and finally two thirds of the envelope mass flowed at supersonic speed from the area of ​​influence of the planet into the surrounding solar nebula.

Would the material flow back again and then perhaps the expected increase in mass occur? It took six months to calculate the further development until - arithmetically - a few decades had passed since the kick. The matter stayed away. The protoplanet had changed into a new state, its mass had decreased and was remarkably similar to that of Uranus and Neptune. The investigations of the calculated currents showed that the pulsations are driven by the so-called k mechanism - named after the measure of the (in) transparency of stellar matter, k. The same mechanism also describes the cause of the star pulsations. The principle of the k-mechanism for the excitation of pressure waves is comparable to that of the diesel engine for performing mechanical work, with the difference that with the k-mechanism in the compressed state of the matter, the heat is not supplied by spontaneous ignition, but by the heating up as a result of the increased light absorption of matter in a condensed, more opaque state.

In autumn 1989 the time had come for the work to be continued at the Institute for Theoretical Astrophysics at the University of Heidelberg within the framework of the project on gas planet formation, funded by the DFG for over four and a half years. Powerful remote data connections made it possible to carry out invoices on the Cray-2 in Stuttgart from Heidelberg; the repetition of the first calculations now only took a few hours instead of several months. This opened up the possibility of comparing the formation of planets, for example calculating events at variable distances from the sun and under the specification of different conditions in the solar nebula. Perhaps at a certain distance from the sun, Jupiter would be possible? But in all cases the mass outflow was shown, and the planetary remnants always resulted in the presumed forerunners of Uranus and Neptune. It seemed as if the genesis of massive gas planets like Saturn and especially Jupiter could not be deciphered. Did we make our calculations using approximations that were too restrictive? Were the uncertainties in the structure of the solar nebula, the cradle of the formation of planets, so great that the possibility of the formation of Jupiter and Saturn was ruled out from the outset? Could the solution methods for the expansion equations be trusted? The question of how Jupiter was formed became more and more urgent.

The expansion of the graphics environment and accessible storage media at the Heidelberg Institute has meanwhile made it possible to set up an archive of the various planetary developments. That was very decisive, since the calculated development histories of the planets were only available as endless oceans of numbers in which the physical information is scattered in tens of thousands of physical quantities at tens of thousands of points in time. With the help of a higher-level graphic language, however, it was possible to leaf through this data pool like in a book, to understand the results better, to develop intuitions and to bring new ideas to maturity via online assessments. It turned out that with regard to some very important properties of the protoplanets found so far, one "looked" like the other, in such a way that an always the same repulsion of the shell is plausible for this type.

The question immediately arose whether the protoplanets could at some point, under certain conditions, be "different". Quite unexpectedly, their comparatively cool outdoor areas turned out to be very close to "boiling", that is, to the convective movements of their atmospheric gases. Convective movements occur when the deeper layers of the gas envelope are significantly hotter than the layers above. As if by coincidence, many of the planetary developments came close to the "boiling point", but none exceeded it! Should the outer areas of a protoplanet begin to "boil", it would presumably - very similarly to stars - suddenly show a different dependency behavior on the environmental conditions. With the onset of convection, a protoplanet would change its character significantly. There was also reason to assume that non-convective protoplanets are very similar to one another for different solar nebula conditions, whereas convective protoplanets are very different.

So it was necessary to determine the conditions under which a protoplanet would have to become convective in the solar nebula. This step opened up a whole new scenario. It turned out that the new, convective protoplanets can grow up to 80 Earth masses, almost up to Saturn's mass, without becoming "critical". They also have the - hitherto unknown - property that their mass, mediated by the conditions in the solar nebula, depends on the distance from the sun. It is also very remarkable that they have very different total masses with almost constant nuclear masses, a property that can also be deduced from observations for today's large planets with the help of "optimized" models.

In the run-up to a closer investigation of this new class of protoplanets, however, it was necessary to clarify the reasons for the deviations of the results of our hydrodynamic calculations from the quasi-static calculations of our Californian colleagues. The unexpected mass repulsion of the critical protoplanets aroused some astonishment and skepticism among fellow researchers. In the meantime, various works on the basis of less general, quasi-static calculations had shown that, under normal conditions in the solar nebula, planets with at least the mass of Saturn should arise. Why was no mass decrease in development beyond critical condition found in these calculations?

The best possible opportunity to clarify such differences was during a research program held from August to December 1992 at the Institute for Theoretical Physics at the University of California, Santa Barbara. It was supposed to bring all planet formation theorists together in one place over a longer period of time in order to intensively discuss the questions at hand and to work together on merging the various concepts. The conditions found there were also ideal for us insofar as the sufficiently fast intercontinental lines made it possible to calculate on the Cray-2 in Stuttgart, to analyze the results with the help of the graphic environment of the Institute for Theoretical Astrophysics in Heidelberg and to record them in Santa Barbara to browse it on the screen. Those hours were particularly productive when day crosses the Pacific and night falls in the other hemisphere, where by far most computers are located. The global networks are then free.

In a comparison of all common gas planet formation models, a good agreement could be found in the areas for which hydrodynamic effects are irrelevant. In the static limit case, the dynamic calculations agreed with the quasi-static ones: where the dynamic protoplanets actually become "alive" and repel their mass via pulsations, they simply continued to grow in the quasi-static calculations, since the underlying assumptions are pulsations from the Exclude consideration. But this also shows that in the static approximation essential properties are excluded from the investigation. The pulsations and the hull repulsion at the critical mass can only be found in hydrodynamic calculations, i.e. quasi-static calculations generally lose their validity at the critical mass. The question of how Jupiter was to come about was still open. Perhaps the newly discovered class of "convective" protoplanets offered a way out, because among them also massive static solutions with up to 80 earth masses could be found, whereas previously static solutions up to a maximum of 20 earth masses were known for the usual "non-convective" protoplanets. Convective protoplanets only occur under conditions in the solar nebula that had not been investigated until then, but which are very similar to the commonly assumed conditions. The question was how the convective protoplanets would behave beyond the critical mass. After all, it could not be ruled out that the massive hulls that had finally been found would repel everything again at the critical moment, and that Jupiter's formation would again be prevented.

When one of the reviewers of a corresponding article did not really want to believe that there could be such massive "static" protoplanets, we decided to repeat the hydrodynamic calculations without the "static" assumption, and continued them immediately after the critical moment . The results were the same up to critical mass. Then the protoplanets went different ways. The "standard planets" showed the repulsion, the dconvective planets "temporarily began to be slightly restless, but then decided on an ever faster contraction. Their mass increased up to a few hundred earth masses, then they collapsed into much more compact ones Configurations. It was thus possible for the first time to show how gas planets could grow to the mass of Jupiter. The new picture of gas planet formation can now be sketched as follows: If a planet core grows in the solar nebula, it begins to bind gas from the nebula in the earth's atmosphere, gas pressure and gravity are balanced in these planetary shells: there is hydrostatic equilibrium. As Matthias Götz recently showed in his Heidelberg dissertation, rotation only plays a relatively minor role. Beyond the critical mass, there are two qualitatively completely different ones Development opportunities. The "standard planets" repel a large part of their accumulated gas envelope and develop into precursors of Uranus and Neptune. The newly discovered "convective" protoplanets continue to accumulate gas until their envelopes ultimately collapse into a more compact state. In this way they grow up to the much larger masses of Saturn and Jupiter. Here, as in the case of the formation of the planetary nuclei, one is forced to assume that the formation of planets is relatively inefficient overall, that is, more matter must have been present in the solar nebula than previously assumed. Usually "uranus and Neptune-like protoplanets are formed. Jupiter and Saturn require conditions for their formation that are typically found in greater proximity to the sun. Earth-like planets are formed where it is so" hot "that gas is no longer bound to the nuclei How closely the history of the formation of the planets is interwoven with events that are still taking place in the solar system will be brought to mind in a dramatic way in summer this year. Jupiter captured a comet unnoticed about twenty years ago. The comet has been on since then an elliptical orbit around Jupiter. About two years ago there was such a close encounter that the comet was torn apart by Jupiter's tidal forces. Finally, in March 1993, the fragments lined up like a string of pearls along their orbit were discovered as comet "P-Shoemaker-Levy 9" It soon turned out that the next encounter with Jupiter would mean the end of the comet dignity. The 22 or so fragments of Comet Shoemaker-Levy collided with Jupiter between July 16 and 22, 1994, providing a unique opportunity for astronomers around the world to observe the process that was formed many times over four and a half billion years ago of the earth and the cores of the gas planets.

Author:
Dr. Günther Wuchterl
Institute for Astronomy at the University of Vienna, Türkenschanzstr. 17, A-1180 Vienna,
Telephone (431) 4706800