High energy astronomy began with observations of the Sun. The first cosmic X-rays were detected from the Sun in the 1940s, and the first X-ray telescopes were used to study the Sun as part of the Skylab Mission in 1973.
These observations of the Sun opened the eyes of the astronomical community to the potential of high energy astronomy in general and of focusing X-ray telescopes in particular. Now, with the HEAO experiments, X-ray astronomy is returning the favor. Before now, astronomers who specialize in the study of the hot outer layers of stars had only one object, the Sun, to study. Their theories as to what made the upper atmosphere or corona of the Sun so hot were frustratingly difficult to test. That has changed dramatically in the last five years. The Einstein Observatory has detected X-ray emission from the coronas of almost 200 stars of all types; the number should go over 1000 by the time all the data are analyzed.
The analysis of all the data may take several years, but it took only a few months to realize that the old theories were wrong.
Surveys using the Einstein Observatory guided by the results of HEAO 1 experiments have shown that all stars are X-ray sources at some level. The level of X-ray power from many of these stars cannot be explained by the old models for the turbulent outer layers of stars. To see why and to get some idea as to what the new results are telling us, let us take a brief look at the structure of the Sun and other stars as we understand it.
The Sun is a hot ball of gas about 1400 000 km in diameter. That is, it is about a hundred times larger than Earth. The visible surface of the Sun, called the photosphere, has a temperature of about 6000°C. Above the photosphere are two layers that are best observed visually during a solar eclipse, the chromosphere, which means color sphere, because of its predominantly red color, and the corona, or "crown." Temperatures in these layers range from 4000° C at the base of the chromosphere to a million degrees in the corona. Because of the high temperature of the corona, most of its radiation is emitted in the X-ray band. Studies of the X-ray emission from the Sun and other normal stars are therefore primarily studies of the coronas of these stars.
The solar corona and now the coronas of other stars provide us with a cosmic laboratory for investigating how hot gases are produced in nature and how magnetic fields interact with hot gases to produce flares, spectacular explosions that release as much energy as a million hydrogen bombs. They also give us a means of probing beneath the surface of the star, since the production of a hot corona depends on what is happening there.
The energy that makes the Sun shine is generated in a huge nuclear fusion power plant buried deep in the core of the star. There, at temperatures in excess of 10 million degrees and densities exceeding that of lead, nuclear reactions are fusing hydrogen nuclei together to form helium nuclei with the release of energy. This energy gradually leaks out of the core and diffuses up toward the surface of the star.
About four-fifths of the way out from the center of the Sun, the energy can no longer be carried upward efficiently by radiation. A rolling, boiling motion does the job better. Columns of hot gas rise, transfer energy to the cooler surface gas, and descend again. Since the energy is carried upward by mass motions or convection, the region in which this occurs is called the convection zone. These motions will generate sound waves; it was suggested that the dissipation of these sound waves is the source of heat for the solar corona. Elegant theories were formulated, and many detailed calculations involving hours of computer time were performed to demonstrate the validity of this concept, which went under the general name of acoustical heating.
The accompanying figures summarize the downfall of the acoustical heating model. The X-ray photo of the Sun from Skylab shows the X-ray emission to be highly structured. Analysis of the Skylab data reveals that the X-rays come from closed tubes of magnetically confined hot gas,....
....contrary to the predictions of the acoustical heating model, according to which the corona should be a featureless fog Iying above the photosphere.
X-ray photographs of star associations in the constellations of Cygnus and Carina show several sources that can be identified with stars that are much hotter and much larger than the Sun. These stars, called O stars, are not expected to have a vigorous convection zone. Therefore, they should not have acoustically heated coronas; however, they obviously have hot coronas.
 Observations with HEAO I and 2 experiments of a number of peculiar double star systems show evidence for very hot coronas. These star systems consist of a star similar to the Sun and a cooler (K type) star that is thought to have a "starspot" analogous to, but much larger than, the sunspots on the surface of the Sun. The intensity of the X-ray emission and the temperature of the gas are both too high to be explained by acoustic heating. The temperatures are so high that strong magnetic fields, presumably near the starspot, are required to prevent the corona from evaporating.
An X-ray photograph of the Alpha Centauri star system shows that all three stars in the system have hot coronas. Visually, the system consists of a star much like our Sun (a G type star), a nearby companion K type star which is slightly smaller and cooler than the Sun, and a very small M type dwarf star, Proxima Centauri. At a distance of slightly over four light years, Proxima Centauri is the nearest star to the solar system. It is also one of the smallest stars known, having a diameter and mass less than one-tenth that of the Sun.
The acoustical heating model predicts that the G type star should be more than 10 times brighter in X-rays than the K type star, but the observations show just the opposite. The K star is the stronger X-ray source of the two.
Proxima Centauri is highly variable in X-rays. This is not surprising, since it is a flare star, as are most red dwarf stars. The surfaces of such stars are apparently covered with huge starspots over which large flares occur almost continually, producing sudden, unpredictable changes in the optical, radio, and X-radiation.
These flares are probably due to convulsions in the twisted magnetic loops of hot gas above the starspots. This is apparently what happens on the Sun, and it seems to fit for flare stars.
These examples show that the standard theory for the origin of hot coronas around stars is inadequate. It is now apparent that magnetic fields must play a key role, not just on flare stars, but in all stars. To see how this might work, we turn again to the example of the Sun.
As the Skylab X-ray photograph shows, the X-ray emission from the Sun comes from groups of hot magnetized loops. What is the connection between the magnetic field and the X-ray emission? In the case of solar flares, the link is strong. A twisting or shearing of the magnetic field seems to occur before the flare.
Hours and sometimes days before a solar flare occurs, the magnetic field in a certain region is observed to be twisted and sheared into increasingly complex shapes. This twisting increases the energy stored in the magnetic field in much the same way as twisting a rubber band or a coiled spring increases the tension and the energy stored in these systems. The twisting also induces electric currents in the magnetized loops in much the same way that the rotating magnetic field in the alternator of an automobile drives an electric current to charge the battery.
 If the loops get too twisted and the currents get too large, the system "overloads." Something analogous to a short circuit occurs, and a violent explosion rips through the stressed region, producing high energy particles, powerful blasts of radiation in all wavelength bands, and majestic prominences surging high above the surface of the Sun. This is roughly what happens in a solar flare and is probably a fair description of what is occuring on flare stars and other flaring stellar coronas.
The same general process can probably also explain the existence of hot magnetized loops on the Sun and other stars. In most cases, the twisting of the magnetic field will not be violent enough to produce a flare. The energy stored by the twisting can be converted into heat. Electric currents can do this in a manner similar to the way electric currents heat the coil on a stove top or electric heater. This heat will appear in the form of hot magnetic loops. If these loops can get rid of any excess energy by radiating it away or by conducting heat down to the cooler surface layers, then the loops will be stable and will produce a more or less steady glow in X-rays.
It appears, then, that the origin of hot coronas around stars can be traced to the twisting of magnetic fields. The strength of the surface magnetic field and the rate that it is twisted or stressed determine the intensity of the X-ray emission from stellar coronas. In turn, the field strength and the rate of twisting are probably determined by a combination of three factors: the field strength deep within the star, the structure of the convection zone, and the rate at which the star is rotating. The interaction of the boiling motion and the star's rotational motion with the deep Iying magnetic field produces magnetic bubbles that float to the surface of the star, forming magnetic loops. These loops are further twisted by a combination of turbulence and rotation on or near the surface.
The strength of the magnetic field, and presumably the coronal X-ray activity, should depend on the amount of twist in the magnetic field. This should depend in some way on the number of times the star has rotated in the time it takes a bubble to rise to the surface. This is because the twisting is accomplished by dragging the magnetic field around with the rotating star in much the same way that a drop of cream is dragged out and twisted around by the rotation produced by stirring coffee in a cup.
This means that stars that are rapidly rotating and stars that have deeper convection zones should have stronger magnetic fields. Red dwarf stars, because of their small masses, are in a boiling state almost throughout their volume. Therefore, the magnetic bubbles can originate deep within these stars and can be amplified many times before they float to the surface. These stars should have enormous starspots associated with these large magnetic bubbles.
Both these effects are observed. Stars similar to the Sun that are rotating much faster than the Sun are more vigorous X-ray emitters than the Sun, and red dwarf stars are known to be covered with large active starspots.
 This might also explain the observation by several groups that very young stars have X-ray emitting coronas. These stars, which are a million years old or less, are still in a state of slow collapse and are probably rotating rapidly and boiling throughout their volume.
This theory of the origin of X-ray coronas around stars is still in a rudimentary stage. Only the bare bones of a theory exist now; much fleshing out remains to be done. For now, though, the structure looks promising. lt carries with it the exciting prospect of using X-ray observations to study the generation of strong magnetic fields and the interaction of these fields with matter over a far wider range than previously possible.
Finally, the verification of relatively intense X-ray emission from the obscure red dwarf stars may ultimately prove to be a most valuable discovery. Although these stars flare frequently, they are intrinsically much fainter than the Sun. This makes them very difficult to detect with even the most powerful telescopes, so not much is known about their number and distribution throughout the galaxy. In the region of space near our solar system, it is known that they are very numerous. Of the 90 nearest stars that have been classified, 62 of them are red dwarf stars.
Because of their intrinsic faintness, no one has a very good idea as to how many red dwarfs the galaxy as a whole contains. Advanced X-ray facilities should provide a powerful tool for determining just how many red dwarfs are in our galaxy and therefore how massive our galaxy really is.