ASTR 103 - Text Supplement

Interstellar Medium

Latest Modification: November 11, 1998

Table of Contents

Interstellar matter is the material lying between stars. Although stars interact with the interstellar medium over the course of their entire lives, we want to divided that interaction into three periods. The first is when stars are born out of interstellar matter which we will discuss at the end of Chapter 17. The second is over that major period of their life when they lose varying amounts of mass to the interstellar medium. Our Sun is undergoing such a loss in the form of the solar wind, which we believe is a process that is not unique to the Sun. We shall inquire into the stellar wind in the first section of this chapter. Finally, the last period is when stars die and is so doing some of them will expel a great portion of their mass out into interstellar space in what astronomers call a supernova outburst. The observational aspects of the expulsion of matter by stars is considered in the second section of this chapter, while the explanation of why the supernova outburst occurs is covered in Chapter 18.

If you exhale your breath once and let it expand into an evacuated cubical enclosure 1 kilometer on a side, the resulting density of your breath will exceed the density in most parts of the interstellar medium. Although this suggests that interstellar space is nearly a vacuum, there is a significant amount of matter lying between the stars because of the vast volume of space. Interstellar matter is primarily a gas, in which hydrogen is the chief component. In regions near very luminous, hot stars the gas is ionized, whereas in other regions it is so cold that molecules exist in it. Thus the interstellar medium is far from uniform in its properties.

Mixed with the interstellar gas is a very fine dust, whose grains are about the size of the particles that are seen as small flashes in a shaft of light coming through a window. Interstellar dust, however, has a very different chemical composition and origin than particles of Earth dust.

Interstellar matter is not uniformly spread throughout the Galaxy but is clumped together in interstellar clouds that vary in size and the complexity of their association in our Galaxy. The stars of our Galaxy--and presumably the stars in all the billions of galaxies in the Universe--are born in interstellar clouds. And when they come to the end of their lives, many stars throw off matter that mixes with the interstellar medium, where it forms new interstellar clouds and finally becomes the matter composing new generations of stars. In summary stars form from interstellar matter, and throughout their lives stars in turn structure and transform the interstellar medium.

Mass Loss by the Sun and Stars

There are many stars that, for one reason or another, lose mass at one or several times during their lives, or possibly even on a continuing basis. Such mass loss is important for its effect on a star's subsequent evolution, which depends on mass. The solar wind is an outgrowth of physical processes in the corona, and it represents a very small amount of mass loss for the Sun (about 10-13 M./y). The evidence that stars are losing matter comes directly from telescopic observations and indirectly from studies of the spectra of stars. For some stars, matter is ejected in one gigantic explosion, such as in supernovae and to a lesser extent novae. Planetary nebulae are another example of a single, but smaller, expulsion of matter by stars. There are also stars for which there is an almost continuous loss of matter, called a stellar wind, over a substantial period of their lives. Stellar winds are probably coupled in part to a range of surface activity, as is the case for the solar wind (Section 15.5). Thus stellar surface activity is the key to understanding a continuous mass loss through stellar winds.

The Solar Wind

The solar corona is a tenuous, spherical halo of very hot gas whose temperature is 1 to 2 million K. These very high temperatures result from the deposition of thermal energy in coronal gases through the dissipation of energy stored in coronal magnetic fields. This high coronal temperature drives a rapid flow of plasma (composed mostly of protons and electrons), which moves out from the base of the corona as the solar wind. The pressure of the gases in the corona must exceed that of the interstellar medium surrounding the Sun, so that the solar corona continuously expands, being replaced by material from the photosphere. Such a wind is said to be thermally driven.

Several radii outside the Sun, the velocity of the solar wind becomes supersonic and continues to increase at least up to 1 AU. By then the solar wind is moving at over 400 km/s, eight times the speed of sound in the gas. Because of the Sun's rotation, magnetic field lines, which confine solar wind particles, spiral outward like water from a rotating sprinkler. Perhaps 600,000 tons of plasma leave the Sun every second, which amounts to about 10-13 of the Sun's mass per year.

The corona is also subject to sudden and very dramatic disruptions of coronal structure, known as coronal transients. Given sufficient cause, such as a solar flare or other type of eruption, what looks like a magnetic bubble or a series of magnetic loops moves outward through the corona with explosive force. These transients are ejecting coronal matter at velocities of thousands of kilometers per second. The amount of matter being ejected has been estimated to be as much as 1016 g, an awesome amount that makes for large variations in the solar wind.

The solar wind rushing out through the Solar System, carrying with it pieces of the Sun`s magnetic field, defines a region about the Sun called the heliosphere. The boundary of the heliosphere occurs where a balance is achieved between the pressure of the solar wind and its magnetic field and the pressure of the interstellar medium and the Galaxy's magnetic field. This outer boundary, which at one time was thought to extend only out to about the orbit of Jupiter or Saturn, apparently lies somewhere beyond the orbits of the planets, as more evidence from active satellites in the process of exiting the Solar System seems to indicate.

Mass Loss by Stellar Winds

Do other stars have thermally driven stellar winds? There is no reason to believe that the Sun is unique in this respect (Table 16.1). The amount of mass carried away by the solar wind is so small that it would take literally trillions of years for the Sun to lose a significant fraction of its total mass. It is therefore unlikely that we can detect such a low rate of mass loss in other late-type, low-mass stars like the Sun.

[Table 16.1]

Red giants and supergiants are stars with very large radii and very low mean densities. Atomic and molecular particles in their atmospheres are not held so tightly by the star's gravity as such particles are in main-sequence stars. More substantial stellar winds can and do occur in red giants and supergiants, but these stellar winds have smaller velocities than the solar wind. Astronomers have actually photographed a faint halo of scattered light from an enormous envelope of gas surrounding the red supergiant Betelgeuse (Figure 16.1); this halo is the result of a stellar wind. In fact, Betelgeuse is losing mass at a rate equivalent to 1M. every million years.

[Figure 16.1]

The recognition that some hot stars of spectral classes O and B continuously lose mass dates from about 1929. Satellite surveys in the ultraviolet part of the spectrum indicate that all stars brighter than bolometric magnitude -6 are losing appreciable amounts of mass through stellar winds. Estimates are that O stars contribute as much as 30 percent of the matter being lost by stars, even though as a class there are only about 10,000 of them in the Galaxy. Their winds may be driven as much by the pressure of the immense numbers of photons they emit (radiation pressure) as by thermal effects (Figure 16.2).

[Figure 16.2]

Explosive Variable Stars

Planetary Nebulae

In contrast to a continuous mass loss in the form of stellar winds is the ejection at one time of a star's surface layers to form a planetary nebula. (The name was given by the eighteenth-century astronomer William Herschel, who noted the resemblance to the disk of a planet; the planetary nebulae certainly have nothing to do with the planets of our Solar System.) In photographs such as Figure 16.3 of the Ring Nebula in Lyra (M57) or of the Owl Nebula in Ursa Major (M97) one can see a small, hot, subluminous central star surrounded by a nebulous shell of ionized gas.

[Figure 16.3]

The shell is expanding slowly outward at speeds of about 30 km/s. The spectrum of its light is an emission spectrum produced by rarefied common gases such as hydrogen, helium, oxygen, neon, and sulfur. The source of energy causing emission from the shell is ultraviolet radiation from the hot central star, whose surface temperature is around 100,000 K. At such temperatures, most of a central star's luminosity is composed of ultraviolet photons, and the luminosity is typically 1000 times that of the Sun, but the star's radius is only a few tenths that of the Sun. Such conditions provide the nebulous shell surrounding the star with sufficient energy to give the gas a kinetic temperature on the order of 10,000 K, a density of several thousand particles per cubic centimeter, a diameter of several tenths of a light year, for a mass that is a few tenths of a solar mass. The degree of ionization in the surrounding nebula is somewhat higher than that in an H II region (discussed in Section 16.3), but otherwise they are very similar. Infrared observations indicate that a great deal of dust accompanies the gases in a planetary nebula's shell. Heated by the absorption of ultraviolet and visible photons, the dust in turn radiates in the infrared.

Planetary nebulae's precursors appear to be red giants of moderate-to-low mass located in the Galactic disk. Many stars may have or will eventually become planetary nebulae, although we have identified only about 1000 in our Galaxy. This phase in stellar evolution is brief--lasting a few tens of thousands of years--reducing our chance of seeing it. We can estimate how long a phase in the life of star is by the relative number of stars that are found in that phase compared to the whole population of stars. An analogy is the problem of how many people are asleep worldwide at any moment. The answer is 1/4 to 1/3 are asleep since the human being is known to sleep from 6 to 8 hours out of each 24 hour period. Turning the argument around, however, if we found 1/4 to 1/3 of the worldwide population asleep at any one moment, then we could conclude that the typical human being sleeps 6 to 8 hours per day. From such reasoning, astronomers estimate that there are actually between 20,000 and 50,000 planetary nebulae in our Galaxy and that a few new ones form each year.


The occurrence of a nova is announced by a rapid rise in a star's brightness, amounting to tens of thousands of times its original brightness in a few hours. This is followed by a slow decline that may persist for a year before the star settles down to its former obscurity. The nova's spectrum shows that matter has been expelled from the star because there is large Doppler shift of the absorption lines toward the blue, indicating a velocity of approach. Soon after the outburst very broad emission lines appear in the spectrum, which indicates that a gaseous shell has been ejected at a high speed, usually a few hundred to several thousand kilometers per second. The broad emission lines result from combining radiation from the front part of the shell, which has a large blueshift, with radiation from the back part, which has a large redshift. The expanding shell has even been seen years later in the case of a few novae, as shown in Figure 16.4. The amount of matter expelled lies somewhere between a few hundred-thousandths and a few tenths of a solar mass. For several novae, radio radiation has been detected, which arises from the thermal energy in their expanding shells of ionized gas. And with the X-ray satellites several novae have also been found to be emitting X-rays. About 30 novae occur in our Galaxy each year; a few even go through recurrent outbursts.

[Figure 16.4]


A supernova is an explosion of a star of such immense proportions that it can be observed in an external galaxy even when the rest of the galaxy cannot be seen. These exploding stars suddenly attain luminosities up to several billion times that of the Sun. As many as five supernova outbursts may occur in our Galaxy each century, according to present estimates; most supernovae in our Galaxy probably escape detection because of heavy obscuration by interstellar dust in the Galactic plane. In other galaxies their occurrence varies from several times a century in the brightest and largest spiral galaxies to one every few centuries in the faintest spirals.

A supernova remnant is the expanding shell of gas resulting from the stellar explosion, such as the Loop Nebula in Cygnus, shown in Figure 16.5. Of the 100 or so supernova remnants that radio astronomers have found in our Galaxy, at least 8 are known X-ray objects and 13 have also been identified optically. Table 16.2 lists some examples of supernova remnants. Identifying old supernova remnants optically is very difficult because the expanding gas shell thins out so rapidly that eventually it blends with the interstellar medium. Therefore, ages for the supernova remnants that are observed in our Galaxy are probably less than 100,000 years.

[Figure 16.5]

[Table 16.2]

One of the priorities of the Einstein Orbiting Observatory was to obtain X-ray images of supernova remnants. (Figure 16.6b shows an X-ray image of the supernova remnant known as the Crab Nebula which is discussed in the adjacent box.) Part of the X-ray emission is synchrotron radiation from high-energy electrons spiraling around magnetic field lines. The other part is the result of the expanding nebulae plowing into interstellar clouds. Somewhat like the sonic boom of jet airplanes, the ejected shell creates a shock front that compresses and pushes interstellar matter ahead of it. This process heats the intermingling gases to temperatures in the millions of degrees, causing them in turn to emit X-rays.

At least two types of supernovae can be discerned. The major difference between them is in their spectra and maximum luminosity, but the general behavior of both is pretty much the same. Type I supernovae have been observed in all types of galaxies, but they occur most often in the disks of spiral galaxies. Their maximum luminosity is about 4 billion times that of the Sun. For Type I supernovae, there is a rapid decline in brightness after maximum luminosity, which is followed by a slowing of the decline with time.

Type II supernovae reach a maximum luminosity of up to 600 million times that of the Sun and exhibit a greater variety of light-curve shapes and spectral changes than do Type I supernovae. They appear most often in the arms of spiral galaxies but rarely in elliptical galaxies.

Although both types of supernovae have very complex and variable spectra that are not yet fully understood, they both show spectroscopic evidence for very high expansion velocities, which are on the order of several tens of thousands of kilometers per second.

How much matter is blown off to return to the interstellar medium? The amounts of mass ejected are not known for certain, but the best estimates suggest that Type I supernovae are among the older population of stars in our Galaxy and lose about a solar mass of material, whereas Type II supernovae are stars belonging to the younger population of stars and may eject more than 5M. (in some cases 50M.) of stellar matter. The question, ever since the supernova phenomenon was first recognized for the extraordinary event that it is, has been "What kind of star has exploded and why?" We will address that question in Chapter 18.

The Crab Nebula and Pulsar

Interstellar Matter

Early in this century astronomers thought that in our Galaxy interstellar space was fairly transparent and any dimming of starlight in general could be ignored. Then in the 1930s astronomers discovered that open clusters contained fewer faint stars and redder stars the farther away the cluster is from us. For that to be a real effect implied something was very strange about our location. The obvious answer is that there must be interstellar matter lying throughout the region between the stars in the plane of our Galaxy. This matter absorbs and scatters starlight thereby diminishing in brightness and reddening in color distant stars compared with nearer ones, which would account for the observations of distant open clusters.

The interstellar matter of our Galaxy, and presumably other galaxies, is a mixture of atomic and molecular gases, mostly hydrogen, along with small solid particles, called grains or dust, concentrated primarily in the plane of the Galaxy. Let us begin by discussing the gaseous component since it is the most abundant part of the interstellar medium.

Interstellar Gas

The gaseous component of the interstellar medium is confined almost entirely to a thin disk in the plane of the Galaxy. In the vicinity of the Sun the disk is only about 1000 ly in thickness. About 90 percent of it by number is hydrogen, of which perhaps half is in molecular form and half in atomic form. Atomic hydrogen occurs in both neutral and ionized forms. Molecular hydrogen and ionized hydrogen are found in only a small fraction of interstellar space compared with the vast volume in which neutral hydrogen atoms are located. And where either hydrogen molecules or hydrogen ions are located, they are the most prevalent form of hydrogen, the other being absent. Because hydrogen is the main ingredient in the interstellar gas, astronomers generally designate a region in which hydrogen is predominantly ionized as an H II region and a region where hydrogen is predominantly neutral as an H I region.

Although most of the mass of interstellar gas is found in interstellar clouds, most of the volume of the interstellar medium consists of warm or hot diffuse gas. Starlight passing through this warm diffuse interstellar gas is selectively absorbed, producing a few absorption lines superimposed on the normal spectra of stars. These interstellar lines can be differentiated from the spectral lines of the O and B stars because they are usually narrow; they are not characteristic of a hot star's photosphere, and they have different Doppler shifts from stellar absorption lines. Interstellar lines are more difficult to identify in stars of later spectral classes, which have many absorption lines. Frequently we see several sets of Doppler-shifted interstellar lines; this means the starlight has passed through several intervening clouds of diffuse gas moving at different speeds along the line of sight, as illustrated in Figure 16.7.

[Figure 16.7]

In the visible part of the electromagnetic spectrum, astronomers have identified absorption lines belonging to such elements as sodium, calcium, and iron and such molecules as cyanogen and methylidine. In ultraviolet spectra, absorption lines for molecular hydrogen, atomic hydrogen, carbon, nitrogen, oxygen, iron, and other elements have been found in data obtained by spacecraft. A surprising find is a large amount of deuterium, the heavy isotope of hydrogen, compared with its low abundance on Earth. Since 1964, in the radio spectrum, discrete lines of hydrogen, helium, and carbon have been observed, lines resulting from electron transitions between energy levels near the ionization limit. For example, a free electron may be captured into level n = 110 of a hydrogen atom, from which it can cascade down to level n = 109 and emit a photon with a wavelength of 6 cm.

Throughout that part of the interstellar medium that has been studied, the abundances of the chemical elements that are heavier than helium are similar generally to what they are in the Sun and other stars. However, since the chemical elements are not spread uniformly throughout the interstellar medium, it is difficult to decide what typical element abundances are. It may be that there are no really typical values. In an H II region nearly 90 percent of the gas is hydrogen. Another way to compare the relative composition is to say that, for every 10,000 hydrogen atoms, there are approximately 1200 helium atoms, 1 or 2 carbon atoms, 1 or 2 nitrogen atoms, 3 or 4 oxygen atoms, 1 neon atom, 1 sulfur atom, and lesser numbers of atoms of heavier elements, particularly iron.

The 21-CM Line

A new means of exploring the interstellar medium and its structure became available to astronomers in 1951, when a spectral line at 21 cm (1420 Mhz) produced by neutral hydrogen was discovered. How are photons formed that have a 21-cm wavelength, which is in the radio region of the electromagnetic spectrum?

As an electron revolves about a proton, it and the proton also spin like tiny rotating tops (Figure 16.8). Once in 11 million years (on average) an electron, if spinning in the same sense as the proton to which it is bound and not disturbed by collisions with other atomic particles, will spontaneously change its spin to the opposite sense. This change drops the atom into a lower-energy state, creating a 21-cm photon, that is one whose wavelength is 21 cm, to carry away the difference in energy. Within an interstellar cloud an electron may actually reverse its spin much sooner, as often as once every several hundred years, during collision with a passing atom. Random collisions between particles in the interstellar medium can also transfer kinetic energy to a bound electron and cause it to flip over and align its spin with that of the proton.

[Figure 16.8]

Even though the time lag for producing a 21-cm photon is inordinately long, a ready supply of 21-cm radiation is always available because of the enormous number of hydrogen atoms along a line of sight through the Galaxy.

The emission of 21-cm photons not only confirms the importance of hydrogen as the primary constituent of the interstellar medium but also provides radio astronomers with a valuable tool for studying the structure of the interstellar medium throughout out Galaxy and neighboring galaxies. Because of its long wavelength, a 21-cm photon can travel greater distances through interstellar space than can photons of visible light. This is so because electromagnetic waves are more likely to interact with bits of matter the closer their wavelengths are in size to the characteristic size of the matter. Thus electromagnetic waves with visible wavelengths more readily interact with atoms, molecules, or very small solid particles than do waves with very long radio wavelengths.

[Biography - Karl Guthe Jansky]

Interstellar Molecules

Since 1963 radio astronomers have found a surprising number of interstellar molecules, including many organic ones (those containing carbon), by searching for their spectral fingerprints, which are emission lines that occur in the centimeter and millimeter regions of the electromagnetic spectrum. From approximately 150 radio spectral lines some 50 or more molecules (Table 16.2)--containing mostly combinations of hydrogen, carbon, nitrogen, and oxygen--have been identified; the number of different molecules discovered is increasing yearly. Some of the newly discovered molecules are familiar inorganic compounds, such as ammonia, water, and several containing sulfur.

[Table 16.2]

A number of interesting organic molecules have also been found, such as formaldehyde, methyl alcohol, and ethyl alcohol. Enough have been found of these interstellar molecules that contain a reasonably complicated arrangement and number of atoms to suggest that, however they are produced, the process is quite capable of forming rather complex molecules. Some of the organic interstellar molecules have not yet been produced in a chemistry laboratory, so that study of the interstellar medium is adding a new dimension to organic chemistry.

Compared with hydrogen, the amounts of other molecules that occur in interstellar space are small--less than one-thousandth that of hydrogen. A few of their spectral lines are observed as absorption lines instead of emission lines whenever enough molecules lie along the line of sight toward a Galactic or extragalactic radio source that emits continuous radiation. Molecules are primarily found in dark cloud complexes such as those in the constellations Orion, Taurus, and Ophiuchus. Other locations for molecules are distributed across the Galaxy in localized regions containing interstellar clouds. Some are even concentrated in tiny high-density regions comparable in size to the Solar System. In addition to being found in our Galaxy, the hydroxyl radical, water, formaldehyde, hydrogen cyanide, ammonia, and carbon monoxide have also been detected in several nearby galaxies. Thus their presence in the interstellar medium of our Galaxy is not a unique event, but probably represents a common feature of most galaxies. When we discuss the structure of our Galaxy in Chapter 20, we shall have more to say about the locations of molecules.

How these molecules were formed is not well understood. A two-atom collision can produce a diatomic molecule, but it is very difficult to imagine a sequence of successive collisions that can produce a polyatomic molecule with as many as 13 atoms. But before we discuss a possible formation mechanism for the molecules, we should introduce interstellar dust.

Interstellar Dust

Interstellar dust consists of solid grains of microscopic size whose composition and properties are very unlike most types of dust on Earth. Photographs of regions along the Milky Way are laced with dark patches that are large clouds containing dust as well as gas. The dimming of starlight is caused almost entirely by interstellar dust, for the gaseous component of interstellar matter is reasonably transparent to starlight. In fact, the interstellar gas is billions of times more transparent to visible light than is air at sea level on Earth. Clinging close to the Galactic plane, within a few hundred light years, interstellar dust completely shuts off our view of the Galactic center in visible wavelengths, and it keeps us from seeing extragalactic objects whose direction is along the Galactic plane. In other spiral galaxies seen edgewise this dust is the dark lane that passes centrally across the galaxy (Figure 20.5).

Dimming by interstellar dust is greatest for ultraviolet light, less for visible light, and least in infrared wavelengths. For visible light, the loss can be as much as 0.7 magnitude per 1000 ly (the average is about half this value) near the Galactic plane. This means that for a star at the center of the Galaxy, about 30,000 ly away, only about 1 photon out of every 100 billion photons reaches us. If we do not correct the observed apparent magnitude of a distant star for this loss of light, its distance calculated from the distance modulus is too large. In the hard X-ray, infrared, and radio spectral regions, however, we can observe all the way to the Galactic center, since the dust is transparent to these wavelengths. Because blue light is affected twice as much as red light by interstellar dust, light from a distant star not only looks dimmer but is also redder than it should be for the spectral type of the star (Figure 16.9). Astronomers refer to this effect as interstellar reddening. Because of this effect, color indices measured for distant stars are in error and must be corrected before they can be used as a measure of the star's temperature.

[Figure 16.9]

About 1 percent of the mass of interstellar clouds is due to dust and 99 percent is due to gas. The average density of the dust is about one grain per 1013 cm3, or one grain in a cube 200 m on each side. This is a very low density when compared with the typical interstellar gas density of one atom per cubic centimeter. The density of the dust grains can be much larger in small, localized regions, such as the heart of an interstellar cloud. But so will the density of gas also be larger, and it appears that the ratio of dust to gas in most of the interstellar medium is constant to within a factor of about 2.

What is the size and composition of dust grains? Because the scattering of photons is strongly dependent on the size of the scattering particle, the strong scattering of visible light by interstellar grains suggests that a grain is comparable in size to the wavelength of light, say, a few hundred-thousandths of a centimeter or smaller. At this size the typical grain should contain about 100 million atoms with most of them being elements heavier than hydrogen and helium. Analyses of data from the Infrared Astronomical Satellite suggests that the diffuse dust grains lying in between interstellar clouds may actually be smaller than the dust grains in clouds. From the reddening, dimming, and polarizing of starlight, astronomers conclude that dust grains, whether large or small, are probably assorted graphite, iron particles, silicon carbide, silicates, and ices; the exact composition is still uncertain.

It seems possible that most of the interstellar dust comes from material that is being blown out of stellar atmospheres. There are several phases in a star's life, starting from birth and ending with death, when the star can lose matter. As an example, among the brightest sources of infrared radiation are the glowing dust shells around some stars, called circumstellar shells (Figure 16.10). Apparently, the grains intercept short-wavelength radiation from the central star, heat up, and reradiate the energy as long-wavelength infrared photons.

[Figure 16.10]

Interstellar dust grains may be the means of forming interstellar molecules. It is thought that hydrogen and other types of atoms can accrete on the cold surfaces of the grains, where they bond together to form molecules. These molecules can escape from the grain surface by absorbing a low-energy photon of starlight (or by some other means). Apparently, the enveloping dust cloud prevents ultraviolet starlight or other energetic photons from reaching the interstellar molecules deep inside interstellar clouds and dissociating them.

16.4. Interstellar Clouds

Diffuse and Dark Clouds

In photographs of the Milky Way, our view of the starry background is partly or wholly blocked by dark interstellar clouds, sometimes called dark nebulae. They contain denser concentrations of interstellar dust than occur generally in the Galactic plane. One such dark region is a long, chainlike complex composed of dozens of isolated and connected dark interstellar clouds that stretches about halfway around the Milky Way from the constellation Cygnus to Crux. This obscuring strip forms the Great Rift dividing the Milky Way into two branches, as can be seen in Figure 14.1. In many regions along its length this dark nebulosity separates into tangled lanes of absorbing material that partially cover bright, glowing, gaseous nebulae (Figure 16.11).

[Figure 16.11]

Even though ground-based observations have provided us with important information about the properties of interstellar clouds, much of our understanding of them has come from ultraviolet studies with the Copernicus and IUE satellites. We find that the clouds can be divided into diffuse clouds, which are thin enough for us to observe stars behind them, and dark clouds, which are so opaque that stars behind them cannot be seen. Some of the dark clouds are of immense extent and are the locations of many different types of molecules; these are known to astronomers as giant molecular clouds. The intercloud region (between clouds) contains a high-temperature, low-density gas, much of it ionized hydrogen, and wispy-structured dust regions.

For astronomers it is still not clear in all the examples under study where individual clouds leave off and groups of clouds (or cloud complexes) begin. This probably accounts for some of the wide variations in properties quoted for interstellar clouds. Both types of clouds are irregularly shaped and are from 0.1 to 50 ly in diameter. Giant molecular clouds can be as much as several hundred light years across. Their temperatures go from about 100 K for diffuse clouds down to 10 to 20 K for dark clouds. Interstellar clouds may take up as much as 4 percent of the space in the Galactic plane, with typical masses of several solar masses up to 104 M. for diffuse clouds and up to 5 x 105 M. for giant molecular clouds. Their densities, which vary from 100 particles per cubic centimeter for diffuse clouds to more than 1 million particles per cubic centimeter for dark clouds, are low compared with the 1019 molecules per cubic centimeter in the air we breathe. Even so, dark clouds can be remarkably opaque because of the accumulative effect of extinction by interstellar dust as starlight traverses their enormous lengths.

Typical separations between clouds appear to be on the order of hundreds of light years. The total number of giant molecular clouds in our Galaxy may run up to several thousand, representing a couple of billion solar masses of interstellar matter. The largest single concentration of giant molecular clouds is a ring of them, lying some 15,000 ly from the center of the Galaxy (Figure 16.12). It has been suggested that this ring may contain as much as 90 percent of all the interstellar matter in our Galaxy.

[Figure 16.12]

Obscuring Effect of Interstellar Clouds

Interstellar dust, concentrated in interstellar clouds in the plane of our Galaxy, obscures distant portions of the disk from our view (Figure 14.1). The center of the Galaxy in the direction toward Sagittarius is completely hidden in visible wavelengths. Light from distant galaxies that lie in a direction vertical to the Galactic plane is dimmed by 40 percent. The dimming becomes even greater in directions farther from the vertical until it is nearly total along the plane of the Galaxy, especially toward the Galactic center. Because interstellar dust tends to concentrate in clouds, the dimming of light by it is not uniform but depends on our line of sight through the scattered interstellar clouds (Figure 16.13). In addition to dimming the light passing through it, interstellar matter also reddens the light by scattering blue-wavelength photons more proficiently than it does red-wavelength photons.

[Figure 16.13]

Carbon Monoxide in Dark Clouds

In dark interstellar clouds, hydrogen is primarily in the form of molecules rather than atoms, so that clouds are not sources of 21-cm radiation. And unfortunately, molecular hydrogen has no spectral features in the visible or near-infrared part of the spectrum. With the discovery about 15 years ago of strong emission in the radio spectrum due to carbon monoxide, radio astronomers have acquired a marker of molecular hydrogen's location and a new probe for investigating dark clouds, as suggested by Figure 16.14. Carbon monoxide can serve as a marker since the conditions that permit it to exist are also suitable for the existence of molecular hydrogen. Dark clouds are the primary locations for interstellar molecules, and the CO molecule is much more abundant in them than in the interstellar medium generally. For every 10,000 hydrogen molecules in clouds, there is approximately 1 CO molecule. Not only can astronomers determine the motions of the molecules from Doppler shifts, but they can also infer densities and temperatures.

[Figure 16.14]

Interstellar Masers, Dark Clouds with Energy Sources

Some of the dark interstellar clouds are detectable from their continuous emission of radio waves and infrared radiation. These clouds obviously have sources of energy within them. In 1965, astronomers accidentally found microwave emissions produced by hydroxyl radicals coming from dark clouds. The character of the emissions was peculiar, and the region from which they came was very small. They found that these small regions were also bright sources of infrared radiation, but that they emit virtually no visible light.

The radio emissions were much stronger than could be accounted for by random thermal collisions. Clearly there was a mechanism that selectively excited the OH molecule. It is thought that infrared radiation from nearby stars excites OH molecules and they are stimulated to deexcite by interaction with stellar photons having the right wavelength. The emitted radiation in turn stimulates other molecules to radiate in the same fashion, producing an avalanche of emissions. Thus the radiation in a normally weak line is greatly amplified. The word maser used the describe this phenomenon is an acronym for microwave amplification by stimulated emission of radiation.

Astronomers know of several hundred OH masers and several dozen H2O masers operating in dark interstellar clouds. They are also found in the atmospheres of red giants that are variable stars. In general, masers in molecular clouds are brighter than those in luminous red stars, but those associated with stars seem to be more numerous. Our interest here, however, is the significance of the presence of masers in clouds where astronomers believe stars are forming. Clearly the masers in clouds signify that energetic events are occurring at specific points in molecular clouds. Such events most likely constitute star formation.

16.5. Emission Nebulae

H II Regions

The emission of ultraviolet photons by O and B stars is so great that even far from these hot stars the number of photons is sufficient to ionize hydrogen gas in interstellar clouds. With the ionization of hydrogen, the H I region becomes an H II region, or an emission nebula (Figure 16.15). Stars of spectral type O5 emit enough ultraviolet photons to ionize hydrogen out to distances of 300 ly from the star. For cooler spectral types the surrounding H II region is smaller; an A0 star creates an ionized region about it that is less than 1 ly in radius. Emission nebulae are among the most beautiful of all astronomical objects as the color pictures in Figure 16.15 clearly show. Table 16.2 lists some of the properties of a few emission nebulae.

[Figure 16.15]

[Figure 16.16]

[Table 16.2]

In panoramic photographs of the plane of the Milky Way such as Figure 14.1, one sees many bright, glowing regions whose spectrum is an emission spectrum. The Balmer alpha line of hydrogen is responsible for the vivid red color of many H II regions. These H II regions are produced by hot stars and are associated with interstellar clouds either by being surrounded by them or by being on the edge of a cloud complex.

H II regions occur in about six distinct categories, depending on their size and the density of free electrons resulting from the ionization of hydrogen. Astronomers refer to the smallest as ultracompact H II regions and the largest as supergiant H II regions. The smallest ones are from a few tenths to a few tens of light years in diameter and their masses range from a few tenths to a few solar masses. These smaller H II regions are generally buried in dark molecular clouds so that in the visible part of the spectrum they are almost totally obscured from view or are heavily reddened if at all visible (the larger ones generally can be seen).

In the three categories of large H II regions, the regions are from a few to 1000 ly in diameter, and they contain anywhere from tens to hundreds of millions of solar masses of ionized matter. These three categories are the types of emission nebulae that are seen most readily, with the supergiant H II regions being by far the brightest objects in the spiral arms of our Galaxy and other spiral galaxies. They occur sometimes in groups and sometimes isolated from each other. As for the shapes of these H II regions, they range from readily definable shapes to large, complex, ill-defined regions. Altogether it is estimated that about 1 percent of the mass of our Galaxy is tied up in the form of H II emission nebulae.

Carina and other gaseous nebulae are suffused with an X-ray glow resulting from many supernova outbursts. In the constellation Cygnus lying about 7000 ly from the Sun (beyond the bright star Deneb) and partially hidden behind the dark interstellar cloud complex known as the Great Rift is a rarefaction in the interstellar medium known as a superbubble. It is about 1000 ly in diameter and contains gas at temperatures of about 2 million K. It appears that this superbubble was created by a chain of supernova explosions and possibly amplified by stellar winds occurring within the last 3 million years. Such bubbles, surrounding many stellar associations of massive stars, occupy at least as much as 10 percent, if not more, of the entire Galactic disk and thus are important components of the Galaxy.

What Type of Interstellar Medium Surrounds the Sun?

Before leaving a discussion of the interstellar medium, we should ask about the nature of the interstellar matter that surrounds our Solar System. From what we have said about giant molecular clouds, the Sun is obviously not sitting in the middle of one of them. Observations with the ultraviolet satellites place the Sun in the low-density (about 0.1 particle per cubic centimeter) and high-temperature gas of the intercloud region. Also our Solar System seems to located on the edge of a "hole" or a bubble (smaller than a superbubble) in the interstellar medium that may well be the result of a supernova outburst. Although the Sun in its motion relative to the stars of the solar neighborhood could encounter a dense cloud (greater than 100 particles per cubic centimeter), it is not likely to happen soon.

Having surveyed stars and the interstellar medium, we should try to bring this data into focus, tying it all together in the life story of stars. Such is the subject of the next three chapters.

Copyright 1995 J. C. Evans
Physics & Astronomy Department, George Mason University
Maintained by J. C. Evans;