Quasars and Active Galaxies:
A Point-Form Summary.
This section of the course notes, and the associated PowerPoint presentation, makes the following critical points:
redshifts in astronomical spectra can arise in three ways: motion through space, gravitational redshift, and by virtue of the cosmological expansion. Only the third of these typically gives rise to large redshifts
in the 1960s, early radio observations led to the realization that there were many radio sources in the sky, but the resolution was too poor in those days to allow a precise identification of the sources with their visible counterparts. The techniques of radio interferometry have since solved that problem, but one arely approach was to use a radio telescope to study the way in which a source might 'blink out' as the moon passed in front. This provides information about the location and structre of the source
in this way, several strong radio sources were identified with objects of basically star-like appearance. The optical spectra of these objects could not be readily interpreted, however: they typically contained emission lines that seemed to correspond to no known elements
Schimdt realized that the emission lines were those of hydrogen, enormously red-shifted. The only plausible reason for such a redshift was that the objects, dubbed quasars, must be carried along by the expansion of the universe, and thus be at very large distances
to be seen at all at these distances, the quasars must be tremendously bright - much more luminous than an entire galaxy. Earlier photographs of the sky showed that the quasars also vary in brightness, rather quickly. This means that they must be relatively small, and the puzzle was to explain the source of so much luminosity from so small a volume
various kinds of 'new physics' were invoked, but the passage of time revealed further evidence that the quasars truly are at cosmological distances. We now realize that there are classes of objects intermediate between ordinary gaalxies and the quasars, and that their discovery 'too soon' led to irrelevant speculation about unconventional physics
the current understanding of quasars is that the energy is extracted from a rapidly rotating supermassive black hole in the centre of a large galaxy: stars and gas fall in towards the hole, and a rotating accretion disk is formed. Much of the gas falls into the hole, but a fair fraction of it is 'squirted out' along the rotation axis, in a way that extracts a great deal of energy (much more than would be provided by simply allowing the stars to undergo ordinary thermonuclear reactions)
the stability provided by the spin axis is an essential ingredient in theoretical models of quasars, and is evidenced as well in the structure and size of many double-lobed radio galaxies
quasars provide one additional service in that they act as remote luminous sources, the light of which passes through intergalactic gas at enormous distances from us. Since we are seeing this gas as it was billions of years ago, the study of the absorption lines gives us information about the composition of material in those early stages in the history of the universe
Associated Readings from the Text.Please look at: Chapter 21, pages 664-674.The Interpretation of Redshift.The study of quasars is now 'mainstream' astrophysics, but when they were first discovered there was a genuine mystery about their nature, and provocative suggestions of a need for some brand-new physics and possibly even the overthrow of most of our conventional understanding of basic principles, despite the other successes. Why was this so? To answer that, it is useful first to consider the various ways in which astronomers may encounter redshifts -- that is, shifts of features to longer-than-expected wavelengths in the spectra of the objects they study. There are three different ways in which these can arise: 1 motions through space, a good example of which is provided by the stars near the Sun. Some are approaching us, yielding blueshifts; others are receding, producing redshifts. These velocities are typically small, up to a few hundred kilometers a second -- barely a thousandth of the speed of light, with a correspondingly small spectral shift. 2 gravitational redshifts caused by the loss of energy experienced by light as it climbs away from a strong gravitational field. This effect is strongest near black holes, but no light escapes; for neutron stars, so little light comes from the star itself that the effect is unobservable. The next best target is a dense white dwarf, and indeed observations of the light coming from Sirius B, the white dwarf companion to Sirius, reveals the expected shift -- but the effect is small and generally of little consequence. 3 the expansion of the universe, which has to be recognized as fundamentally different from the motions of stars mentioned above. (The distinction is that the stars, along with us, are found in the gravitationally bound Milky Way, and are moving freely through space. The galaxies are being 'carried' by the expansion of space itself, rather than racing through a pre-existing space.) Depending on the distance, these cosmological redshifts can be very large indeed. Let us quantify that last remark. Imagine a galaxy which is emitting (among other things) photons of blue light. Suppose that galaxy lies at a distance of about 8 billion light years, so that we are seeing it as it was before the universe had reached even half of its present age of about 14 billion years. The universal expansion means that light from that galaxy will reach us with a redshift of about 1: in other words, the wavelength will have shifted from the red to about twice that wavelength. (Blue light has a wavelength of 400 nanometres; red light lies at about 800 nanometres, so the shift is 400 nanometres.) In other words, those blue photons, if present in enough abundance to stimulate the colour receptors in your eyes, would be red in appearance and energy. (Of course, astronomers do not rely on these subjective assessments of colour in measuring redshifts, but instead study the apparent wavelengths of absorption lines in the spectra.) And there are galaxies at even larger distances, with correspondingly larger redshifts, although these are relatively recent discoveries. (Up until the mid-1960s, for example, the largest known redshift for any galaxy was only about 0.5, seen in the spectrum of light coming from a galaxy about 5 billion light years away.)Radio Sources.The next stage of this interesting history dates from the 1960s, during the rapid development of radio astronomical observing techniques and technology. You will remember that the long wavelength of radio waves means that a single 'dish' (telescope) has limited resolution, and indeed the early radio observations could do little more than identify the approximate region of the sky from which the radiation was coming. (The subsequent development of interferometry has changed all that, and radio telescopes working in tandem, like those in the Very Large Array, can pin down the position and map the detailed structure of radio sources in exquisite detail.) As noted, though, the first radio surveys of the sky could do little more than demonstrate that there were some strong sources of radiation 'somewhere out there.' It was not even clear whether the sources were associated with single stars, or entire remote galaxies. What was wanted was a clear-cut identification between a given strong radio source and some object that could be identified on a conventional visible-light photograph so that one could try to work out what the sources were.Enter the Moon.In general, the radio sources are sprinkled at random over the sky. Inevitably, then, the positions of some of them lie close to the ecliptic plane, which means that from time to time there is a pretty good chance that the moon, in its orbit around the Earth, will pass in front of one source or another. This possibility was recognized by astronomers Cyril Hazard and John Bolton, who went to the Parkes radio telescope in Australia to observe the consequence of this happening to a source called 3C273. (This is the 273rd source listed in the Third Cambridge Catalog of radio sources.) There are two important things to be learned from this kind of observation: 1 The way in which the source disappears tells you about its apparent size -- not how big it truly is, but how big it looks to us from the Earth. (A huge galaxy can look quite small if it is sufficiently far away!) The principle is quite simple: as the moon passes in front, the source will vanish in an instant if it is small. If it is more extended in appearance, it will fade away more slowly. 2 More important, the moment of disappearance limits the range of positions where the source can lie: it must be somewhere along the arc defined by the 'leading edge' of the moon as it passes across the background. The moment of reappearance gives you another such arc, and in general you will be able to use this information to narrow down the possible locations to just a couple. (It depends on how extended the source is, and the precision with which you can time the events.) Hazard and Bolton successfully made these observations for 3C273 and several other sources (and then, in rather melodramatic fashion, made independent trips back to Sydney in case the data should be lost in a plane crash! -- that's how important they believed their observations to be). As a consequence, the position of 3C273 was pinned down quite precisely. And what did they see in a standard optical image? Well, their expectation had been that they would see something dramatic, perhaps a pair of galaxies in collision or something equivalently exciting. To their surprise and interest, they found that 3C273 was essentially star-like ('quasi-stellar') in appearance, except for a mysterious faint 'jet' coming off on one side. What on Earth had they discovered?Back to Astrophysics.As you have learned, the deepest insight into an astronomical source comes from a study of its spectrum, which contains much more information than a direct image. With that in mind, astronomers at the Palomar Observatory pointed the Hale 5-metre telescope at 3C273 (and other similar targets) and spread the light out into a spectrum, which they recorded on photographic plates -- the standard technology of the time. The problem was that they could not readily interpret the spectra that they acquired. What they saw was that the object was giving off light in a 'continuum' (that is, light of a range of wavelengths) but that there were prominent emission lines superimposed, rather like that of a gaseous nebula of the sort we see in Orion. But the pattern of emission lines corresponded to no known common elements, and everyone was perplexed. Various imaginative suggestions were made as astronomers grappled their way towards an explanation. Were they seeing the exposed core of some exotic star, perhaps at the endpoint of its life or after a supernova explosion? Were the emission lines coming from excited ionized states of usually rare elements like uranium or some other very heavy species? An explanation was soon to follow, one which had a pleasant simplicity, but it introduced as many new questions as it had solved.Maarten Schmidt's Discovery.A young Dutch astronomer named Maarten Schmidt, then working as a post-doctoral fellow at Palomar Observatory, had a flash of insight one night in 1963: the puzzling emission lines were nothing other than those caused by the ubiquitous element hydrogen! They had resisted identification because they were red-shifted far from their usual rest wavelengths, but once the recognition was in place, the same solution was found to apply to the other such objects. Because of the quasi-stellar nature of their appearance, and the fact that they were first identified as a follow-up to the radio surveys, these objects were dubbed 'quasi-stellar radio sources', or quasars for short. Subsequently, other detection techniques revealed that there are many more again that do not emit large amounts of radio radiation, but the ungainly sobriquet radio-quiet quasi-stellar objects soon gave way to the more tractable term QSOs....and the Problem.The identification of the emission lines led to the happy conclusion that the QSOs were not so mysterious after all -- at least, they were made of the common elements. The sticking point came from a consideration of the large redshifts they exhibited. As I noted above, there are three possible sources of astronomical redshift, but only the cosmological expansion can explain the very large redshifts of the QSOs. (Indeed, the record-holder now is a redshift of about 6!) This might seem unexceptional -- after all, the Cosmological Principle suggests that there should be galaxies out to as far as we can see, and perhaps to infinity. So as our telescopes enlarge in size and power, we might expect to find remote objects with rather large redshifts (corresponding to large cosmological distances and 'look-back' times). The problem, however, is that the newly-discovered quasars were quite bright, and if they lay at a distance of several billion light years, they must be very much brighter than entire galaxies. What could the source of the luminous output be? This problem was unexpectedly made considerably worse when astronomers went back to the great collections of old photographic plates which had been collected over the decades. The QSOs showed up on many such plates, completely unremarked-upon (since they are unexceptional in appearance). The perplexing finding, however, was that for many of them, the images brightened and faded on rather short timescales -- often in less than a year or so. What this means, of course, is that the principal region of luminosity (the 'engine' of the QSO) could only be a light-year or so in size, as compared, say, to our 100,000-light-year-diameter Milky Way galaxy. [You might like to reread the discussion of pulsars to remind yourself of how this argument works. Fundamentally it is a question of light-travel times, but there is a helpful analogy. Suppose you are the recipient of letters from scores of friends and admirers all around the world, all the way from Toronto to London, from Paris to Cairo, from New Delhi to Sydney, Australia. Let us further suppose that each admirer writes and sends a letter a day, and that the letters are delivered to you as airmail on aircraft that travel at constant speed. You can imagine that the letters from Toronto reach you on the day after being posted, those from London and Paris two days later, those from Cairo after three days, those from New Delhi after five days, and those from Sydney after a full week (say). Now suppose that you do something so utterly discreditable that you alienate all those thousands of admirers, and that the news reaches them instantly, over the internet and news broadcasts. As an immediate consequence, all of your now-former admirers decide not to send so much as a single letter to you ever again. What will happen? Well, today's mail will bring letters from all over, as usual. But tomorrow there will be no letters from fans in Toronto, although you will still receive adulatory messages from fans in more remote locations. Five days from now, Toronto, London, and Cairo will have fallen silent, but letters from New Delhi and Sydney will still be arriving - they were, after all, sent before you alienated so many people! After a week, however, you will receive the final messages from Sydney, and that will be it -- nothing more! Notice that this slow falloff is because of the time it takes you to learn (by the cessation of correspondence) that even the Australians have no further use for you, even though they were just as quick as everyone else to reach that decision. In like fashion, a very large astronomical body cannot be seen to 'turn on and off' all at once. Even if it could somehow do so physically, you would not learn about the fading away of the more remote parts of it until some time after the nearer parts were seen to fade away. Roughly speaking, the timescale of variability of an object sets a reasonable limit on its size, expressed in light-travel times. That is the source of the inference that a QSO that varies significantly in brightness on one-year timescales can only be a light-year or so in size.] Now suppose that you do something so utterly discreditable that you alienate all those thousands of admirers, and that the news reaches them instantly, over the internet and news broadcasts. As an immediate consequence, all of your now-former admirers decide not to send so much as a single letter to you ever again. What will happen? Well, today's mail will bring letters from all over, as usual. But tomorrow there will be no letters from fans in Toronto, although you will still receive adulatory messages from fans in more remote locations. Five days from now, Toronto, London, and Cairo will have fallen silent, but letters from New Delhi and Sydney will still be arriving - they were, after all, sent before you alienated so many people! After a week, however, you will receive the final messages from Sydney, and that will be it -- nothing more! Notice that this slow falloff is because of the time it takes you to learn (by the cessation of correspondence) that even the Australians have no further use for you, even though they were just as quick as everyone else to reach that decision. In like fashion, a very large astronomical body cannot be seen to 'turn on and off' all at once. Even if it could somehow do so physically, you would not learn about the fading away of the more remote parts of it until some time after the nearer parts were seen to fade away. Roughly speaking, the timescale of variability of an object sets a reasonable limit on its size, expressed in light-travel times. That is the source of the inference that a QSO that varies significantly in brightness on one-year timescales can only be a light-year or so in size.] What could possibly explain the emission of the light of a thousand ordinary galaxies from so tiny a volume? It is no exaggeration to say that this was seen as an outstandingly important and challenging problem in the astrophysics community in the late 1960s -- coincidentally about the time I began my own career in the field (although I do not carry out quasar research).No Lack of Imagination.Imagination is an important element in science, as much in astronomy as anywhere else. There was certainly no lack of imaginative solutions brought to the topic of the QSOs in the 1960s and 1970s, including suggestions along the following lines: 1 Perhaps the large redshifts of the QSOs was caused by their genuinely high speed through the nearby parts of the universe, rather than a consequence of the cosmological expansion. It could be, for example, that they had been shot out of the nucleus of our own galaxy, like luminous 'bullets', by some unknown mechanism. (If they are nearby instead of billions light years away, they are clearly not as intrinsically bright as the cosmological explanation implies.) One problem with this explanation, however, is that we should expect equivalent behaviour in other galaxies. (Remember the Cosmological Principle, and our feeling that we should not be in any special location, such as membership in the only galaxy that happens to be spitting out QSOs!) But then we would expect to see lots of blue-shifted QSOs near other galaxies -- those that were ejected in our particular direction. No such objects are seen. 2 Perhaps some 'new physics' is needed, either at the margins (when, for example, velocities comparable to the speed of light are encountered) or in some deeply profound way, including a complete overthrow of Newtonian and Einsteinian physics. Needless to say, this would be a disturbing conclusion, given the successes of these explanations in a host of other contexts. It is no exaggeration to say that these matters engendered some controversial discussion and issues, and indeed there was a decade or more of hotly-contested arguments over some often arcane points, typically based on what in retrospect was often fragmentary evidence. Certainly the subject brought into prominence some fascinating characters, many of whom had extremely unconventional views. It was a fascinating time!Some Simple Predictions.The question of the nature of the QSOs has long since been essentially resolved, at least to the satisfaction of all but a small minority of astronomers, but it is interesting to put this in a historical context. Suppose you were convinced, in the 1970s, that the QSOs were indeed at 'cosmological distances'. What might you have predicted? The answers include the following: As our telescopes grow in power, you might expect to discover that the QSOs are not isolated in empty space. On large scales, we see clustering and superclustering, so you might expect eventually to be able to detect faint galaxies in the vicinity of the QSO. Moreover, sufficiently large telescopes might even be able to spread the faint light of these galaxies out into a spectrum, which you might expect to exhibit a redshift close to that of the QSO itself. Indeed, this is borne out. If QSOs are associated with galaxies, they might even be (say) the active cores of massive galaxies. In that case, very high-resolution images (taken perhaps with the Hubble Space Telescope, which was just a dream in the 1970s) might show the 'fuzz' of the parent galaxy surrounding the bright QSO itself. This too is borne out. Since the QSOs are inferred to be very far away, their light presumably passes through the low-density distributed gas between the galaxies. If so, we might expect to seen narrow absorption lines from these clouds of gas, with the QSOs of highest redshift having more such absorption lines than those of lower redshift since the light travels a longer distance, and presumably intercepts more gas clouds. This is also borne out. The interesting point is that all of these lines of evidence support the cosmological hypothesis, but have no straightforward explanation in any other model. As the data came in, therefore, the case became extraordinarily persuasive: the redshift was cosmological; the QSOs were remote, very luminous and rather small (remember the variability); and no 'new physics' was needed.A Growing Understanding.As the years passed, the perplexing nature of the QSOs became less a problem, not simply because of the developments I have just described. There were parallel developments in other areas which led us to realize that, in a sense, we had over-reacted to the questions thrown our way by the QSOs. We had the bad luck -- or the good fortune, if you enjoy the stimulus of perplexing problems that lead down many blind alleys before being resolved! -- to discover the QSOs too soon, before we realized that they were merely the most extreme example of many objects that possess some their attributes on somewhat reduced scales. One example is provided by developments in our ability to map out the structures of the so-called radio galaxies. Many of the radio sources detected in surveys like the Third Cambridge Catalog turn out to coincide with galaxies which are often of quite ordinary appearance. (Not all 3C sources correspond to objects like 3C273!) But further study reveals that the radio radiation is not confined to the optical parts of the galaxy: indeed, there are many 'double-lobed' radio sources which extent vastly beyond the optical confines of the galaxies. (If you had eyes sensitive to radio radiation, you would see these as luminous 'dumbbells' in the sky, many of them larger in appearance than the full moon.) The radiation from these lobes is synchrotron radiation very like that which we met when considering the pulsars, and the fundamental ingredients are the same: fast-moving electrons spiralling around magnetic field lines. (The difference is that the galaxies are not rotating once a second, and we see no pulses!) The big question, however, is how the lobes can be created. Apparently something in the cores of these gaalxies is shooting out high-speed electrons, but the lobes may be many hundreds of thousands of light years away from the galaxy center. Whatever the ejection mechanism is, it has to point steadily in that fixed direction for a very long time indeed. (Imagine using a garden hose to water a pea patch. If you let the head of the hose wander aimlessly, the water will go all over the place.) What can maintain the stability of direction? The obvious candidate is the stability that is provided by spin, by the conservation of angular momentum. The present picture, then, is that the core of a radio galaxy contains a supermassive black hole which is itself spinning and around which there is an rotating accretion disk of material created by the inward fall and tidal disruption of entire stars. Although nothing can come out of the supermassive black hole itself, there are ways in which some of the infalling material gets squirted out along the rotation axis (while much of it falls in). In fact, this turns out to be an amazingly efficient way of extracting energy from the black hole -- much more efficient than the fractional conversion of mass to energy which is entailed in thermonuclear reactions in stars. This very much resolves the energy generation problem in the QSOs. You will, I am sure, remember the presumably relevant observation that some QSOs (like 3C273) are seen to have small 'jets' -- manifestations of the same fundamental behaviour. The energy problem is also reduced by the realization that the light is not broadcast uniformly in all directions: the apparent brightness of a source depends on whether we are looking 'straight down the beam' of a given QSO or seeing it instead at some angle. Given all these developments in our understanding, it is no exaggeration to say that the study of quasars has moved from the mysterious and inexplicable to the mainstream and conventional, all in the space of twenty-five years or so! Previous chapter:Next chapter0: Physics 016: The Course Notes, spring 2005. 1: The Properties of the Sun: 2: What Is The Sun Doing? 3: An Introduction to Thermonuclear Fusion. 4: Probing the Deep Interior of the Sun. 5: The Sun in More Detail. 6: An Introduction to the Stars. 7: Stars and Their Distances: 8: The HR Diagram: 9: Questions Arising from the HR Diagram: 10: The Importance of Binary Stars: 11: Implications from Stellar Masses: 12: Late in the Life of the Sun: 13: The Importance of Star Clusters in Understanding Stellar Evolution: 14: The Chandrasekhar Limit: 15: Supernovae: The Deaths of Massive Stars, 16: Pulsars: 17: Novae: 18: An Introduction to Black Holes: 19: Gravity as Geometry: 20: Finishing Off Black Holes: 21: Star Formation: 22: Dust in the Interstellar Medium: 23: Gas in the ISM: 24: The Size and Shape of Our Galaxy: 25: The Discovery of External Galaxies: 26: Galaxies of All Kinds: 27: The Expanding Universe: 28: Quasars and Active Galaxies: 29: The Hot Big Bang: 30: The Geometry of the Universe: 31: Closing Thoughts: Part 1:Part 2:Part 3: |
(Wednesday, 22 April, 2026.)
