1 00:00:50,610 --> 00:00:55,410 Welcome and thank you all for coming. 2 00:00:56,220 --> 00:00:59,610 In some cases, considerable distances. 3 00:00:59,610 --> 00:01:05,160 It's heartwarming to see so many friends and colleagues, both old and new. 4 00:01:07,530 --> 00:01:12,180 So if we could dim the lights so I can try that. 5 00:01:13,740 --> 00:01:19,860 Okay, very good. So I'm going to talk today about accretion processes in the universe. 6 00:01:20,970 --> 00:01:24,730 I think probably some of you will be hearing about the subject for the first time. 7 00:01:24,750 --> 00:01:32,670 I know that several of you in the audience are among the world's most distinguished researchers in the topic. 8 00:01:33,180 --> 00:01:36,420 So I hope you'll bear with me and indulge a little bit. 9 00:01:39,970 --> 00:01:43,780 I hope you can. So. 10 00:01:45,950 --> 00:01:49,850 Let's begin with the big picture. All structure in the universe. 11 00:01:50,330 --> 00:01:58,070 Stars, galaxies and clusters of galaxies formed as a consequence of accretion. 12 00:01:58,550 --> 00:01:59,420 What is accretion? 13 00:01:59,420 --> 00:02:14,330 Accretion simply refers to the fact or to the process of a dilute gas following on to being concentrated in some kind of a central object. 14 00:02:15,950 --> 00:02:20,780 Stars and planets form in gaseous cocoons and interstellar matter. 15 00:02:21,800 --> 00:02:26,120 Gas streams flow from one star to another in binary systems. 16 00:02:26,270 --> 00:02:31,490 A different type of accretion process and huge collapsed stars. 17 00:02:32,750 --> 00:02:40,130 Black holes, in fact, sit in the centre of galaxies, drawing gas continuously from their surroundings. 18 00:02:43,700 --> 00:02:50,780 Now it must be understood that accretion itself is really all but inevitable. 19 00:02:51,560 --> 00:03:00,020 It can't be stopped. The gas that is accreting is not being blown away by some other accretion event. 20 00:03:01,190 --> 00:03:12,830 Eventually it must collapse to either solids or ultracompact objects known as white dwarfs. 21 00:03:14,000 --> 00:03:22,700 The end states of most stars, which holds itself together with a pressure force similar to those that hold up 22 00:03:22,700 --> 00:03:27,920 atoms and white dwarf star in a sense as a great big atom or a neutron star, 23 00:03:28,760 --> 00:03:32,450 which is like a great big atomic nucleus. I say great big. 24 00:03:32,930 --> 00:03:39,110 But of course, the neutron star itself would fit comfortably within the ring road around Oxford. 25 00:03:39,860 --> 00:03:46,010 You have the mass of the sun. Sometimes it seems a bit like that, even under ordinary circumstances. 26 00:03:47,120 --> 00:03:55,250 Within an area which, as I say, could be comfortably encircled by the ring road or. 27 00:03:57,020 --> 00:04:00,890 The matter could collapse down to the ultimate singularity. 28 00:04:01,610 --> 00:04:05,390 A black hole puncturing spacetime itself. 29 00:04:07,670 --> 00:04:19,900 Now. Allow me to set the tone of the lecture in these early stages by invoking the spirit of Monty Python. 30 00:04:21,630 --> 00:04:38,250 Some of you are of a certain age or sensibility may recover or may remember the Royal Society for putting things on top of other things, 31 00:04:38,760 --> 00:04:42,930 which is really not a bad description of what accretion theorists do. 32 00:04:44,460 --> 00:04:49,470 The sheer numbers have put more things on top of other things than ever before. 33 00:04:49,800 --> 00:04:55,590 This was a good year, so that's a good way to think of astrophysical accretion. 34 00:04:56,730 --> 00:05:02,100 That's what theorists do. They want to figure out how to put things on top of other things now. 35 00:05:02,880 --> 00:05:16,740 In fact, the notion of. Particularly the large scale structure in the universe forming from accretion does have a somewhat more noble line of descent. 36 00:05:17,340 --> 00:05:21,420 There's a very famous quotation from Isaac Newton and later to Richard Bentley. 37 00:05:22,440 --> 00:05:26,710 He writes, If matter were evenly disposed throughout an infinite space. 38 00:05:26,760 --> 00:05:35,310 Newton is trying to figure out now, just by pure thought, no less than the origin of the stars in the universe. 39 00:05:37,390 --> 00:05:44,740 If matter were evenly disposed throughout an infinite space, it could never convene into one mass, but some of it would convene into one mass, 40 00:05:45,010 --> 00:05:52,790 some into another, so as to make an infinite number of great masses scattered great distances from one to another. 41 00:05:52,810 --> 00:05:57,800 Throughout all that infinite space. That's the basic idea. 42 00:05:57,850 --> 00:06:00,850 The pictures we know somewhat more complicated. 43 00:06:02,140 --> 00:06:12,280 And in particular, if we want to understand the first major accretion process, how did the large scale structure of the universe form? 44 00:06:13,270 --> 00:06:20,380 We need to take into account the fact that the universe is expanding. 45 00:06:21,830 --> 00:06:25,490 It began small. It's growing large. 46 00:06:26,300 --> 00:06:31,880 The birth of the universe is in some sense the ultimate anti accretion event. 47 00:06:32,630 --> 00:06:38,420 It's explosive release, which is going on to this day. 48 00:06:38,420 --> 00:06:42,020 Indeed, as we know now, it continues to accelerate. 49 00:06:43,280 --> 00:06:47,510 And within that structure, we need to somehow turn around and re collapse. 50 00:06:51,400 --> 00:06:55,990 This is a difficult problem to form something by self gravity. 51 00:06:57,840 --> 00:07:00,210 We have to start off with a little over density. 52 00:07:00,510 --> 00:07:09,340 It must first before it does anything else, stop expanding with the rest of the universe and only then turn around and collapse back on itself. 53 00:07:09,360 --> 00:07:15,310 This is not so easy. And either a physical. Or a mathematical sense. 54 00:07:16,420 --> 00:07:23,049 The first calculations of how to do this and how small over densities would behave in an expanding 55 00:07:23,050 --> 00:07:30,610 background were done by the Russian physicist Ian Lipschitz using the full machinery of general relativity. 56 00:07:30,940 --> 00:07:37,840 And I've put up the equations there to give you some sense of the complexity of the problem. 57 00:07:37,900 --> 00:07:47,590 If one wants to do it rigorously now, fortunately, it is easy to state the fundamental result in very simple terms. 58 00:07:48,610 --> 00:07:52,660 If the universe grows by a factor of, let's say, F. 59 00:07:54,120 --> 00:08:00,780 The density must go down by a factor of one over f cubed, its ordinary matter being conserved. 60 00:08:02,040 --> 00:08:09,300 Now an embedded growing disturbance and over density does not initially in fact grow. 61 00:08:10,620 --> 00:08:15,540 It too goes down, but only by a factor of one over f squared. 62 00:08:16,380 --> 00:08:21,870 So it's actually decreasing, but relative to the background increasing. 63 00:08:23,550 --> 00:08:32,130 And it's that behaviour which has to turn itself into something more dramatic if we're to get to where we are today. 64 00:08:33,740 --> 00:08:37,700 So here is a representation of the expanding universe. 65 00:08:39,990 --> 00:08:47,700 And what we want to do is to start off of the region slightly over dance and then let it self accrete. 66 00:08:48,630 --> 00:09:00,390 So as the universe is expanding, the over density has to turn around and accrete if we're going to get the ball rolling. 67 00:09:01,970 --> 00:09:06,830 Now these days, people have done rather sophisticated simulations. 68 00:09:07,370 --> 00:09:12,230 This is the Greco simulation from astronomers in Chile. 69 00:09:13,370 --> 00:09:18,470 It is a simulation which involves 134 million particles. 70 00:09:18,980 --> 00:09:21,440 And in what we'll be looking at here, 71 00:09:21,440 --> 00:09:30,170 it's essentially the growth of this kind of accretion structure with the background expansion taken out of the calculation. 72 00:09:30,770 --> 00:09:38,600 So you're watching the formation of structure relative to this expanding background, and you can see something interesting. 73 00:09:40,650 --> 00:09:46,380 It doesn't tend to form little blobs, nice little spheres, but it. 74 00:09:47,920 --> 00:09:52,060 Forms these elongated filaments and sheets. 75 00:09:52,870 --> 00:09:59,229 And that's really what the large scale structure of the universe looks like when observers go 76 00:09:59,230 --> 00:10:07,450 out and actually look at how the galaxies themselves and the clusters of galaxies are disposed. 77 00:10:08,050 --> 00:10:14,290 You don't see a nice little spherical collapse. Generally, one axis prevails and things tend to be flat. 78 00:10:14,320 --> 00:10:17,860 Gravity makes things flat in this context, not round. 79 00:10:19,240 --> 00:10:25,960 Well, we have a long way to go. We have some idea of how things got get started. 80 00:10:28,040 --> 00:10:31,940 But if we want to move on, we're trying to make the whole universe. 81 00:10:33,800 --> 00:10:37,880 We have to get the galaxies. We have to get the stars. We have to get the planets. 82 00:10:40,070 --> 00:10:47,150 If we want to detach from the expanding universe within a galaxy, there are two principal obstacles that have to be overcome. 83 00:10:48,140 --> 00:10:54,740 One, and I think these are self-evident. If you try to squeeze something, it doesn't like to be squeezed. 84 00:10:54,740 --> 00:11:00,530 There will be an opposing pressure due to the contraction process itself that will continue to grow. 85 00:11:02,640 --> 00:11:13,980 The other very important dynamical consideration is that if there is any angular momentum in the object to begin with when it contracts, 86 00:11:13,980 --> 00:11:19,260 the conservation of angular momentum will cause it to spin more and more rapidly. 87 00:11:19,650 --> 00:11:27,600 And there's an opposing force that results from this which counter effects and eventually prevails over gravity. 88 00:11:28,860 --> 00:11:36,600 Let's take the first one first. Let's consider situations where we have some degree of symmetry in the problem. 89 00:11:39,490 --> 00:11:44,379 Exactly how does the pressure depend upon the density? So P Here is the pressure rho. 90 00:11:44,380 --> 00:11:50,760 The Greek letter Rho stands for the mass density. Proportional to Roe? 91 00:11:51,750 --> 00:11:54,780 No, that's not the relationship. 92 00:11:54,810 --> 00:12:05,010 This would be true for a gas where the temperature remain fixed, but the contraction process itself heats the gas be proportional to ROE squared. 93 00:12:05,250 --> 00:12:09,390 That's a bit too stiff. The correct answer is actually something in between. 94 00:12:10,710 --> 00:12:15,510 So if we want to actually understand what the build-up of the pressure is opposing the accretion, 95 00:12:17,070 --> 00:12:24,150 it helps to think of the molecules in the gas a little bit like a ball bouncing between two walls. 96 00:12:24,720 --> 00:12:30,360 The physics of it is actually very, very similar. So if we imagine a ball. 97 00:12:33,320 --> 00:12:40,760 Moving at velocity V into walls contracting from either side with velocity W each 98 00:12:40,760 --> 00:12:46,700 time the wall strikes the I should say the ball strikes the wall it picks up. 99 00:12:47,290 --> 00:12:51,300 If you can show this, this is a problem we would ask students to do. 100 00:12:51,320 --> 00:13:04,220 The ball picks up a velocity two times, w twice the walls velocity and it comes back, picks up another two to W and so on as the walls contract. 101 00:13:04,640 --> 00:13:15,230 And what you find when you do the calculation is that the product of the separation of the walls l times, the velocity V remains are constant. 102 00:13:16,370 --> 00:13:23,030 And that's the fundamental process that builds up the pressure in a contracting ball of the gas. 103 00:13:23,330 --> 00:13:30,370 I actually have an animation of it. To give you a physical sense. 104 00:13:32,140 --> 00:13:37,180 On how this builds up. So you see the wall bouncing back and forth and boom, boom, boom, bah bah bah bah bah. 105 00:13:38,620 --> 00:13:42,400 So you get a sense that, in fact, you're doing a lot of heating in this process. 106 00:13:43,600 --> 00:13:48,160 And that makes the gas rather stiff. The gas doesn't feel too stiff. 107 00:13:48,940 --> 00:13:53,650 But if you're going to try to contract it uniformly, it's actually extremely stiff. 108 00:13:55,660 --> 00:14:03,790 So how does P depend upon Roe? Well, if we take three simple equations, if we write down that P is Roe v squared, 109 00:14:04,510 --> 00:14:10,570 simply stating that the pressure is the rate of momentum transfer per unit area against the wall. 110 00:14:11,410 --> 00:14:16,209 If we write V times L is a constant which we just found out. 111 00:14:16,210 --> 00:14:19,840 If we write down that mass has conserved the product of the density times, 112 00:14:19,840 --> 00:14:26,169 l cubed is a conserved mass and you can instantly eliminate the velocity and 113 00:14:26,170 --> 00:14:33,190 you find that the pressure is proportional to the density rho the 5/3 power. 114 00:14:33,220 --> 00:14:37,090 That's an interesting kind of an unusual result. It's a funny exponent. 115 00:14:38,550 --> 00:14:42,820 It may not feel that way, but it's a quite the gases are really quite stiff. 116 00:14:44,050 --> 00:14:51,450 Now, interestingly enough. If I tried to do this in relativity, I can go through exactly the same kind of argument. 117 00:14:51,460 --> 00:14:55,560 I have to be a little careful to use momentum instead of velocity, 118 00:14:56,310 --> 00:15:03,570 and it helps to work with the number density instead of the mass density, but it's still basically the same kind of calculation. 119 00:15:04,020 --> 00:15:14,190 What one finds at the end of the day is that the pressure is proportional, not to the 5/3 power of the density, 120 00:15:14,190 --> 00:15:19,920 but the 4/3, which seems like a significant, significant technical difference. 121 00:15:20,490 --> 00:15:32,280 But as most astrophysicists, I should say, I hope all astrophysicists know, it actually has extremely profound consequences, as we will see. 122 00:15:34,190 --> 00:15:37,250 Summarise. How does P depend upon Roe? 123 00:15:37,490 --> 00:15:48,200 If I try to make gas accrete without loss of energy, either the pressure is proportional to density for the 5/3 for an ordinary gas, 124 00:15:48,650 --> 00:15:53,690 or to the 4/3 if the particles are moving near the speed of light in relativity. 125 00:15:54,800 --> 00:16:00,800 So armed with this knowledge, we can really start to think about how to put things on top of other things. 126 00:16:01,520 --> 00:16:05,210 We can start to try to make matter in the universe. Accrete. 127 00:16:07,410 --> 00:16:13,860 Well, it seems to me there are two types of interesting spherical accretion processes. 128 00:16:14,160 --> 00:16:18,030 Processes where we have a high degree of symmetry. 129 00:16:18,030 --> 00:16:25,770 Nothing depends upon direction. Everything is just moving radially, accreting on to some kind of a central object. 130 00:16:26,250 --> 00:16:31,350 One is known as bond, the accretion. I won't talk too much about bond the accretion in this lecture. 131 00:16:32,070 --> 00:16:37,170 It's kind of a technical subject. It's an interesting problem from a mathematical point of view, 132 00:16:37,530 --> 00:16:42,840 probably a little bit too idealised, I think, to be of profound astrophysical significance. 133 00:16:43,860 --> 00:16:51,090 So the second type of accretion process, which I will discuss, may not strike you as an accretion process at all. 134 00:16:52,830 --> 00:17:03,900 But I would like to think of stars. Ordinary stars like the Sun as a sort of extended accretion event, because in fact, 135 00:17:04,650 --> 00:17:10,920 the sun, even as we speak, is in fact at least the core of the sun. 136 00:17:10,950 --> 00:17:17,760 I have to make that distinction. It's actually getting smaller and smaller with time at a very slow rate. 137 00:17:18,690 --> 00:17:23,850 But there's no question that the core of the sun is contracting, which comes. 138 00:17:25,430 --> 00:17:31,280 And it's quite interesting to think of stars in this way as kind of these extended accretion events. 139 00:17:32,930 --> 00:17:38,450 So to understand what's going on, it's imagining we have a gigantic ball of gas. 140 00:17:39,440 --> 00:17:45,340 If we make it too big, I take the same amount of mass and spread it over to bigger volume. 141 00:17:45,770 --> 00:17:50,240 It'll gather itself back up by its own self gravity, but only to a certain point. 142 00:17:50,840 --> 00:17:57,139 Because if I now take the gas and try to squeeze it, we know what's going to happen. 143 00:17:57,140 --> 00:18:00,920 The density will excuse me. The pressure will rise. 144 00:18:00,920 --> 00:18:13,330 Like the density to the 5/3 power. And ultimately the pressure will push back harder and the gravity can attract. 145 00:18:13,900 --> 00:18:18,550 In effect, the pressure force is going like one over the size cubed. 146 00:18:19,060 --> 00:18:27,030 The gravitational force no more than one over R squared. So if it's not too big, not too small. 147 00:18:27,030 --> 00:18:32,160 I have a just right point on the just right point, Goldilocks point. 148 00:18:33,350 --> 00:18:44,020 Is the far. Now, a very interesting question is what happens if we squeeze so hard and those molecules are like that 149 00:18:44,200 --> 00:18:49,660 ball hitting the wall that we just saw and it's going so fast that it's approaching the speed of light. 150 00:18:49,930 --> 00:18:56,260 Well, we know what happens then when the pressure is proportional to the density, to the 4/3 power. 151 00:18:57,160 --> 00:19:00,170 That's not an idle thought. 152 00:19:00,190 --> 00:19:04,690 Exercise actually happens in the course of real stars. 153 00:19:05,380 --> 00:19:09,460 The core mass exceeds 1.4 times the mass of the sun. 154 00:19:10,900 --> 00:19:15,120 In that case, the pressure force and the gravity force behave similarly. 155 00:19:15,130 --> 00:19:25,840 They are both one over r squared forces, and if the mass is in excess of 1.4 solar masses, the pressure is unable to push back. 156 00:19:26,500 --> 00:19:29,830 Accretion wins, gravity wins. 157 00:19:30,700 --> 00:19:34,929 And there's 1.4 solar masses as a very important number in astrophysics. 158 00:19:34,930 --> 00:19:47,130 It's known as the Chandrasekhar limit. Stellar cores with masses in excess of 1.4 solar masses ultimately will lose the battle to accretion. 159 00:19:49,620 --> 00:19:54,900 To summarise, here is a star. If I make it too big for its mass. 160 00:19:57,370 --> 00:20:06,760 Dad. The gravity rig gathers its two big. If I squeeze it down too much, the star is too small. 161 00:20:07,720 --> 00:20:14,300 Pressure inflates. Just right. 162 00:20:15,410 --> 00:20:19,140 The real star. It's optimising something. 163 00:20:20,550 --> 00:20:31,800 Now, in fact, the first serious models of stars in the 19th century by Calvin and Helmholtz were explicit accretion models. 164 00:20:32,490 --> 00:20:38,490 Accretion was actually invoked. The idea was that the star was squeezing itself. 165 00:20:39,210 --> 00:20:42,630 The whole star was contracting. It was heating up. 166 00:20:43,230 --> 00:20:52,590 But rather than having the pressure go like the density to the 5/3 power, the energy leaked out and there was an energy loss. 167 00:20:54,210 --> 00:20:57,450 And the energy loss to the star was our game. 168 00:20:57,600 --> 00:21:01,020 That was the origin of the stars luminosity. 169 00:21:04,170 --> 00:21:10,260 The source of a star's energy in the view of Kelvin and Helmholtz is the contraction itself, 170 00:21:11,910 --> 00:21:18,510 the work done to do the contraction and the release of the potential energy of the star. 171 00:21:20,700 --> 00:21:24,270 Half of the liberated energy is released as heat. 172 00:21:24,600 --> 00:21:34,440 And if that is so, the star does its arithmetic and always stays at the just right point to maintain its equilibrium. 173 00:21:34,440 --> 00:21:42,330 It's a very elegant model. There you have it in schematic form. 174 00:21:44,240 --> 00:21:47,600 Now there is ultimately a problem with the model. 175 00:21:47,600 --> 00:21:53,329 The lifetime of a star like the Sun would only be 10 million years. 176 00:21:53,330 --> 00:21:59,600 Ten to the seven years. Well, for Calvin in particular, this was not a problem. 177 00:22:01,010 --> 00:22:11,570 This was a feature. Calvin used this result together with arguments pertaining to the cooling of the earth, to argue explicitly, 178 00:22:12,140 --> 00:22:22,070 and I would say vehemently against the geological sedimentation timescales in the late 19th century for the age of the earth, 179 00:22:23,060 --> 00:22:27,860 and thus indirectly against Darwinian evolution. 180 00:22:29,180 --> 00:22:32,990 Among other critics, was, in fact, T.H. Huxley. 181 00:22:34,080 --> 00:22:42,330 Who criticise the work on the grounds that Calvin had hidden in and unwarranted assumptions in the calculations. 182 00:22:42,840 --> 00:22:49,860 The mathematics was correct. The assumptions were in error, which in fact turned out to be exactly the case. 183 00:22:50,220 --> 00:22:56,070 But I doubt for reasons that actually could have foreseen, Calvin and Helmholtz did get the basics right. 184 00:22:56,430 --> 00:23:07,049 These guys were not going to screw up at that level. But what they did not and could not get right was the fact when the contraction process hits, 185 00:23:07,050 --> 00:23:11,070 the gas above 10 million degrees tended the seven degrees. 186 00:23:11,610 --> 00:23:16,260 Nuclear reactions are triggered, providing a profound feedback. 187 00:23:16,980 --> 00:23:23,160 Greatly slowing, but in fact not absolutely stopping the core contraction. 188 00:23:24,000 --> 00:23:30,210 The sun will last not ten to the seven years. Not 10 million years, but 10 billion years. 189 00:23:33,150 --> 00:23:38,880 Now on our to do list, I talked a bit about pressure forces and accretion. 190 00:23:39,510 --> 00:23:41,490 I like to talk about. 191 00:23:43,090 --> 00:23:55,660 Rotational forces and the build-up of opposing centrifugal force and how objects in the universe manage to accrete even when they're rotating. 192 00:23:57,970 --> 00:24:05,110 So in reality, accreting gas cannot fall directly onto a central mass. 193 00:24:06,910 --> 00:24:10,170 Always falls in with some angular momentum. 194 00:24:12,980 --> 00:24:25,910 As I've said, as the gas contracts it grows or the rotational energy grows like one over R squared, the gravitational energy like one over R. 195 00:24:25,910 --> 00:24:31,100 It's a situation similar to the pressure. At small distances. 196 00:24:31,100 --> 00:24:36,380 The opposing force is growing much more rapidly than the gravitational force, 197 00:24:36,650 --> 00:24:45,200 and less angular momentum is extracted from the elements of fluid somehow and transported outward. 198 00:24:45,440 --> 00:24:52,170 Sustained accretion is simply impossible. Well, how does accretion occur? 199 00:24:54,020 --> 00:24:58,940 The gas has to lose angular momentum and generally can do so as we'll see. 200 00:24:58,940 --> 00:25:06,370 But it's a process that happens very slowly. Radiation is much more efficient. 201 00:25:06,370 --> 00:25:10,630 The gas cools under these circumstances. 202 00:25:10,640 --> 00:25:18,760 It is natural for the gas to flatten and cool, but maintain a reservoir of angular momentum. 203 00:25:20,600 --> 00:25:26,270 In which the centrifugal force v squared over r balances the gravitational force 204 00:25:26,270 --> 00:25:32,360 GM over r squared where m is the central mass r is location within the disk. 205 00:25:32,750 --> 00:25:36,050 So we speak of an accretion disk. 206 00:25:37,400 --> 00:25:41,270 We can see today we can see accretion disks. 207 00:25:41,840 --> 00:25:46,390 They can be imaged. Here is a an accretion disk. 208 00:25:46,400 --> 00:25:53,600 The colours that you see which is forming around a star in a dark cloud in the constellation cepheus the well 209 00:25:54,650 --> 00:26:00,830 what you're looking at is heat emission coming from the star and then you see some white lines in the picture. 210 00:26:00,830 --> 00:26:09,200 And those are thought to come from a shockwave, from a much larger accretion process which is raining down on the disk itself. 211 00:26:10,310 --> 00:26:18,350 Here's a rather dramatic edge on view of a disk around a forming star on the left. 212 00:26:18,950 --> 00:26:23,880 You see the disk itself, a rather thick looking disk. 213 00:26:23,900 --> 00:26:27,800 It looks a little bit like a Frisbee coming at you on the right. 214 00:26:28,070 --> 00:26:31,400 It's taken through a filter in which the disk emission has been suppressed. 215 00:26:31,850 --> 00:26:37,760 And you can see reflection from the central star quite easily popping out of the top. 216 00:26:39,410 --> 00:26:43,580 We can even see evidence of disks inside the nuclei of galaxies. 217 00:26:44,150 --> 00:26:55,310 This is a picture of a dusty Taurus and a little bright spot, which is actually an inner accretion disk in the core of an active galaxy. 218 00:26:55,670 --> 00:26:57,110 So we can see these things. 219 00:26:59,820 --> 00:27:11,790 Now, interestingly, even before we had any hope were before the days of adaptive optics and before the days of the Hubble telescope. 220 00:27:12,870 --> 00:27:22,920 Discs were already an iconic presence in astrophysics, intimately associated with high energy physics astrophysics, excuse me. 221 00:27:23,490 --> 00:27:27,210 And the search for black holes in particular. 222 00:27:27,960 --> 00:27:38,310 The two gentlemen that you see, Roger Penrose, emeritus, Ralph Spall, professor of mathematics at the University of Oxford, and Stephen Hawking, 223 00:27:39,090 --> 00:27:53,970 Lucasian, professor of mathematics at the other place, were in fact instrumental in making black holes part of the fabric of physics. 224 00:27:54,840 --> 00:28:06,090 They proved that under rather general circumstances, it was inevitable that black holes had to form these punctures in space. 225 00:28:06,090 --> 00:28:12,450 Time were inevitable, and it didn't require such unusual conditions. 226 00:28:13,720 --> 00:28:21,070 I think that was the surprising thing and this was quite an exciting result in the 1960s, 227 00:28:21,070 --> 00:28:27,460 and astrophysicists were in a mad scramble to try to find evidence for the existence of black holes. 228 00:28:27,490 --> 00:28:32,080 Now, black holes, you recall, are black because they don't emit anything. 229 00:28:33,310 --> 00:28:42,130 So the problem is, how do you see them? Well, the thought was that many black holes may be surrounded by accretion disks. 230 00:28:42,580 --> 00:28:48,430 And by looking at the accretion disks, you could gather evidence for the existence of the black hole itself. 231 00:28:50,080 --> 00:28:56,530 Now, they had a rather rough history. You remember this Chandrasekhar limit that I mentioned a few slides back. 232 00:28:57,130 --> 00:29:05,140 Well, Chandrasekhar's advisor Arthur Eddington had a few choice words to say about his students work. 233 00:29:06,900 --> 00:29:16,230 Referring to the Chandrasekhar limit. Dr. Chandrasekhar has gotten this result before, but he has rubbed it in in his last paper. 234 00:29:17,250 --> 00:29:25,860 Various accidents may intervene to save the star from collapsing to a singular point a puncture in space time. 235 00:29:26,400 --> 00:29:38,070 But I want more protection than that. I think there should be a law of nature to prevent the star from behaving in this absurd way. 236 00:29:38,820 --> 00:29:46,470 Eddington did not mince words. This was 1935, and in our meeting, I wish I had been a fly on the wall there. 237 00:29:49,470 --> 00:29:57,570 Well, black holes are ubiquitous now in theory and in the even in some sense in a real sense and observations today, 238 00:29:57,570 --> 00:30:04,350 this was far from so as we've just seen in the 1930s, and it was far more so in the 1960s. 239 00:30:06,210 --> 00:30:09,840 And an important impetus for the development of accretion. 240 00:30:09,840 --> 00:30:13,890 This theory was as a means to search for black holes. 241 00:30:14,640 --> 00:30:23,910 Why? How did this get started? Well, it really began in 1962 with the birth of X-ray astronomy. 242 00:30:24,390 --> 00:30:29,340 Two seminal figures in that field, Riccardo Giacconi and Bruno Rossi. 243 00:30:30,030 --> 00:30:42,870 And they had the bright idea simply to strap a couple of Geiger counters into the payload of a rocket and send it up and see what they could see. 244 00:30:44,010 --> 00:30:49,500 And much to everyone's surprise, they found a sky full of X-rays. 245 00:30:50,660 --> 00:30:53,660 Orders of magnitude more than anybody thought at the time. 246 00:30:58,130 --> 00:31:03,380 What is the significance of this for accretion? Well, black holes are invisible. 247 00:31:03,410 --> 00:31:08,030 How does one search for them? The answer is in close binary systems. 248 00:31:08,180 --> 00:31:10,370 Two stars in orbit around one another. 249 00:31:13,800 --> 00:31:20,750 When the stars are close enough, matter is drawn from the normal star into an accretion disk around the black hole. 250 00:31:22,180 --> 00:31:26,740 It's the disk we observe as a proxy for the whole itself. 251 00:31:26,750 --> 00:31:32,380 So this is the physical picture that astronomers have in mind. 252 00:31:33,040 --> 00:31:38,830 So you have a more or less normal star. It's very, very close to a black hole. 253 00:31:40,770 --> 00:31:46,380 If in fact the black hole is close enough, it can actually pull matter off of the star. 254 00:31:47,040 --> 00:31:51,210 And because of the presence of relative angular momentum and accretion, disk must form. 255 00:31:52,050 --> 00:31:56,820 And the accretion disk itself, as we'll see in our modelling, can get very, very hot. 256 00:31:59,370 --> 00:32:04,910 In the course of its accretion. The disk must somehow dissipate. 257 00:32:05,690 --> 00:32:10,190 The rotational energy of the gas doesn't dissipate angular momentum. 258 00:32:10,190 --> 00:32:15,590 It has to move angular momentum outwards, but it dissipates the energy and radiates it. 259 00:32:16,900 --> 00:32:23,350 If we dissipate the rotation of the gas, that will allow it to slow down and spiral inwards. 260 00:32:23,740 --> 00:32:36,400 The accretion process can continue, but like an automobile brake where we're trying to stop differential rotation, the gas becomes very, very hot. 261 00:32:37,180 --> 00:32:45,970 Indeed, it is a generic source of X-rays, which is why X-ray astronomy is the vehicle to search for these black holes. 262 00:32:46,390 --> 00:32:49,780 So are there truly black holes with disks around them? 263 00:32:51,070 --> 00:32:55,480 Well, the transformational year was 1968. 264 00:32:57,790 --> 00:33:01,570 Jocelyn Bell Burnell, who I wish could have been in the audience tonight. 265 00:33:01,600 --> 00:33:11,980 She's judging the Physics Olympiad in London. Jocelyn is a visiting professor in the department here in the Astrophysics Department. 266 00:33:12,670 --> 00:33:16,580 As a graduate student in Cambridge, she discovered radio pulsars. 267 00:33:17,320 --> 00:33:28,930 That's a radio pulsar. Radio pulsars, when they were discovered, are radio signals coming from space at incredibly precise intervals, 268 00:33:28,930 --> 00:33:33,400 more precise than any clock on earth could measure. 269 00:33:34,090 --> 00:33:37,480 And when they were found, it was stunning. 270 00:33:39,380 --> 00:33:43,280 There was even speculation that it was a sign of extraterrestrial intelligence. 271 00:33:44,560 --> 00:33:47,380 Until the sky turned out to be full of water. 272 00:33:50,650 --> 00:34:00,640 Well, very quickly, within a few months of the discovery, Tommie Gold identified pulsars with rotating neutron stars, 273 00:34:01,210 --> 00:34:10,060 these very compact objects that would rotate as a result very, very rapidly without any dissipative process slowing them down. 274 00:34:10,060 --> 00:34:16,360 It was the only clock anybody could think of. That would have the accuracy. 275 00:34:17,750 --> 00:34:26,600 To produce these radio signals and exactly how the radio signals are produced is still a complicated problem that's not completely understood. 276 00:34:28,230 --> 00:34:39,180 Later on in the same year in the Crab Nebula, in the Constellation, the crab, a fantastically rapid pulsar, was discovered. 277 00:34:40,990 --> 00:34:51,670 Going around once every 30 milliseconds and there was nothing that could go around that fast and be an astronomical source except a neutron star. 278 00:34:51,970 --> 00:35:00,459 And in fact, there were other very, very powerful and compelling arguments that link the rotating neutron stars with pulsars. 279 00:35:00,460 --> 00:35:08,980 Now in rotating neutron star is practically a rather a neutron star in itself is practically a black hole. 280 00:35:09,040 --> 00:35:21,190 So if neutron stars were real, it's just a relatively small leap to have gravity become the ultimate vector and leave behind a black hole. 281 00:35:21,200 --> 00:35:25,390 So this was powerful evidence for the existence of these objects. 282 00:35:25,400 --> 00:35:29,470 And after 1968, they were taken extremely seriously. 283 00:35:31,000 --> 00:35:36,160 More than ever, it was important to understand the physics of accretion disks. 284 00:35:37,030 --> 00:35:40,810 What makes the friction in an accretion disk? 285 00:35:41,380 --> 00:35:45,040 How does it slow down? The earth goes around the sun. 286 00:35:45,040 --> 00:35:49,300 The planets go around the sun. They don't spiral under the sun. 287 00:35:49,840 --> 00:35:54,220 Why is it that gas spirals in to a central object? 288 00:35:55,720 --> 00:35:59,080 It's not just a matter of a gas being a continuous fluid. 289 00:36:00,070 --> 00:36:05,350 When you work out the details, you still have difficulties. 290 00:36:06,960 --> 00:36:12,210 Big difficulties. Well, for more than 20 years, 291 00:36:12,220 --> 00:36:22,750 the best that theorists could do was to account for this so-called anomalous friction by adding an ad hoc term to the fundamental equations. 292 00:36:23,800 --> 00:36:27,580 Just saying something was making friction in the disk. 293 00:36:28,000 --> 00:36:32,770 Perhaps there was some sort of a subtle instability in the disk. 294 00:36:33,250 --> 00:36:37,660 We set it up in a certain way. It wouldn't maintain the configuration. 295 00:36:39,820 --> 00:36:49,450 That was speculation. And then these sort of instabilities resulted in a large internal friction because they would trigger turbulence. 296 00:36:50,080 --> 00:36:53,470 And turbulence with all of its mixing would be highly dissipative. 297 00:36:54,520 --> 00:37:04,270 They even gave this kind of a discourse process, the sort of friction resulting from rubbing of adjacent orbits in the disk. 298 00:37:05,080 --> 00:37:08,980 It's called the viscosity. They call this viscosity alpha viscosity. 299 00:37:11,010 --> 00:37:17,790 Now the most successful implementation of this Alpha Formalism by Shakira in San Diego. 300 00:37:19,260 --> 00:37:30,250 Gave quite useful predictions. And in fact, some of the most interesting predictions required this ad hoc term to be added to the equations. 301 00:37:30,570 --> 00:37:33,810 But then the results turned out to be independent of it. 302 00:37:35,010 --> 00:37:37,590 An interesting set of mathematical affairs. 303 00:37:38,610 --> 00:37:47,250 Many results would depend upon this artificial term that was not well understood, but only in a very marginally dependent way. 304 00:37:49,020 --> 00:37:57,410 Still absent any understanding of what was really going on in the fluids, matters remained entirely unsatisfactory. 305 00:37:57,420 --> 00:38:00,510 If you did a calculation. This is what you would find. 306 00:38:00,510 --> 00:38:07,950 There would be your unperturbed orbit. You would do a perturbation, and the perturbation would look like this. 307 00:38:08,910 --> 00:38:14,490 You would go back and forth. Nothing very exciting would happen when you disturb the gas. 308 00:38:15,690 --> 00:38:20,760 Well, the prevailing approach, I would say circa 1980. 309 00:38:22,490 --> 00:38:28,410 Don't worry. Be happy. Some, however, were a bit more nettled. 310 00:38:29,430 --> 00:38:39,540 And this is a wonderful quotation by Professor Kip Thorne, one of the leading theorists of the 1970s, one of fact, 311 00:38:39,540 --> 00:38:45,180 one of the leading relativists who was extremely active at the frontlines of the search for black holes. 312 00:38:45,990 --> 00:38:53,310 And he was extremely nettled, he said. The chief stumbling block at this point is the friction in the disk. 313 00:38:53,880 --> 00:39:01,230 We do not know whether the friction is generated by turbulence in the spiralling gas by magnetic fields. 314 00:39:01,240 --> 00:39:05,190 We haven't talked about magnetic fields. But I will come back to them. 315 00:39:06,180 --> 00:39:11,010 Or by a combination of turbulence and magnetic fields. 316 00:39:12,270 --> 00:39:15,960 December 1974, evidently. 317 00:39:18,250 --> 00:39:21,650 Astronomers have been doing this thing kind of thing for a while. 318 00:39:21,670 --> 00:39:25,930 It turns out, as I was reading this book by Dorothy Sellers, by the way, 319 00:39:25,930 --> 00:39:30,550 an excellent read for scientists, if you've not seen it, the documents in the case at one point, 320 00:39:30,880 --> 00:39:40,960 one of the characters in the book utters, I hate being accounted for, as though I were some incalculable quantity in an astronomical equation. 321 00:39:42,220 --> 00:39:47,770 Evidently, this was something that astronomers were forced to do from time to time. 322 00:39:50,400 --> 00:39:52,260 It is a remarkable fact, however. 323 00:39:54,220 --> 00:40:05,110 That if we consider the effects of an electrical current being present in the flow it and let me remind you especially in a disk around a black hole, 324 00:40:05,110 --> 00:40:14,860 these are completely ionised gases. These are gases where there are ions and electrons that are free to move around. 325 00:40:16,040 --> 00:40:23,870 And if that's the case, it turns out that the dynamical, the rotational properties of the fluid are dramatically altered. 326 00:40:25,160 --> 00:40:28,850 The system is in fact completely destabilised. 327 00:40:31,370 --> 00:40:43,550 If the rotation rate is faster on the inside than on the outside, which almost all rotating systems are if they're not in a state of uniform rotation. 328 00:40:45,930 --> 00:40:55,589 Differentially rotating systems are prone to this instability, which is now known as the magneto rotational instability, instability. 329 00:40:55,590 --> 00:41:02,490 And if you look at the dates, you notice it had precursors in the literature dating back to the 1950s. 330 00:41:04,250 --> 00:41:09,680 And that's an interesting question as to why it wasn't appreciated earlier. 331 00:41:10,880 --> 00:41:17,720 And the answer seems to be that the work itself was extremely mathematical, 332 00:41:18,230 --> 00:41:26,120 and neither of the earlier authors made any real attempt to elicit the physics of the instability itself. 333 00:41:26,660 --> 00:41:35,540 And so, although it was in a classic textbook by Chandrasekhar for many, many years, it just stayed on people's shelves, 334 00:41:36,020 --> 00:41:45,740 completely unappreciated until John Hawley and I went back totally unaware of the earlier work founded on our own, 335 00:41:46,010 --> 00:41:48,740 and then traced it back to these earlier origins. 336 00:41:50,280 --> 00:41:58,770 So if we take another look at Professor Thorne's quotation back in 1974, we're now in a position to answer him. 337 00:41:59,670 --> 00:42:05,460 The answer to his question is yes. Turbulence, magnetic fields. 338 00:42:06,000 --> 00:42:18,480 Combination of turbulence and magnetic fields. Yes. In fact, we now know that it is the magnetic field itself that introduces the turbulence. 339 00:42:19,230 --> 00:42:31,650 To understand this, we can go back to the work of Professor Hans Al-Fayed, who won a Nobel Prize for eliciting this in the 1970s. 340 00:42:32,370 --> 00:42:36,540 And in fact, it really is remarkable. 341 00:42:38,570 --> 00:42:43,580 Many of us have worked with these equations long enough now that it's sort of second nature to us. 342 00:42:44,600 --> 00:42:49,610 But it's amazing that we can do anything at all with a magnetic field in a gas. 343 00:42:49,670 --> 00:42:55,960 I think you have a gas which is full of electrons and electrons are doing electron things. 344 00:42:56,690 --> 00:43:01,720 And then you have the ions and the ions behave a different set of equations. 345 00:43:01,730 --> 00:43:06,380 They have their iron things that they're doing, they're moving in different ways. 346 00:43:06,710 --> 00:43:12,650 And if they don't have the same velocities, if there's a difference between the ions and electrons, you get a current, 347 00:43:13,010 --> 00:43:22,490 and then the current makes magnetic fields, and then the magnetic fields act back on the ions and electrons and it seems a hopeless can of worms. 348 00:43:23,540 --> 00:43:28,640 And in reality, it turns out to be remarkably simple. 349 00:43:30,440 --> 00:43:38,960 Because if we take into account the fact that this gas behaves like an excellent electrical conductor. 350 00:43:40,330 --> 00:43:43,900 It turns out that the magnetic field lines of force. 351 00:43:44,260 --> 00:43:47,540 You should think of these lines of force as kind of physical things. 352 00:43:47,800 --> 00:43:53,710 These are the lines that you see when you take a bar magnet and sprinkle it with iron filings. 353 00:43:54,140 --> 00:44:04,870 You see these characteristic patterns or merge those magnetic lines of force, act as though they were painted in, frozen in to the gas. 354 00:44:05,260 --> 00:44:11,260 If the gas swirls around the magnetic field, lines of force swirl around as well. 355 00:44:12,830 --> 00:44:18,800 Moreover, if we take one of these magnetic lines of force and we bend it. 356 00:44:20,650 --> 00:44:27,550 It acts like a taut spring. Its physical effect is to provide attention. 357 00:44:28,300 --> 00:44:32,050 Like a plucked spring. Excuse me. A plucked string. 358 00:44:33,070 --> 00:44:43,060 Or even more simply, when you have a magnetic field present, the fluid elements feel like they are hooked together by a spring. 359 00:44:43,750 --> 00:44:48,640 Hook is a good word to use in that context because they obey Hooke's law. 360 00:44:49,510 --> 00:44:55,570 They feel an attractive force. I try to separate them, which is proportional to their separation. 361 00:44:56,560 --> 00:45:03,700 So magnetic fields are difficult to think of it difficult for me to think of when I've been struggling with them for decades. 362 00:45:04,360 --> 00:45:09,160 I find it easier to think of the masses on springs as something I can understand. 363 00:45:09,970 --> 00:45:15,760 So imagine we have two masses connected by a spring in orbit around the star. 364 00:45:15,760 --> 00:45:22,080 What will happen? You're going to be surprised if you haven't seen this before. 365 00:45:22,770 --> 00:45:29,140 Let's focus in on the process. Let's take a close up view of those two masses. 366 00:45:30,070 --> 00:45:39,280 And we have a mass which is orbiting a little bit farther out, rotating a little bit more slowly. 367 00:45:39,280 --> 00:45:44,530 So it has a small arrow. And then I have another mass over here which is going faster. 368 00:45:44,800 --> 00:45:49,600 My rotational centre is down here and they're hooked by this spring. 369 00:45:50,200 --> 00:45:55,510 The spring is pulling back on this inner mass, which is trying to go faster, 370 00:45:56,080 --> 00:46:01,720 but it is losing its angular momentum to the outer mass, which is moving more slowly. 371 00:46:02,880 --> 00:46:05,010 What will happen under those circumstances? 372 00:46:05,580 --> 00:46:15,480 Well, the inner mass will spiral in the outer mass will acquire the angular momentum and start to spiral out. 373 00:46:17,550 --> 00:46:25,650 And the spring will stretch. But that just makes the connection between them, the talk between them even more powerful. 374 00:46:27,040 --> 00:46:36,520 That's the basis of the instability. If you put a weak spring, I mean, this is a purely mechanical process between two masses in orbit. 375 00:46:37,210 --> 00:46:41,380 It doesn't hook them together. It makes them fly apart. 376 00:46:42,220 --> 00:46:49,810 And that's the basis of the origin of turbulence in these disks when even a very weak magnetic field is present. 377 00:46:51,040 --> 00:46:55,270 If we go back to the case of the magnetic field, here's a schematic picture. 378 00:46:56,320 --> 00:46:59,680 The field lines are pointing upwards. Here's my equilibrium. 379 00:47:00,250 --> 00:47:09,969 If I imagine bending those lines and then letting the equations take over and let it evolve, this is what I find. 380 00:47:09,970 --> 00:47:15,280 When I actually do the simulation, things start to slide. 381 00:47:15,280 --> 00:47:18,490 And of course, a real disk doesn't do that indefinitely. 382 00:47:18,970 --> 00:47:28,660 I have pieces of the disks going out. They encounter other pieces of the disk coming in and it becomes happily a turbulent mess. 383 00:47:30,380 --> 00:47:34,940 That's what we want. That's the origin of the friction. 384 00:47:39,520 --> 00:47:45,400 Once we knew this, it was actually embarrassingly easy to simulate on a computer. 385 00:47:46,960 --> 00:48:00,920 Here is. A doughnut starting off from an initial condition, which is not a disc like condition, but it is threaded by a magnetic field. 386 00:48:00,920 --> 00:48:04,780 And on the left there is a meridian, meridian or slice. 387 00:48:07,150 --> 00:48:17,020 From the right side. You're viewing it from above, and you're looking at how the density evolves in a disk which is subject to this instability. 388 00:48:17,530 --> 00:48:20,830 And it does precisely what it should do. 389 00:48:21,190 --> 00:48:24,510 Mixes different fluid elements. Makes the disk hot. 390 00:48:26,690 --> 00:48:37,190 It fits the bill perfectly. It's even possible to get more dramatic looking outflows now on the left side of this picture. 391 00:48:37,210 --> 00:48:47,560 I'll let this run. I have presented a diagram which displays curves known as echoey potential surfaces. 392 00:48:47,560 --> 00:48:48,940 That sounds complicated. 393 00:48:49,240 --> 00:48:58,000 Aqui Potential surfaces are simply the surfaces that if I were to make the disc out of water, that would be the surfaces that the water would fill up. 394 00:48:58,620 --> 00:49:05,920 I could potential surfaces on the earth like the ocean, follow spherical shells in a disc. 395 00:49:06,100 --> 00:49:11,890 They follow unusual shapes because you have both gravity and strong rotation to contend with. 396 00:49:12,640 --> 00:49:19,330 And one of the things that is interesting is that there are some surfaces which are quite happy and are completely bound. 397 00:49:19,330 --> 00:49:24,549 They look like teardrops and then other surfaces near the disc where you get a 398 00:49:24,550 --> 00:49:30,220 significant well which open up and extend to infinity so that you can have an outflow. 399 00:49:30,790 --> 00:49:39,119 And this is a. An excellent example of exactly that process of material that has come in through the body of 400 00:49:39,120 --> 00:49:45,660 the disk and then managed to hop on to one of these other surfaces and turns into an outflow. 401 00:49:46,770 --> 00:49:57,180 So. One of the ubiquitous properties that we see with disks is not only accretion itself into a central object, 402 00:49:57,660 --> 00:50:02,430 but a lot of material is coming out both in jets and winds. 403 00:50:03,360 --> 00:50:11,520 And this is something that even if you didn't want to put it into, this simulation emerges extremely naturally. 404 00:50:11,880 --> 00:50:17,190 So there's something right about this explanation. You're getting many, many things out that you didn't put in. 405 00:50:21,180 --> 00:50:26,149 Let me summarise. In preparing this talk. 406 00:50:26,150 --> 00:50:31,130 It was kind of fun to remember what the big problems were when I was a graduate student. 407 00:50:32,400 --> 00:50:35,550 Not that long ago. I hope I'm. 408 00:50:37,180 --> 00:50:45,610 1970s, late 1970s. And in fact, there has really been major historical progress. 409 00:50:47,020 --> 00:50:56,320 We have seen the origin of large scale structure in the universe via the tiny fluctuations that missions like 410 00:50:57,490 --> 00:51:06,520 WMAP more recently Planck have revealed all these little tiny specks essentially correspond to fluctuations. 411 00:51:07,830 --> 00:51:09,840 Really fluctuations in the radiation, 412 00:51:09,840 --> 00:51:19,720 but they're tracing fluctuations in the matter that will eventually grow into the large scale structure that has made the universe what it is. 413 00:51:19,740 --> 00:51:24,330 We can see that these are tiny fluctuations and part intended. 414 00:51:24,330 --> 00:51:30,660 The four part content of a five. But we still don't know how to make galaxies. 415 00:51:31,990 --> 00:51:40,810 I can't give you a lecture on how galaxies are made. We don't even understand a lot of simple accretion processes. 416 00:51:41,170 --> 00:51:48,910 I'm embarrassed to say we don't understand how hot gas accretes within clusters of galaxies. 417 00:51:49,420 --> 00:51:53,020 Galaxies cluster together in groups with thousands of members. 418 00:51:53,440 --> 00:52:05,510 They say gas in a galaxy spills out. It occupies the space in between the galaxies emitting X-rays, and it's behaving in a way we don't understand. 419 00:52:06,500 --> 00:52:14,720 We seem to have kind of a carbon problem again in the sense that there's a heat source which is keeping the gas hot that we don't understand. 420 00:52:15,500 --> 00:52:20,330 An embarrassing theoretical failure, one that I myself have spent time working on. 421 00:52:22,390 --> 00:52:29,530 Here's a lovely picture. This is the centre of the galaxy and what you see in these kinds of rings. 422 00:52:29,860 --> 00:52:35,020 These are stars that are whose orbits have been tracked. 423 00:52:35,050 --> 00:52:39,490 These are the proper motions of the stars near the centre of the galaxy. 424 00:52:39,520 --> 00:52:43,480 Look at this orbit. It's incredibly elliptical. 425 00:52:43,780 --> 00:52:48,160 It's almost like a straight line in and out. These are the actual orbits. 426 00:52:49,180 --> 00:52:55,630 This is the cleanest evidence that we have, that there is a black hole in the centre of the galaxy. 427 00:52:56,110 --> 00:53:06,550 You can determine the mass quite accurately 4.6 million solar masses in a region on the scale of the size of the solar system. 428 00:53:08,110 --> 00:53:12,970 It's got to be a black hole. It's quiet as a whisper. 429 00:53:14,930 --> 00:53:22,790 There's nothing going on. At one point, black holes were invoked to explain things called quasars. 430 00:53:22,790 --> 00:53:25,940 The brightest objects in the room in the in the universe. 431 00:53:26,480 --> 00:53:31,520 They were incredibly luminous. If you had a black hole present, fireworks resulted. 432 00:53:31,970 --> 00:53:41,990 My previous part of this lecture was to explain why the disks around black holes emit X-rays 4.6 million solar masses and little the galaxy. 433 00:53:43,220 --> 00:53:54,270 Nothing. It's not understood. Our understanding of many observational facts surrounding black hole accretion sources is minimal or nil. 434 00:53:56,940 --> 00:54:03,720 The disks are constantly erupting and quieting down and changing their spectral appearances completely. 435 00:54:03,990 --> 00:54:09,280 There's a famous example. Here's a black hole candidate and there is a spectrum. 436 00:54:09,790 --> 00:54:14,920 This is called a state power law. Seven months later, the spectrum looks like this. 437 00:54:15,350 --> 00:54:20,800 Now, this spectrum, we don't understand. We do understand this is pretty much what you would get from an accretion disk. 438 00:54:22,000 --> 00:54:28,480 Here. The spectrum seems to be dominated by a population of relativistic particles. 439 00:54:28,930 --> 00:54:32,770 And why within a period of seven months we should go from one to the other, 440 00:54:33,610 --> 00:54:38,980 completely changing, at least the radiative state of the disc is not understood. 441 00:54:39,990 --> 00:54:46,230 So there will be a lot to keep people like me, I hope, gainfully employed for a while. 442 00:54:48,750 --> 00:54:55,510 Parasitology. Oh, this is an embarrassment. It's like all my failures in front of my eyes. 443 00:54:55,530 --> 00:54:58,920 But a stellar accretion disk or a major town challenge. 444 00:54:59,070 --> 00:55:02,310 Too big to succeed. Big disk is bad. 445 00:55:02,430 --> 00:55:07,560 It rotates slowly, it's cold, it's dusty, and it's incredibly neutral. 446 00:55:08,190 --> 00:55:12,960 And it's very difficult to see how magnetic fields can work in a situation like this. 447 00:55:13,560 --> 00:55:19,740 It appears as though it can just barely do it. But the physics is not well understood. 448 00:55:19,740 --> 00:55:27,600 There really is a long way to go. At the same time, things are not all bleak. 449 00:55:27,600 --> 00:55:37,679 The combination of, I would say, improved physical analysis and what is truly dazzling progress in computational power allows us to do, 450 00:55:37,680 --> 00:55:39,280 among other things I should say, 451 00:55:39,300 --> 00:55:51,840 three dimensional general relativistic magneto, hydro, dynamical, turbulent disk calculations of an accreting disk around a rotating core black hole. 452 00:55:51,840 --> 00:55:56,819 For those of you who know what those words means, that is simply stunning. 453 00:55:56,820 --> 00:56:06,210 We are in an era we may not be able to do that calculation well, but we can do it and get some kind of results out of it and. 454 00:56:07,750 --> 00:56:10,540 It will only get better with time and I think it's quite exciting. 455 00:56:12,250 --> 00:56:18,970 You just saw in this earlier calculation with the jet emerging, we start with any old kind of random initial equilibrium. 456 00:56:19,480 --> 00:56:24,160 Let the magneto rotational instability run and it turns into a cat. 457 00:56:24,160 --> 00:56:29,800 Larry and Jeff, excuse me, kept Larry in disk with the jet popping out of it right where it should be. 458 00:56:30,430 --> 00:56:36,700 So this is progress. This really is progress. And just remember. 459 00:56:39,010 --> 00:56:42,310 That's what the universe would look like without accretion. 460 00:56:45,610 --> 00:56:55,900 Without accretion, the universe would be a big, boring bag of gas doing nothing hurtling into oblivion. 461 00:56:56,830 --> 00:57:07,030 So let's be glad some of us are taking the trouble to put things on top of other things. 462 00:57:09,140 --> 00:57:09,980 Thank you very much.