1
00:00:00,010 --> 00:00:04,070
[music]

2
00:00:04,090 --> 00:00:08,190
[music]

3
00:00:08,210 --> 00:00:12,200
Narrator: When a massive star

4
00:00:12,220 --> 00:00:16,250
explodes as a supernova, its core may be crushed into one of two types

5
00:00:16,270 --> 00:00:20,360
of compact remnant: a black hole, or a neutron star.

6
00:00:20,380 --> 00:00:24,510
Neutron stars are the size of a city, but

7
00:00:24,530 --> 00:00:28,660
contain more mass than our sun. They rotate rapidly, host 

8
00:00:28,680 --> 00:00:32,720
powerful magnetic fields, and produce beams of radiation that emit a wide 

9
00:00:32,740 --> 00:00:36,880
range of energy. When we detect pulses as the beams sweep over 

10
00:00:36,900 --> 00:00:41,060
Earth, the object is known as a pulsar. Paul Ray: They can spin at

11
00:00:41,080 --> 00:00:45,150
many times per second on their axis; the fastest pulsars spin over 700 times 

12
00:00:45,170 --> 00:00:49,180
per second. And that rapidly spinning massive object, generates 

13
00:00:49,200 --> 00:00:53,260
extremely strong magnetic fields and accelerates particles to 

14
00:00:53,280 --> 00:00:57,400
high energies. And we see that those accelerated particles 

15
00:00:57,420 --> 00:01:01,440
emitting energy in the form of gamma rays, x-rays, and radio waves.

16
00:01:01,460 --> 00:01:05,480
And when that beam sweeps past the line of sight to the Earth we see it pulse on 

17
00:01:05,500 --> 00:01:09,540
and that's why they're named pulsars. Narrator: The most

18
00:01:09,560 --> 00:01:13,670
sensitive tool for observing pulsars in gamma-ray light is NASA's 

19
00:01:13,690 --> 00:01:17,840
Fermi Gamma-ray Space Telescope. Fermi scans the entire 

20
00:01:17,860 --> 00:01:21,950
sky for high-energy sources and has found many previously undetected

21
00:01:21,970 --> 00:01:26,130
gamma-ray emitters. Scientists have identified many of these, 

22
00:01:26,150 --> 00:01:30,220
but for some, the source of the gamma rays remains unknown.

23
00:01:30,240 --> 00:01:34,320
Roger W. Romani: I got interested a couple years ago in trying to find the 

24
00:01:34,340 --> 00:01:38,470
limits of what Fermi can discover, how extreme these objects can be, and 

25
00:01:38,490 --> 00:01:42,530
in order to do that I focused on the set of objects that are 

26
00:01:42,550 --> 00:01:46,640
relatively bright and well measured by Fermi and 

27
00:01:46,660 --> 00:01:50,720
found that virtually all of them have now been identified. At present, when I started 

28
00:01:50,740 --> 00:01:54,870
this project,there were only six objects which we hadn't figured out what they were yet.

29
00:01:54,890 --> 00:01:58,930
Despite intense searches, at radio, with radio wave lengths,

30
00:01:58,950 --> 00:02:03,070
which is the standard way in which people find pulsars--and also looking at the gamma rays 

31
00:02:03,090 --> 00:02:07,130
themselves--no pulsations had been seen. So something was unique

32
00:02:07,150 --> 00:02:11,250
about these six objects and I thought, that's where the discovery space is going to be.

33
00:02:11,270 --> 00:02:15,270
If we can track down what those are, we will have a good chance of finding something new.

34
00:02:15,290 --> 00:02:19,330
We took this small set of six objects and attacked them

35
00:02:19,350 --> 00:02:23,390
with a number of wave bands, but I think the thing that helped us make the greatest progress

36
00:02:23,410 --> 00:02:27,490
was looking in the optical, in visible light. Now this may seem a little bit unusual 

37
00:02:27,510 --> 00:02:31,580
for studying the high-energy gamma-ray universe, but, it turns out that 

38
00:02:31,600 --> 00:02:35,750
many of these objects seem to have optical counterparts. And if you can figure out what the

39
00:02:35,770 --> 00:02:39,800
visible light counterpart of an object is, you're long ways along the 

40
00:02:39,820 --> 00:02:43,870
track to understanding what it's all about. Paul Ray: It was Roger 

41
00:02:43,890 --> 00:02:48,000
Romani's optical observations that discovered a counterpart 

42
00:02:48,020 --> 00:02:52,120
to the gamma-ray source that showed a binary period that was 

43
00:02:52,140 --> 00:02:56,180
indicative of this potentially being a binary millisecond pulsar. Alice Harding: It brightened, and it 

44
00:02:56,200 --> 00:03:00,290
dimmed, and brightened, and so this looked like we were

45
00:03:00,310 --> 00:03:04,300
looking at possibly something which was 

46
00:03:04,320 --> 00:03:08,350
irradiated by a companion pulsar. And that every 

47
00:03:08,370 --> 00:03:12,420
time you're looking at the bright face you see a bright optical source and when 

48
00:03:12,440 --> 00:03:16,510
it rotates away from you, and you see the dark face, you don't see anything.

49
00:03:16,530 --> 00:03:20,510
Roger Romani: We managed to get enough observations of the object to piece together its orbital 

50
00:03:20,530 --> 00:03:24,560
period, and found, remarkably, that it was an incredibly

51
00:03:24,580 --> 00:03:28,660
heated object--blue white on one side, deep, deep red on the other--

52
00:03:28,680 --> 00:03:32,750
that was orbiting around something invisible with an orbital period of about

53
00:03:32,770 --> 00:03:36,790
one and a half hours. Now, that's faster then any spin powered

54
00:03:36,810 --> 00:03:40,850
pulsar ever known, and indicates that it's a really, really tight system 

55
00:03:40,870 --> 00:03:44,880
and that the gamma rays are blasting the companion at point-blank range.

56
00:03:44,900 --> 00:03:48,970
Our colleagues in Germany managed to use the orbital period

57
00:03:48,990 --> 00:03:53,100
that we measured to search in the gamma rays directly and, with a computational 

58
00:03:53,120 --> 00:03:57,160
tour de force, managed to find the pulse signal of the pulsar

59
00:03:57,180 --> 00:04:01,210
directly in the gamma rays themself. Holger Pletsch: What I'm

60
00:04:01,230 --> 00:04:05,240
doing is blind searches for pulsars so that we

61
00:04:05,260 --> 00:04:09,330
try to find pulsars that have not been seen before. So you don't know 

62
00:04:09,350 --> 00:04:13,370
how fast the pulsar is spinning, where exactly it is sitting in the sky.

63
00:04:13,390 --> 00:04:17,420
To do that, you have basically to try, every possible combination of 

64
00:04:17,440 --> 00:04:21,510
parameters--if they match your data output stream. So the problem is 

65
00:04:21,530 --> 00:04:25,530
that the number of possible combinations is tremendously high, so the

66
00:04:25,550 --> 00:04:29,590
straightforward brute force approach is impossible. The computation power you would

67
00:04:29,610 --> 00:04:33,640
need would be in excess of what is available in the whole planet.

68
00:04:33,660 --> 00:04:37,820
So our work is to invent more efficient methods to do that.

69
00:04:37,840 --> 00:04:41,850
The basic method is analogous to zooming.

70
00:04:41,870 --> 00:04:45,920
It's similar to changing your objectives

71
00:04:45,940 --> 00:04:49,990
of your microscope, in favor of one of higher magnification, so you look at

72
00:04:50,010 --> 00:04:54,020
one interesting point on the slide, and then you zoom in on that.

73
00:04:54,040 --> 00:04:58,110
And then you further zoom in if it still is interesting. To find the pulsations 

74
00:04:58,130 --> 00:05:02,260
in the gamma ray data requires about

75
00:05:02,280 --> 00:05:06,300
5,000 cpu days. So if you do it 

76
00:05:06,320 --> 00:05:10,420
on your laptop you need 5,000 days, but if you 

77
00:05:10,440 --> 00:05:14,590
have 5,000 laptops, you only need one day. And so 

78
00:05:14,610 --> 00:05:18,710
that is the path we took because we have a computing cluster that is called Atlas

79
00:05:18,730 --> 00:05:22,880
at the Albert Einstein Institute in Hannover and that computing facility

80
00:05:22,900 --> 00:05:26,960
we used for this analysis and it was immediately clear.

81
00:05:26,980 --> 00:05:31,090
This is a detection, so it's, it cannot be a noise fluctuation

82
00:05:31,110 --> 00:05:35,240
because it's so loud in the data.

83
00:05:35,260 --> 00:05:39,280
Narrator: A pulsar, that was a strong gamma ray source yet showed no radio signature

84
00:05:39,300 --> 00:05:43,380
intrigued researchers. Among them was Paul Ray of the 

85
00:05:43,400 --> 00:05:47,530
Naval Research Laboratory. He and his team thought they might have a solution to the

86
00:05:47,550 --> 00:05:51,600
puzzling lack of radio emission. Paul Ray: When we first discovered the 

87
00:05:51,620 --> 00:05:55,690
system I looked back at our archival radio observations and none of them

88
00:05:55,710 --> 00:05:59,800
showed detections of this pulsar. We think that nearly all pulsars do emit 

89
00:05:59,820 --> 00:06:03,840
radio waves. The radio beam is emitted for most pulsars 

90
00:06:03,860 --> 00:06:07,890
from a region above the polar cap of the star, and that means it's a tightly concentrated 

91
00:06:07,910 --> 00:06:12,040
flashlight-type beam. In a system like this where there's wind being blown

92
00:06:12,060 --> 00:06:16,080
off the companion star, there's a lot of obscuring material along the line of sight.

93
00:06:16,100 --> 00:06:20,170
It might be that it is a radio pulsar and we just couldn't see it.

94
00:06:20,190 --> 00:06:24,230
And the one way to confront that is to use a higher radio frequency,

95
00:06:24,250 --> 00:06:28,360
that's more penetrating, that's less affected by the scattering in the intervening 

96
00:06:28,380 --> 00:06:32,460
material. And so we went and made an observation with the Robert C. Byrd Green Bank

97
00:06:32,480 --> 00:06:36,650
Telescope run by the National Radio Astronomy Observatory in West Virginia,

98
00:06:36,670 --> 00:06:40,690
at a much higher frequency than typical radio observations. And it was in one 

99
00:06:40,710 --> 00:06:44,750
of those observations that we first saw the signal from the system. And it 

100
00:06:44,770 --> 00:06:48,770
appears that it is most of the time obscured by the material 

101
00:06:48,790 --> 00:06:52,880
from its companion. Narrator: A combination of radio 

102
00:06:52,900 --> 00:06:56,900
optical, and gamma-ray data allowed astronomers to assemble a 

103
00:06:56,920 --> 00:07:00,980
complete picture of the system. It turned out to be a rare black widow

104
00:07:01,000 --> 00:07:05,140
binary where a rejuvenated pulsar is gradually evaporating a low 

105
00:07:05,160 --> 00:07:09,170
mass companion star. Alice Harding: They get this name because they are in

106
00:07:09,190 --> 00:07:13,270
very close systems, with the companion star being close enough to the 

107
00:07:13,290 --> 00:07:17,450
neutron star, that the neutron star is irradiating the companion.

108
00:07:17,470 --> 00:07:21,500
So the neutron star is producing a wind of energetic particles and

109
00:07:21,520 --> 00:07:25,610
magnetic fields, and also all the gamma rays that are radiated. All this 

110
00:07:25,630 --> 00:07:29,790
hits the companion star and heats it up to very

111
00:07:29,810 --> 00:07:33,840
high temperatures, but only on one side. So the side that's

112
00:07:33,860 --> 00:07:37,940
towards the neutron star gets blasted by this pulsar wind.

113
00:07:37,960 --> 00:07:42,080
Paul Ray: And it has been whittled away over billions of years to where it now is 

114
00:07:42,100 --> 00:07:46,140
only about 8 times the mass of Jupiter. Alice Harding: This whole system is about the size

115
00:07:46,160 --> 00:07:50,250
of the Earth-Moon system, so it's very compact.

116
00:07:50,270 --> 00:07:54,280
Paul Ray: We see the pulsar at the center, spinning, and emitting beams of

117
00:07:54,300 --> 00:07:58,350
radio and gamma rays. The radio waves are represented by the green, 

118
00:07:58,370 --> 00:08:02,480
and the gamma rays are represented by the magenta. That radiation that 

119
00:08:02,500 --> 00:08:06,510
impinges on the star is blowing off clouds of ionized material 

120
00:08:06,530 --> 00:08:10,600
that are collecting around the system, and that's what obscures the radio emission.

121
00:08:10,620 --> 00:08:14,760
So we see that most of the time, the radio represented in green only

122
00:08:14,780 --> 00:08:18,800
makes it to that obscuring material, and not through it. While the gamma rays

123
00:08:18,820 --> 00:08:22,890
which are much more penetrating go right through.

124
00:08:22,910 --> 00:08:27,010
Roger W. Romani: It turns out that in as far as it's a pulsar, it's not so very unusual.

125
00:08:27,030 --> 00:08:31,040
What's unusual about it is this binary system, and the binary 

126
00:08:31,060 --> 00:08:35,130
system seems to have--through its history--allowed this neutron star

127
00:08:35,150 --> 00:08:39,210
pulsar to accrete enormous amounts of mass. The measurements to date

128
00:08:39,230 --> 00:08:43,340
suggest that it is very heavy indeed, and heavy neutron stars push

129
00:08:43,360 --> 00:08:47,380
the absolute extreme of the densest matter in our visible universe. I say 

130
00:08:47,400 --> 00:08:51,440
this because many people think of black holes as being exotic and the most extreme objects 

131
00:08:51,460 --> 00:08:55,550
known, but after all, a black hole has collapsed to the point where 

132
00:08:55,570 --> 00:08:59,680
nothing is visible, it's black. A neutron star is an object that's 

133
00:08:59,700 --> 00:09:03,710
on the hairy edge of becoming a black hole, yet is still visible in our universe.

134
00:09:03,730 --> 00:09:07,780
Hence the study of the these ultra-massive neutron stars 

135
00:09:07,800 --> 00:09:11,830
gives us the opportunity to study the most extreme matter in our visible universe.

136
00:09:11,850 --> 00:09:15,930
If this fellow is as heavy as he seems, he pushes

137
00:09:15,950 --> 00:09:20,090
that study to a new horizon, to a region of density 

138
00:09:20,110 --> 00:09:24,150
and pressure which has never previously been seen.

139
00:09:24,170 --> 00:09:28,230
[music]

140
00:09:28,250 --> 00:09:32,390
[beeping]

141
00:09:32,410 --> 00:09:36,450
[beeping]

142
00:09:36,470 --> 00:09:40,394