WEBVTT 1 00:00:01.480 --> 00:00:05.599 Welcome to the quart Side Quantum Physics Podcast, an exploration 2 00:00:05.679 --> 00:00:10.880 of the fundamental structure of reality, where quantum laws govern matter, energy, 3 00:00:11.199 --> 00:00:16.239 and information. Here, uncertainty is a feature, not a flaw, 4 00:00:16.719 --> 00:00:19.239 and understanding begins at the smallest scales. 5 00:00:27.679 --> 00:00:33.399 So there is this really persistent idea in astronomy, and honestly, 6 00:00:33.439 --> 00:00:35.200 I think it's just human nature at this point. If 7 00:00:35.200 --> 00:00:37.840 you want to see something better, or you see something 8 00:00:37.920 --> 00:00:40.200 further away, you just need to build something bigger. 9 00:00:39.920 --> 00:00:42.359 Right the classic light bucket philosophy. 10 00:00:41.920 --> 00:00:43.960 Exactly the light bucket. You want to see a faint star, 11 00:00:44.439 --> 00:00:46.399 build a bigger bucket to catch the light. You want 12 00:00:46.439 --> 00:00:50.159 to see a smaller detail, build a wider mirror to 13 00:00:50.200 --> 00:00:52.640 sharpen the image. It's a very brute force. 14 00:00:52.439 --> 00:00:55.280 Approach, it is. But to be fair, it has served 15 00:00:55.359 --> 00:00:58.679 us incredibly well for about four hundred years. I mean, 16 00:00:58.880 --> 00:01:01.719 from Galileo's first spyglass all the way to the James 17 00:01:01.719 --> 00:01:06.640 webspased telescope. The fundamental logic has remained entirely consistent. More 18 00:01:06.719 --> 00:01:10.599 surface area equal more photons and more photons equal better resolution. 19 00:01:10.879 --> 00:01:13.200 Yeah, that makes total sense. But we are doing a 20 00:01:13.200 --> 00:01:16.239 deep dive today into a stack of papers specifically focusing 21 00:01:16.319 --> 00:01:19.239 on a proposal published in Physical Review Letters in February 22 00:01:19.280 --> 00:01:22.760 twenty twenty six that suggests we are finally hitting. 23 00:01:22.519 --> 00:01:25.400 A wall, a hard physical wall, yes right. 24 00:01:25.480 --> 00:01:28.200 A wall where building bigger just isn't feasible anymore. The 25 00:01:28.200 --> 00:01:31.319 team behind this research involves folks from the University of Arizona, 26 00:01:31.400 --> 00:01:34.719 the University of Maryland, and the NASA Goddard Space Flight Center, 27 00:01:35.319 --> 00:01:38.159 and their argument is absolutely fascinating. 28 00:01:38.280 --> 00:01:41.200 It really is. They are basically saying that to take 29 00:01:41.239 --> 00:01:44.519 the next massive leap in seeing the universe, we don't 30 00:01:44.599 --> 00:01:47.319 need to pour more glass. We need to completely hack 31 00:01:47.760 --> 00:01:49.519 the physics of light itself. 32 00:01:49.200 --> 00:01:51.560 Which brings us to the core mission of our deep 33 00:01:51.599 --> 00:01:54.840 dive today. For you listening, we are going to unpack 34 00:01:54.920 --> 00:01:58.319 how scientists plan to build Earth sized telescopes without actually 35 00:01:58.400 --> 00:02:00.719 building a single Earth sized physical object. 36 00:02:00.879 --> 00:02:03.280 It is a complete paradigm shift, and I really don't 37 00:02:03.359 --> 00:02:05.480 use that phrase lightly. We're looking at a proposal that 38 00:02:05.519 --> 00:02:09.680 effectively suggests replacing the physical transportation of gathered starlight with 39 00:02:10.000 --> 00:02:11.120 quantum teleportation. 40 00:02:11.400 --> 00:02:15.240 Quantum teleportation, that is the exact headline that stopped me 41 00:02:15.240 --> 00:02:17.840 in my tracks. It sounds like science fiction or a 42 00:02:17.840 --> 00:02:21.520 comic book. Movie plot. But this is serious, peer reviewed 43 00:02:21.599 --> 00:02:25.719 academic work. The paper is titled super Resolution Imaging with 44 00:02:26.000 --> 00:02:27.759 Entanglement Enhanced Telescopy. 45 00:02:28.120 --> 00:02:30.960 And if this protocol works, we aren't just talking about 46 00:02:30.960 --> 00:02:33.879 getting a slightly sharper picture of a distant star cluster. 47 00:02:34.280 --> 00:02:38.240 We are talking about decoupling the resolution of a telescope 48 00:02:38.280 --> 00:02:41.039 from its physical size and location, entirely. 49 00:02:40.759 --> 00:02:44.400 Creating a virtual giant using the laws of quantum mechanics. 50 00:02:44.960 --> 00:02:47.520 So before we get to the quantum magic, and it 51 00:02:47.560 --> 00:02:50.240 really does feel like magic, we need to rigorously understand 52 00:02:50.280 --> 00:02:53.240 the status quo. Why can't we just keep building bigger mirrors, 53 00:02:53.479 --> 00:02:57.840 or more specifically, why is the current method of combining 54 00:02:57.919 --> 00:03:00.960 multiple telescopes so incredibly difficult. 55 00:03:00.719 --> 00:03:02.800 To understand that we have to start with the concept 56 00:03:02.840 --> 00:03:07.159 of long baseline interferometry. This is currently the gold standard 57 00:03:07.199 --> 00:03:09.280 for high resolution imaging in astronomy. 58 00:03:09.319 --> 00:03:12.039 Okay, let's unpack that definition for the listener. What is 59 00:03:12.199 --> 00:03:14.719 long baseline interferometry technically speaking? 60 00:03:14.960 --> 00:03:17.960 Right, So, in its technical application, interferometry is the practice 61 00:03:17.960 --> 00:03:21.639 of combining electromagnetic signals, which in this context is visible 62 00:03:21.680 --> 00:03:26.599 starlight collected by multiple spatially separated telescopes. So imagine you 63 00:03:26.639 --> 00:03:27.919 have telescope A and. 64 00:03:27.960 --> 00:03:30.919 Telescope B, and they are separated by a good distance, right, 65 00:03:30.960 --> 00:03:32.159 say a kilometer. 66 00:03:32.080 --> 00:03:35.919 Exactly separated by one kilometer. Now separately, there are just 67 00:03:35.960 --> 00:03:38.759 two modest instruments. But if you can take the light 68 00:03:38.840 --> 00:03:42.080 caught by telescope A and a light cought by telescope 69 00:03:42.120 --> 00:03:45.039 B and physically combine those light beams at a central point, 70 00:03:45.560 --> 00:03:47.479 you trick the physics. You create what we call an 71 00:03:47.520 --> 00:03:48.439 effective aperture. 72 00:03:48.560 --> 00:03:52.280 An effective aperture, so the universe essentially behaves as if 73 00:03:52.280 --> 00:03:55.400 it's being observed by one giant mirror that spans that 74 00:03:55.599 --> 00:03:57.680 entire one kilometer distance between them. 75 00:03:57.759 --> 00:04:01.599 Precisely, you achieve the resolution the sharpness of a telescope 76 00:04:01.599 --> 00:04:04.840 with a one kilometer wide mirror, even though physically you 77 00:04:04.960 --> 00:04:08.039 mostly just have empty dirt and air between the two facilities. 78 00:04:08.479 --> 00:04:12.159 The baseline is that distance between the furthest individual telescopes 79 00:04:12.199 --> 00:04:15.120 in your array, the longer the baseline, the sharper the 80 00:04:15.159 --> 00:04:16.360 image you can reconstruct. 81 00:04:16.560 --> 00:04:18.199 Now this sounds a lot like how we got that 82 00:04:18.240 --> 00:04:20.439 famous image of the black hole a few years back, 83 00:04:20.519 --> 00:04:23.839 right the event horizon telescope that was basically an earth 84 00:04:23.959 --> 00:04:26.079 sized virtual telescope that is a. 85 00:04:26.120 --> 00:04:29.160 Very crucial comparison to make, but there is a major 86 00:04:29.240 --> 00:04:33.480 catch there. The event horizon telescope use radio waves. Radio 87 00:04:33.480 --> 00:04:37.199 waves are long, lazy wavelengths. They are roughly a millimeter 88 00:04:37.360 --> 00:04:40.279 or more long, and because they are so slow and 89 00:04:40.360 --> 00:04:43.279 large in a physical sense, you can literally record the 90 00:04:43.319 --> 00:04:46.519 incoming wave data onto hard drives at each telescope. 91 00:04:46.600 --> 00:04:48.879 Oh, I see, you just stamp a timecode on the data. 92 00:04:48.959 --> 00:04:51.839 Exactly, you stamp a GPS timecode on it, put the 93 00:04:51.879 --> 00:04:54.040 hard drives in a FedEx box, shift them to a 94 00:04:54.079 --> 00:04:57.439 supercomputer facility, and combine the waves digitally. Months later. 95 00:04:57.560 --> 00:04:59.279 You can just record the wave and play back. But 96 00:04:59.279 --> 00:05:02.560 we are talking about optical astronomy today visible light. 97 00:05:02.759 --> 00:05:06.120 Right with visible light, the wavelengths are tiny. We are 98 00:05:06.160 --> 00:05:09.959 talking hundreds of nanometers. They oscillate at frequencies of hundreds 99 00:05:10.000 --> 00:05:13.120 of terror herds. You absolutely cannot record the wave of 100 00:05:13.160 --> 00:05:16.120 a single photon on a hard drive and combine it later. 101 00:05:16.560 --> 00:05:19.560 The technology to digitize a LightWave at that speed simply 102 00:05:19.560 --> 00:05:20.319 doesn't exist. 103 00:05:20.560 --> 00:05:23.920 So to do enterferometry with visible light, you have to 104 00:05:23.920 --> 00:05:27.199 combine the actual physical light waves in real time. 105 00:05:27.439 --> 00:05:29.959 You do, which means if you have a telescope in 106 00:05:29.959 --> 00:05:33.279 Arizona and a telescope in Maryland. You literally have to 107 00:05:33.319 --> 00:05:37.399 transport the photon from Arizona to a central lab and 108 00:05:37.480 --> 00:05:40.600 the photon from Maryland to that same lab without those 109 00:05:40.639 --> 00:05:42.600 photons dying or getting out of sink. 110 00:05:42.959 --> 00:05:45.279 And that brings us to the primary villain of our 111 00:05:45.319 --> 00:05:49.000 story today, the physical optical link, or as the source 112 00:05:49.040 --> 00:05:53.279 material frames it in engineering terms, the beam splitter bottleneck. 113 00:05:53.959 --> 00:05:57.279 It is an absolute nightmare for engineers. Just think about 114 00:05:57.279 --> 00:06:00.519 the logistics involved. You need to transport a f fragile 115 00:06:00.600 --> 00:06:04.759 optical beam over miles of terrain. Usually this is done 116 00:06:04.959 --> 00:06:08.680 using vacuum tubes, literally burying light pipes in the ground 117 00:06:08.759 --> 00:06:11.439 containing a series of perfect mirrors to bounce the light along. 118 00:06:11.920 --> 00:06:14.639 Or alternatively, you try to use fiber optic cables. 119 00:06:14.879 --> 00:06:16.839 But wait, we use fiber optics for the Internet all 120 00:06:16.879 --> 00:06:18.639 the time. I can send an email to Australia and 121 00:06:18.639 --> 00:06:20.680 it gets there in milliseconds. Why can't I just send 122 00:06:20.720 --> 00:06:23.560 a star photon to Australia through the same cable. 123 00:06:23.839 --> 00:06:26.959 Because when you send an email, you are sending classical 124 00:06:27.040 --> 00:06:30.240 digital pulses, a bright flash for a one, a dim 125 00:06:30.240 --> 00:06:32.839 flash for a zero. If that signal gets weak after 126 00:06:32.879 --> 00:06:36.199 fifty kilometers of fiber. A repeater station reads the one 127 00:06:36.279 --> 00:06:40.480 or zero boosts it creates a fresh, loud copy and 128 00:06:40.560 --> 00:06:41.480 sends it on its way. 129 00:06:42.439 --> 00:06:45.879 But you can't copy a photon's quantum state bingo. 130 00:06:46.560 --> 00:06:49.839 The no cloning theorem in quantum mechanics strictly forbids it. 131 00:06:50.319 --> 00:06:53.720 You cannot amplify or copy an unknown quantum state without 132 00:06:53.720 --> 00:06:57.160 destroying the original information encoded within it. And in this case, 133 00:06:57.439 --> 00:07:00.399 the specific phase information of the starlight is exactly which 134 00:07:00.399 --> 00:07:02.439 you need to preserve for the interferometry to work. 135 00:07:02.759 --> 00:07:05.319 So you are forced to send the original exact photon 136 00:07:05.360 --> 00:07:07.920 that hit the telescope mirror. You have to guide that 137 00:07:08.000 --> 00:07:11.480 one fragile little packet of energy through kilometers of glass, 138 00:07:11.480 --> 00:07:12.199 fiber or. 139 00:07:12.199 --> 00:07:15.480 Vacuum tube, and no material in the universe is perfectly 140 00:07:15.480 --> 00:07:19.000 transparent over a few kilometers. The glass simply absorbs or 141 00:07:19.040 --> 00:07:21.800 scatters the photon. It's a process called attenuation. 142 00:07:22.279 --> 00:07:24.519 Ah So by the time the light from two very 143 00:07:24.560 --> 00:07:27.439 distant telescopes actually meets in the middle lab, you've lost 144 00:07:27.439 --> 00:07:29.639 so much signal that there's nothing left to combine. 145 00:07:29.800 --> 00:07:33.240 The signal is effectively gone. The transport loss places a 146 00:07:33.279 --> 00:07:36.480 hard physical cap on how big our telescope array can be. 147 00:07:37.079 --> 00:07:40.360 We simply can't build a ten kilometer optical array using 148 00:07:40.360 --> 00:07:43.480 classical methods because the light dies in the pipe before 149 00:07:43.519 --> 00:07:45.160 it ever reaches the mixing station. 150 00:07:45.160 --> 00:07:48.399 Which explains why current arrays like the Chara array or 151 00:07:48.399 --> 00:07:51.360 the VLTI are limited to baselines of just a few 152 00:07:51.439 --> 00:07:55.000 hundred meters. We want to go to kilometers or even thousands. 153 00:07:54.560 --> 00:07:58.639 Of kilometers, and conventional interferometry simply cannot scale to those distances. 154 00:07:58.879 --> 00:08:02.079 The transport losses are two high, and the engineering complexity 155 00:08:02.079 --> 00:08:05.319 of maintaining phased ability keeping the light waves perfectly in 156 00:08:05.399 --> 00:08:09.279 sync in a vacuum tube that long is functionally impossible. 157 00:08:09.399 --> 00:08:12.160 So we are stuck. We desperately want the resolution of 158 00:08:12.199 --> 00:08:14.600 a giant telescope, but we can't transport the light to 159 00:08:14.600 --> 00:08:17.600 combine it. This is exactly where the research from Padilla, 160 00:08:17.759 --> 00:08:21.879 Guha and Sogdod comes. In their paper super Resolution Imaging 161 00:08:21.920 --> 00:08:25.920 with entanglement Enhanced Telescopy proposes a radical solution to the 162 00:08:26.000 --> 00:08:27.079 light pipe problem. 163 00:08:27.199 --> 00:08:30.360 Their solution is brilliant in its sheer audacity. They propose 164 00:08:30.439 --> 00:08:32.480 that we simply don't transport the light at all. 165 00:08:32.639 --> 00:08:35.360 Okay, but if I don't transport the light to a 166 00:08:35.360 --> 00:08:39.360 central hub. I can't physically interfere the beams. And if 167 00:08:39.399 --> 00:08:41.480 I can't interfere the beams, I don't get the high 168 00:08:41.519 --> 00:08:46.759 resolution image. I mean, that is the entire definition of interferometry. 169 00:08:45.960 --> 00:08:49.240 In classical physics. Yes, you are completely correct, but this 170 00:08:49.279 --> 00:08:54.000 team is applying quantum information theory to optics. They are 171 00:08:54.039 --> 00:08:58.200 positing a fundamental foundational shift in how we think about 172 00:08:58.200 --> 00:08:58.919 gathering light. 173 00:08:59.240 --> 00:09:01.519 They are saying we should stop treating light like a 174 00:09:01.559 --> 00:09:03.960 classical wave that needs to be guided through a plumbing 175 00:09:03.960 --> 00:09:05.639 system of mirrors and tubes. 176 00:09:06.000 --> 00:09:09.120 Exactly, they are suggesting we treat the light strictly as 177 00:09:09.159 --> 00:09:12.360 a quantum object that carries information. This is a very 178 00:09:12.360 --> 00:09:15.919 important distinction that doctor Psychott Guha makes in the foundational material. 179 00:09:16.519 --> 00:09:20.440 He rigorously separates quantum optics from quantum information theory. 180 00:09:20.600 --> 00:09:23.720 Right, let's look at that distinction. He defines quantum optics 181 00:09:23.759 --> 00:09:27.120 as the quantum theory of light itself, studying how photons 182 00:09:27.159 --> 00:09:29.039 physically interact with atoms and mirrors. 183 00:09:29.360 --> 00:09:33.759 Yes, whereas he defines quantum information theory or qiit as 184 00:09:33.799 --> 00:09:37.879 the mathematical science of quantifying the actual information carried by 185 00:09:37.879 --> 00:09:43.360 those inherently quantum physical media Qit asks an entirely different 186 00:09:43.440 --> 00:09:45.360 question than classical optics. 187 00:09:45.600 --> 00:09:49.039 It asks what is the absolute theoretical maximum amount of 188 00:09:49.080 --> 00:09:52.159 data I can extract from this quantum state, regardless of 189 00:09:52.200 --> 00:09:53.919 the physical hardware I used to measure it. 190 00:09:54.279 --> 00:09:57.600 Precisely, they're moving away from a hardware problem, how do 191 00:09:57.679 --> 00:10:01.399 I build a longer clear glass pipe, to a fundamental 192 00:10:01.440 --> 00:10:05.320 data processing problem. They're looking strictly at nature's fundamental limits 193 00:10:05.360 --> 00:10:06.200 regarding information. 194 00:10:06.440 --> 00:10:09.240 Let's talk about those limits. In classical optical theory, astronomers 195 00:10:09.240 --> 00:10:11.840 are always battling something called the Railey limit or the 196 00:10:11.840 --> 00:10:12.919 diffraction limit. 197 00:10:12.720 --> 00:10:15.120 The blur limit. If you look at two stars that 198 00:10:15.159 --> 00:10:18.320 are extremely close together in the sky, your telescope just 199 00:10:18.559 --> 00:10:22.600 is one blurry, elongated blob of light. You mathematically cannot 200 00:10:22.600 --> 00:10:24.039 tell if it's one star or two. 201 00:10:24.120 --> 00:10:26.759 And for centuries we just accepted that blur as an 202 00:10:26.919 --> 00:10:30.120 unbreakable law of physics, a limitation of the universe. 203 00:10:30.480 --> 00:10:33.519 But the researchers at the Center for Quantum Networks argue 204 00:10:33.519 --> 00:10:35.840 that the Railey limit is actually just a bad habit. 205 00:10:36.080 --> 00:10:39.279 It is a limitation of our specific receivers, our lenses 206 00:10:39.320 --> 00:10:43.120 and camera sensors, not a limitation of the light itself. 207 00:10:43.000 --> 00:10:46.519 Meaning the information about the two distinct stars is actually 208 00:10:46.879 --> 00:10:49.240 there in the light hitting the mirror, we just aren't 209 00:10:49.279 --> 00:10:50.200 catching it properly. 210 00:10:50.440 --> 00:10:53.440 That is exactly the case. The photons arriving at your 211 00:10:53.440 --> 00:10:58.159 telescope contain the full necessary quantum information required to distinguish 212 00:10:58.240 --> 00:11:02.840 those two stars. Do we smash those delicate photons onto 213 00:11:02.840 --> 00:11:04.240 a focal plane camera. 214 00:11:04.000 --> 00:11:07.000 Sensor which just measures brightness intensity? 215 00:11:07.080 --> 00:11:09.360 Right, a camera pixel just counts how many photons hit it. 216 00:11:09.759 --> 00:11:13.440 In doing so, it aggressively destroys the delicate quantum information, 217 00:11:13.720 --> 00:11:17.399 the phase, the spatial distribution. We're essentially using a blunt 218 00:11:17.399 --> 00:11:20.639 instrument to measure something incredibly fragile. 219 00:11:20.200 --> 00:11:23.080 So we need a sharper instrument, And the paper introduces 220 00:11:23.080 --> 00:11:26.000 a concept called the quantum Fisher information mandate. 221 00:11:26.360 --> 00:11:30.759 The quantum Fisher information or QFI is the theoretical ceiling 222 00:11:30.799 --> 00:11:35.000 of knowledge. It represents the absolute maximum amount of information 223 00:11:35.080 --> 00:11:38.120 you can possibly extract from a quantum system about a 224 00:11:38.120 --> 00:11:39.120 specific parameter. 225 00:11:39.399 --> 00:11:43.559 And in this astrophysical context, that parameter is the angular 226 00:11:43.559 --> 00:11:46.799 separation of the two stars, the tiny distance between them 227 00:11:46.799 --> 00:11:47.240 in the sky. 228 00:11:47.519 --> 00:11:51.039 Correct, the researchers proved mathematically that if you build a 229 00:11:51.080 --> 00:11:55.159 receiver that respects the QFI mandate. A receiver that extracts 230 00:11:55.240 --> 00:11:58.600 the quantum information rather than just the classical intensity. The 231 00:11:58.639 --> 00:12:00.960 traditional RAILI limit effectctively disappears. 232 00:12:01.200 --> 00:12:04.759 That is staggering. You can resolve objects that astronomers previously 233 00:12:04.799 --> 00:12:07.240 considered entirely unseeable. 234 00:12:06.879 --> 00:12:10.639 Provided you collect enough data over time. Yes, standard receiver 235 00:12:10.759 --> 00:12:14.399 techniques are totally suboptimal. Quantum limits allow for a resolution 236 00:12:14.480 --> 00:12:17.840 that makes the traditional diffraction limit look like a historical artifact. 237 00:12:18.000 --> 00:12:21.279 Okay, so that covers the theoretical framework. The theory says 238 00:12:21.320 --> 00:12:24.159 we can see the unseeable if we stop using standard 239 00:12:24.159 --> 00:12:27.519 cameras and start processing quantum information. But I want to 240 00:12:27.519 --> 00:12:30.360 get into the actual mechanics of this. If we aren't 241 00:12:30.440 --> 00:12:33.799 using light pipes to connect the observatories, how do the 242 00:12:33.799 --> 00:12:36.919 two distant telescopes actually talk to each other to form 243 00:12:36.960 --> 00:12:37.600 an image. 244 00:12:37.840 --> 00:12:41.600 This brings us to the architecture of entanglement enhanced telescopy. 245 00:12:42.480 --> 00:12:47.559 The proposed receiver design eliminates the physical optical link entirely. Instead, 246 00:12:47.919 --> 00:12:51.159 two or more distant telescopes share a unified quantum state. 247 00:12:51.639 --> 00:12:55.639 And this architecture relies on three major cutting edge components. 248 00:12:55.919 --> 00:13:00.879 A spatial mode sorter atomic, quantum memories, and teleportation via entanglement. 249 00:13:01.279 --> 00:13:03.799 Let's break those down sequentially, starting with the first thing 250 00:13:03.840 --> 00:13:06.879 the starlight hits after the primary mirror component One is 251 00:13:06.919 --> 00:13:09.639 the spatial mode sorder or Spady. 252 00:13:09.559 --> 00:13:12.639 Spade replaces the traditional camera sensor at the focal plane. 253 00:13:12.720 --> 00:13:15.279 But looking at the diagrams and the source material specifically 254 00:13:15.360 --> 00:13:17.480 Figure one B, it really doesn't operate anything like a 255 00:13:17.519 --> 00:13:18.240 standard camera. 256 00:13:18.600 --> 00:13:21.679 It doesn't think about what a traditional camera does. It 257 00:13:21.759 --> 00:13:25.279 asks a very simple question, where did the photon land 258 00:13:25.320 --> 00:13:28.320 on the grid? It maps X and y coordinates to 259 00:13:28.360 --> 00:13:31.720 create an image. A spatial mode sorder asks a completely 260 00:13:31.720 --> 00:13:35.600 different question. It asks what is the specific geometric shape 261 00:13:35.759 --> 00:13:37.039 of the photon's wavefront? 262 00:13:37.240 --> 00:13:40.120 The shape of the wavefront. When I think of a photon, 263 00:13:40.159 --> 00:13:43.080 I usually just think of a tiny, localized point of light, 264 00:13:43.759 --> 00:13:44.399 a particle. 265 00:13:44.559 --> 00:13:47.039 In quantum optics, you have to think of light arriving 266 00:13:47.080 --> 00:13:51.039 from a distant source as having a spatial distribution. It 267 00:13:51.120 --> 00:13:54.919 exists as a wave that excites specific spatial modes. I 268 00:13:55.000 --> 00:13:57.000 find a musical analogy works best here. 269 00:13:57.080 --> 00:13:58.559 Oh, I like that, Let's hear it. 270 00:13:58.639 --> 00:14:00.879 Think of the light entering the telescope aperture as a 271 00:14:00.919 --> 00:14:04.440 complex musical chord. A normal camera center just acts like 272 00:14:04.480 --> 00:14:06.879 a simple decibel meter. It just hears loud noise and 273 00:14:06.879 --> 00:14:08.039 measures the total volume. 274 00:14:08.159 --> 00:14:10.399 It just measures the intensity of the light exactly. 275 00:14:10.480 --> 00:14:12.960 But the spade device acts like a highly trained musician 276 00:14:13.000 --> 00:14:16.200 with perfect pitch. It listens to that same chord, breaks 277 00:14:16.200 --> 00:14:18.519 the noise down and says that sound is composed of 278 00:14:18.519 --> 00:14:21.120 a C sharp an E and a G it D 279 00:14:21.320 --> 00:14:25.600 multiplexes the incoming photon into orthogonal spatial patterns prior to detection. 280 00:14:26.000 --> 00:14:29.840 Can we visualize these spatial patterns? What do these modes 281 00:14:29.879 --> 00:14:30.799 actually look like? 282 00:14:31.000 --> 00:14:34.840 Imagine pointing a laser pointer at a wall. The bright, 283 00:14:34.960 --> 00:14:37.679 perfectly round dot in the center is what we call 284 00:14:37.720 --> 00:14:39.039 the fundamental mode. 285 00:14:39.000 --> 00:14:40.960 Just a single solid circle of light. 286 00:14:41.320 --> 00:14:45.759 Right now, imagine a different shape. Imagine two distinct lobes 287 00:14:45.759 --> 00:14:49.000 of light side by side with a dark empty line 288 00:14:49.080 --> 00:14:51.480 right down the middle. It looks a bit like a dumbbell. 289 00:14:52.159 --> 00:14:55.200 That is a higher order spatial mode. Or imagine a 290 00:14:55.240 --> 00:14:57.360 pattern that looks like a four leaf clover, and the 291 00:14:57.360 --> 00:15:00.600 starlet actually naturally takes on these complex shape. The wave 292 00:15:00.600 --> 00:15:03.279 function the light does. Yes, when you have a single 293 00:15:03.320 --> 00:15:06.919 perfectly centered star, the light mostly arrives in that simple 294 00:15:07.159 --> 00:15:11.399 fundamental circular mode. But when two stars are very close together, 295 00:15:11.480 --> 00:15:14.480 closer than the Railey limit, their combined light field is 296 00:15:14.519 --> 00:15:16.120 slightly displaced. 297 00:15:15.600 --> 00:15:18.879 And that tiny physical displacement alters the wave exactly. 298 00:15:18.919 --> 00:15:22.120 The displacement causes the incoming light to excite those higher 299 00:15:22.200 --> 00:15:25.679 order shapes, the Dumbbell mode or the Clover mode. The 300 00:15:25.720 --> 00:15:28.960 Spade device is designed to physically separate these different shapes 301 00:15:29.000 --> 00:15:30.320 into different output channels. 302 00:15:30.559 --> 00:15:32.840 So if your detector sitting at the end of the 303 00:15:32.919 --> 00:15:36.559 Dumbell channel registers of photon, you know instantly that the 304 00:15:36.639 --> 00:15:38.759 source isn't just a single centered star. 305 00:15:39.120 --> 00:15:41.320 You know instantly that there is a second star there, 306 00:15:41.720 --> 00:15:45.480 or that the singular source is slightly off center. Unlike 307 00:15:45.519 --> 00:15:48.440 a camera that just records a blurry blob, the Spade 308 00:15:48.480 --> 00:15:52.159 device analyzes the specific mode to extract a much richer 309 00:15:52.240 --> 00:15:55.759 layer of information about the angular separation of the source. 310 00:15:56.279 --> 00:15:59.840 Knowing which mode was excited tells you exponentially more about 311 00:15:59.879 --> 00:16:02.480 the target than just knowing where a dot landed on 312 00:16:02.519 --> 00:16:05.440 a CCD chip. That is how they begin to reach 313 00:16:05.480 --> 00:16:07.320 that quantum Fisher information. 314 00:16:06.960 --> 00:16:10.320 Limit correct, So that is step one. Telescope A catches 315 00:16:10.320 --> 00:16:12.639 a photon and runs it through the Spady sorder which 316 00:16:12.679 --> 00:16:15.519 identifies the spatial mode. Say it's the Dumbbell mode in 317 00:16:15.559 --> 00:16:16.440 time bin number five. 318 00:16:16.519 --> 00:16:18.320 But now we hit a wall again. You have this 319 00:16:18.519 --> 00:16:21.440 vital piece of mode information at Telescope A in Arizona, 320 00:16:21.840 --> 00:16:24.320 but you can't just text or email that data to 321 00:16:24.320 --> 00:16:28.679 Telescope B in Maryland, right because true inefrometry requires comparing 322 00:16:28.720 --> 00:16:31.759 the quantum phase of the light, and any classical measurement 323 00:16:31.759 --> 00:16:33.120 you make destroys that phase. 324 00:16:33.480 --> 00:16:35.679 Right. If you measure the photon right then and there 325 00:16:35.720 --> 00:16:39.240 to see its phase, you collapse its wave function. You 326 00:16:39.320 --> 00:16:43.080 permanently lose the quantum correlation with the light hitting the 327 00:16:43.120 --> 00:16:43.840 other telescope. 328 00:16:43.879 --> 00:16:47.080 So you have to somehow save that fragile quantum state 329 00:16:47.200 --> 00:16:49.399 until you can properly compare it with the state at 330 00:16:49.399 --> 00:16:51.000 the other facility. 331 00:16:50.559 --> 00:16:54.799 Which brings us to component two atomic quantum memories. You 332 00:16:54.919 --> 00:16:56.600 need a storage. 333 00:16:56.159 --> 00:16:59.000 Medium, essentially a hard drive for starlight. 334 00:16:58.720 --> 00:17:01.320 A quantum hard drive. You have to take that fragile 335 00:17:01.320 --> 00:17:03.639 photonic state that just came out of the Spade device 336 00:17:03.960 --> 00:17:07.319 and map it onto an atomic structure. The source material 337 00:17:07.400 --> 00:17:11.640 specifically focuses on using atomic quantum memories such as silicon 338 00:17:11.720 --> 00:17:13.359 vacancies in a diamond lattice. 339 00:17:13.440 --> 00:17:16.680 Silicon vacancies. These are engineered flaws inside a diamond. 340 00:17:16.839 --> 00:17:21.200 Right, Yes, they are highly specific, artificially created defects. These 341 00:17:21.279 --> 00:17:24.519 vacancies have free electrons that act as quibots. They have 342 00:17:24.559 --> 00:17:27.279 a quantum spin state that can be precisely manipulated by 343 00:17:27.359 --> 00:17:28.559 lasers and microwaves. 344 00:17:28.920 --> 00:17:31.119 How do you get the starlight into the diamond though? 345 00:17:31.160 --> 00:17:34.359 The process involves routing the sordied photon into the memory 346 00:17:34.759 --> 00:17:37.680 using what is called a photon memory cnot gate. 347 00:17:37.839 --> 00:17:42.480 A cnot gate that is a logic operation from quantum computing, 348 00:17:42.799 --> 00:17:44.880 controlled not exactly. 349 00:17:45.240 --> 00:17:48.680 It operates as a fundamental quantum logic gait. The system 350 00:17:48.720 --> 00:17:51.440 takes the incoming photon and forces it to interact with 351 00:17:51.480 --> 00:17:55.039 the electron and the diamond defect. Through this interaction, the 352 00:17:55.160 --> 00:17:58.480 quantum state of the photon, its phase, its mode information 353 00:17:59.000 --> 00:18:01.839 is effectively mapped onto the spin state of that local 354 00:18:01.880 --> 00:18:02.640 memory quibit. 355 00:18:02.839 --> 00:18:05.799 So the original starlight photon is absorbed and gone, but 356 00:18:05.880 --> 00:18:10.160 its exact quantum information, it's ghost basically is now safely 357 00:18:10.200 --> 00:18:13.200 trapped in the electrons spin of this diamond atom. 358 00:18:13.240 --> 00:18:17.640 At telescope A precisely and Concurrently, the exact same process 359 00:18:17.680 --> 00:18:20.319 is happening with a different photon from the same star 360 00:18:20.519 --> 00:18:21.440 over at telescope B. 361 00:18:21.640 --> 00:18:24.319 So now you have two diamonds sitting in labs potentially 362 00:18:24.359 --> 00:18:27.279 thousands of kilometers apart, and each one is holding half 363 00:18:27.319 --> 00:18:29.599 of the quantum information from the star system. 364 00:18:29.680 --> 00:18:32.160 And now you face the ultimate challenge. How do you 365 00:18:32.200 --> 00:18:34.920 combine the information in those two diamonds without moving them 366 00:18:35.119 --> 00:18:37.079 and without a physical wire connecting them. 367 00:18:37.119 --> 00:18:39.319 This is the real cliffhanger of the architecture, and the 368 00:18:39.319 --> 00:18:42.920 answer is component three teleportation via entanglement. 369 00:18:43.359 --> 00:18:47.480 Before the astronomical observation even begins, the system operators have 370 00:18:47.559 --> 00:18:51.640 to predistribute entangled pairs of quibbits to both telescope sites. 371 00:18:52.160 --> 00:18:55.880 Spooky action at a distance, as Einstein called it. 372 00:18:55.400 --> 00:18:59.160 It serves as the invisible bridge. Station A holds one 373 00:18:59.279 --> 00:19:02.599 half of an intangled pair and station B holds the 374 00:19:02.640 --> 00:19:06.680 other half. As doctor Guha notes, in the theoretical framework, 375 00:19:07.039 --> 00:19:10.720 quantum entanglement is a correlation that is fundamentally stronger than 376 00:19:10.759 --> 00:19:14.160 any probabilistic correlation allowed by classical physics. 377 00:19:14.400 --> 00:19:16.480 So how does the bridge actually work. You have the 378 00:19:16.519 --> 00:19:19.480 starlight memory and you have the entangled memory. Sitting next 379 00:19:19.480 --> 00:19:20.240 to it, you. 380 00:19:20.119 --> 00:19:24.079 Execute a teleportation protocol. At each telescope site, you perform 381 00:19:24.079 --> 00:19:27.480 a very specific joint measurement, an X basis measurement on 382 00:19:27.599 --> 00:19:30.440 both the local diamond memory holding the starlight and the 383 00:19:30.519 --> 00:19:32.839 local half of the entangled pair simultaneously. 384 00:19:32.920 --> 00:19:35.599 You measure them together. Does that transport the starlight to 385 00:19:35.640 --> 00:19:36.200 the other lab. 386 00:19:36.319 --> 00:19:39.640 It doesn't transport the physical atom or a photon, but 387 00:19:39.720 --> 00:19:43.640 it mimics the exact mathematical result of physical optical interference. 388 00:19:44.279 --> 00:19:47.440 By performing these measurements at both sites and then sharing 389 00:19:47.480 --> 00:19:51.160 the results of those measurements over a standard classical internet connection. 390 00:19:51.480 --> 00:19:54.319 Wait, a normal internet connection the classical channel mentioned in 391 00:19:54.359 --> 00:19:55.640 the outline, Yes. 392 00:19:55.559 --> 00:19:59.559 A standard fiber optic internet line. By combining the local 393 00:19:59.599 --> 00:20:02.480 measurem data with the data sent over the classical channel, 394 00:20:02.920 --> 00:20:06.559 the supercomputer can reconstruct exactly what would have happened if 395 00:20:06.599 --> 00:20:09.920 you had physically overlapped the two starlight beams in a 396 00:20:10.000 --> 00:20:10.680 central lab. 397 00:20:10.920 --> 00:20:12.759 I want to make sure the listener really grasps this, 398 00:20:12.880 --> 00:20:16.839 because it is mind bending. The physical starlight from telescope 399 00:20:16.880 --> 00:20:19.799 A never touches the starlight from telescope B. 400 00:20:20.039 --> 00:20:20.279 Never. 401 00:20:20.400 --> 00:20:23.960