WEBVTT - Small Tech, Big Deal 0:00:04.400 --> 0:00:07.800 Welcome to text Stuff, a production from I Heart Radio. 0:00:12.039 --> 0:00:14.800 Hey there, and welcome to tech Stuff. I'm your host, 0:00:15.000 --> 0:00:18.360 Jonathan Strickland. I'm an executive producer with I Heart Radio 0:00:18.560 --> 0:00:22.160 and I love all things tech. And if you guys 0:00:22.200 --> 0:00:24.960 have listened to tech Stuff or any real length of time, 0:00:25.400 --> 0:00:28.840 you know that I talk a lot about how miniaturization 0:00:29.200 --> 0:00:33.360 really changed everything. You could argue it fueled a new 0:00:33.400 --> 0:00:39.680 industrial revolution. So in the early twentieth century, technology like radios, televisions, 0:00:39.680 --> 0:00:43.240 and computers were all bigger because they had to be, 0:00:43.240 --> 0:00:47.400 because the internal components inside these technologies, the things that 0:00:47.880 --> 0:00:53.280 made these technologies work, were themselves much larger. That's why 0:00:53.400 --> 0:00:55.600 you would buy a television that had a tiny ten 0:00:55.720 --> 0:00:59.320 inch screen housed inside a cabinet large enough to be 0:00:59.480 --> 0:01:03.320 a full piece of furniture. The invention of the transistor 0:01:03.480 --> 0:01:06.759 would lead to manturization, and in less than one years 0:01:06.760 --> 0:01:09.200 we would find ourselves holding a device in our hands 0:01:09.640 --> 0:01:13.319 that was vastly more powerful than the massive computers that 0:01:13.360 --> 0:01:17.399 took up entire floors of buildings back in the day. 0:01:17.440 --> 0:01:21.000 But what if we keep going down that path. What 0:01:21.080 --> 0:01:25.360 if we were to mantorize things even more? What if 0:01:25.400 --> 0:01:28.800 we could get technology down to a scale so small 0:01:29.319 --> 0:01:32.160 that it would be too tiny for us to see. 0:01:32.600 --> 0:01:37.040 What have we conquered the nano scale? So in today's episode, 0:01:37.280 --> 0:01:41.560 I'm going to explain what nanotechnology is all about and 0:01:41.600 --> 0:01:44.679 how the idea evolved, and a bit about where we 0:01:44.720 --> 0:01:47.840 are now. We'll also talk about how stuff gets really 0:01:48.320 --> 0:01:52.280 weird when you get really small, which I think any 0:01:52.320 --> 0:01:55.640 toddler would attest to, but I mean, it gets really 0:01:55.760 --> 0:01:59.920 weird when you get really small. In fact, if you 0:02:00.080 --> 0:02:04.080 want to get super lucy goosey with the term nanotechnology, 0:02:04.120 --> 0:02:06.800 it gives us a chance to talk about those weird 0:02:06.840 --> 0:02:13.320 things right now. But first a definition. Technically speaking, nanotechnology 0:02:13.480 --> 0:02:17.120 encompasses tech that is on a size scale of one 0:02:17.400 --> 0:02:22.840 d nanometers or smaller down to one nanometer. A nanometer 0:02:23.280 --> 0:02:27.320 is one billionth of a meter. A strand of human 0:02:27.440 --> 0:02:33.160 hair ranges between eighty thousand and one hundred thousand nanometers 0:02:33.200 --> 0:02:36.240 in width. So if you take one of your hairs 0:02:36.680 --> 0:02:39.480 because you can't take mine I'm bald, and you were 0:02:39.520 --> 0:02:42.720 to hold your hair and look at how wide that 0:02:42.880 --> 0:02:46.919 strand of hair is, not how long? How wide? That's 0:02:47.280 --> 0:02:51.679 eight thousand to ten thousand times wider than what we're 0:02:51.720 --> 0:02:56.800 talking about here. In addition, we often think of nanotechnology 0:02:56.880 --> 0:03:00.079 today as being a branch of science and tech that 0:03:00.240 --> 0:03:03.799 is exploring the possibility of manipulating matter on the molecular 0:03:04.000 --> 0:03:07.840 or even atomic scale. The classic example of this in 0:03:07.880 --> 0:03:11.480 science fiction is the universal Assembler, a device that can 0:03:11.520 --> 0:03:16.800 construct macro sized objects atom by atom or molecule by molecule. 0:03:17.320 --> 0:03:20.079 And we'll cover those in more detail a little bit later, 0:03:20.120 --> 0:03:22.960 but this is sort of how the replicators are on 0:03:23.080 --> 0:03:25.880 Star Trek are supposed to work. Right, You say t 0:03:26.160 --> 0:03:30.120 earl gray hot, and then the device takes all the 0:03:30.200 --> 0:03:33.400 atoms that are necessary to make that, puts them together 0:03:33.639 --> 0:03:36.960 right there when you're waiting, and boom, you have t 0:03:37.320 --> 0:03:39.880 on demand. But we don't have to wait until the 0:03:39.920 --> 0:03:44.440 twenty second century to talk about our work in nanotechnology. 0:03:44.480 --> 0:03:47.280 In fact, we can go back more than three thousand 0:03:47.440 --> 0:03:51.320 years ago in China and talk about lamp black. Now, 0:03:51.360 --> 0:03:55.520 this material is a byproduct of burning oil, typically a 0:03:55.640 --> 0:04:00.120 coal based oil, and burning oil in a shallow and 0:04:00.680 --> 0:04:04.040 where you produce really heavy smoke was the typical production 0:04:04.080 --> 0:04:07.440 method for lamp black. You would use a collection pan 0:04:07.880 --> 0:04:11.760 that you would put near the flame, and the collection 0:04:11.800 --> 0:04:16.000 pan begins to accumulate very small particles of carbon they 0:04:16.000 --> 0:04:20.360 are deposited on that pan. Some of those particles are 0:04:20.400 --> 0:04:25.919 around twenty nanometers in diameter. So the lamp black has 0:04:26.000 --> 0:04:31.440 a pretty phenomenal surface area to volume ratio, right, because 0:04:31.880 --> 0:04:36.360 the particles are very small, So there's more of the 0:04:36.400 --> 0:04:39.960 surface of the particle exposed to the air than there 0:04:40.160 --> 0:04:43.120 is under the surface. One way to think about this 0:04:43.160 --> 0:04:46.560 is if you have a gold brick that has a 0:04:46.600 --> 0:04:49.599 certain amount of surface that's exposed to the air. Right, 0:04:49.839 --> 0:04:52.760 But if you were to make that gold brick into 0:04:52.839 --> 0:04:56.280 gold foil, right, if you were to flatten it out 0:04:56.640 --> 0:04:59.520 so that it's much much, much much wider, much much longer, 0:04:59.560 --> 0:05:02.400 but very e than well, now, way more of the 0:05:02.440 --> 0:05:06.160 surface of that gold is exposed to the outside world 0:05:06.520 --> 0:05:09.760 as a much larger amount of surface area compared to 0:05:09.839 --> 0:05:12.599 its volume. Well, that's kind of how things are on 0:05:12.600 --> 0:05:17.160 the nano scale. Nanoparticles have way more surface area exposed 0:05:17.200 --> 0:05:20.400 to the world compared to their volume than stuff that's 0:05:20.440 --> 0:05:22.960 on the macro scale. All right, let's get back to 0:05:23.040 --> 0:05:26.960 lamp black. So because of this amazing amount of surface area, 0:05:27.120 --> 0:05:31.200 it became a very popular black pigment for inks. And paints. 0:05:31.600 --> 0:05:33.240 You didn't need a lot of it in order to 0:05:33.320 --> 0:05:36.640 be able to cover a surface. Well, if it's black, 0:05:36.920 --> 0:05:40.120 then you could use that to be an ink. Right 0:05:40.520 --> 0:05:43.960 centuries later, this same sort of stuff, which we typically 0:05:44.040 --> 0:05:47.880 now call carbon black, is used in all kinds of applications, 0:05:47.960 --> 0:05:52.600 including printer toner. So even to this day we're using 0:05:52.640 --> 0:05:56.400 the same sort of stuff, these tiny, tiny particles of carbon. 0:05:57.279 --> 0:06:00.760 Way back in the three as in the four century 0:06:00.800 --> 0:06:05.040 Common era, some Roman artisan crafted a cup made out 0:06:05.080 --> 0:06:08.919 of glass. Now that in itself isn't incredibly special, but 0:06:08.960 --> 0:06:11.800 this particular cup had a really cool quality to it. 0:06:12.160 --> 0:06:15.039 So let's say you had the glass sitting on a 0:06:15.080 --> 0:06:18.600 table and you placed an oil lamp in front of 0:06:18.720 --> 0:06:22.000 the glass, so it's between you and the glass. Then 0:06:22.120 --> 0:06:24.440 to you, the glass would appear to be green. But 0:06:24.520 --> 0:06:26.560 let's say you position the glass so that the light 0:06:26.600 --> 0:06:30.400 from the oil lamp was actually going into the glass 0:06:30.520 --> 0:06:33.920 rather than onto it. Well, now the glass would appear 0:06:33.960 --> 0:06:38.120 to be red. The color of the glass changes depending 0:06:38.200 --> 0:06:42.479 upon how light hits it. Now today we call the 0:06:42.480 --> 0:06:45.720 glass by the name the Lakergus cup, and you can 0:06:45.760 --> 0:06:48.160 see the Lakergus cup if you ever go into the 0:06:48.200 --> 0:06:52.960 British Museum. Maybe not right now, but you know, things 0:06:53.000 --> 0:06:54.760 being what they are, but when things get better, you 0:06:54.800 --> 0:06:57.200 could see it there. That's where the cup is. It's 0:06:57.200 --> 0:07:00.400 called the Lakergus cup because the figure on the cup 0:07:00.560 --> 0:07:04.000 is that of King like Urgus. He's being dragged into 0:07:04.040 --> 0:07:08.200 the underworld by the nymph ambrosia. So that's fun. So 0:07:08.320 --> 0:07:10.680 why does the glass change color and what does it 0:07:10.760 --> 0:07:14.000 have to do with nanotechnology? The answers had to wait 0:07:14.080 --> 0:07:18.040 more than fifteen hundred years before we really sussed it out. 0:07:18.640 --> 0:07:23.840 In n scientists used an atomic force microscope more on 0:07:24.080 --> 0:07:27.080 those later to examine the like Hurgis cup, and they 0:07:27.120 --> 0:07:33.120 found that this glass contained extremely tiny particles of copper, gold, 0:07:33.160 --> 0:07:37.560 and silver. The particles were in the nanoscale range, and 0:07:37.560 --> 0:07:40.120 they were mixed in with the glass itself. The red 0:07:40.200 --> 0:07:44.280 light came from gold's absorption of light. The glass was 0:07:44.760 --> 0:07:50.520 a type of nanocomposite material. In the following centuries, glassmakers 0:07:50.560 --> 0:07:54.080 would experiment by adding different types of metals to glass 0:07:54.120 --> 0:07:57.320 mixtures to produce various colors of glass a k A 0:07:57.680 --> 0:08:01.000 stained glass. But while we were are able to grind 0:08:01.080 --> 0:08:03.760 stuff down to such a fine powder as to have 0:08:04.000 --> 0:08:08.800 individual particles on the nano scale suspended in glass. It 0:08:08.840 --> 0:08:12.360 wasn't like we were building machines at that same scale. 0:08:12.480 --> 0:08:16.400 That would have been unthinkable. In fact, I would argue 0:08:16.440 --> 0:08:20.080 that before the transistor, most folks weren't really thinking about 0:08:20.120 --> 0:08:24.360 going small with technology. Before the electronic era, we were 0:08:24.400 --> 0:08:28.600 building mechanical systems, and generally the power of machines scaled 0:08:28.960 --> 0:08:32.240 with their size. You could do stuff with gear ratios 0:08:32.280 --> 0:08:34.960 to help boost output without making an entire piece of 0:08:35.000 --> 0:08:39.000 technology bigger, but that only worked down to a certain point. 0:08:39.600 --> 0:08:43.160 In the early era of computers, even as we moved 0:08:43.200 --> 0:08:47.680 from the electro mechanical systems to pure electronic ones, the 0:08:47.760 --> 0:08:51.360 general thought was that the more powerful machines of tomorrow 0:08:51.800 --> 0:08:55.040 would be at least the same size, if not larger, 0:08:55.240 --> 0:08:59.160 than the behemoths of that era. Mantorization was something that 0:08:59.760 --> 0:09:03.720 most people just didn't really anticipate. That, by the way, 0:09:04.240 --> 0:09:06.400 is something that we should keep in mind whenever we 0:09:06.440 --> 0:09:09.920 make any predictions about the future, is that frequently things 0:09:09.960 --> 0:09:13.640 we don't anticipate will end up being a much larger 0:09:14.200 --> 0:09:18.000 influence on the way technology develops than what is currently 0:09:18.080 --> 0:09:22.080 going on. So in the nineteen twenties, if you were 0:09:22.120 --> 0:09:24.640 predicting what the future of technology was going to be, 0:09:25.120 --> 0:09:29.439 You probably weren't thinking in terms of electronics. That was unanticipated. 0:09:29.960 --> 0:09:33.480 And just like if we project out now, we say 0:09:33.760 --> 0:09:35.520 was it going to be like fifty years from now? 0:09:36.080 --> 0:09:38.840 If we're basing it on the technologies we're using right 0:09:38.880 --> 0:09:41.680 at this moment, chances are we're going to miss something 0:09:42.320 --> 0:09:44.920 because it's something we haven't early anticipated that's going to 0:09:45.000 --> 0:09:49.360 change everything between now and then. Okay, anyway, in our 0:09:49.400 --> 0:09:53.040 history we get up to nineteen forty seven when William Shockley, 0:09:53.280 --> 0:09:57.680 Walter Britain, and John Bardine, among others, developed the first 0:09:57.800 --> 0:10:01.000 transistor in a T and T S research and development 0:10:01.000 --> 0:10:04.840 division that would be Bell Labs. The transistor could step 0:10:04.880 --> 0:10:07.600 in and do the job that was previously performed by 0:10:07.720 --> 0:10:11.320 larger components like vacuum tubes. Vacuum tubes are still in 0:10:11.400 --> 0:10:16.120 use today, but the transistors largely replaced them in many technologies. 0:10:16.440 --> 0:10:20.640 So early transistors were large and impractical for any real application. 0:10:21.040 --> 0:10:24.640 They were, you know, a demonstration of a scientific principle, 0:10:24.760 --> 0:10:28.200 but you wouldn't actually use them for something like a radio. However, 0:10:28.840 --> 0:10:32.280 it did prove that the science underlying the transistors was sound, 0:10:32.440 --> 0:10:34.600 and it was only a matter of time before companies 0:10:34.640 --> 0:10:38.760 began to refine the technology and built smaller transistors and 0:10:38.800 --> 0:10:42.199 develop new manufacturing processes to do so at a scale 0:10:42.440 --> 0:10:45.480 large enough for them to be actually be useful. We'll 0:10:45.520 --> 0:10:49.640 get to a famous observation that Gordon Moore made because 0:10:49.800 --> 0:10:52.720 of this particular trend in a little bit, but there's 0:10:52.760 --> 0:10:55.800 another person that I need to talk about first. In 0:10:55.920 --> 0:11:00.679 nineteen fifty nine, physicist Richard Feynman gave a presentation at 0:11:00.679 --> 0:11:05.040 the American Physical Society at the California Institute of Technology 0:11:05.080 --> 0:11:09.240 also known as cal Tech. He called the presentation There's 0:11:09.320 --> 0:11:13.200 plenty of room at the bottom. It would retroactively become 0:11:13.320 --> 0:11:17.720 one of the foundational arguments in support of nanotechnology, the 0:11:17.840 --> 0:11:20.920 discipline and the pursuit of it. Now, it helps if 0:11:20.960 --> 0:11:24.320 we understand how things had developed by the time Feinman 0:11:24.440 --> 0:11:28.560 gave this talk. It took centuries for humans to develop 0:11:28.600 --> 0:11:31.560 technologies that allowed us to observe the world of the 0:11:31.720 --> 0:11:37.760 very small. From magnifying glasses to microscopes, we gradually peeled 0:11:37.800 --> 0:11:41.400 back the unknown, and we kept finding, to our amazement 0:11:41.800 --> 0:11:46.360 that things could get even smaller. But light based or 0:11:46.559 --> 0:11:52.520 optical microscopes have fundamental limitations that are dictated by physics. 0:11:53.080 --> 0:11:56.319 It's not because the limitations of the materials we used. 0:11:56.320 --> 0:11:59.600 It's not that we couldn't find clearer lenses or anything. 0:12:00.080 --> 0:12:03.400 It's rather due to the fact that light waves themselves 0:12:03.400 --> 0:12:07.439 have limitations. Now, we can see stuff because light bounces 0:12:07.520 --> 0:12:11.280 off of it, and light waves are very short. They 0:12:11.320 --> 0:12:16.679 are tiny, but they're not as tiny as say, individual atoms. 0:12:17.280 --> 0:12:20.320 Light waves are too big to reflect off of stuff 0:12:20.360 --> 0:12:24.600 as small as atoms and most molecules, and so no 0:12:24.640 --> 0:12:28.840 matter how good your optical microscope is, you're not going 0:12:28.880 --> 0:12:32.400 to be able to resolve images at that smallest scale 0:12:32.920 --> 0:12:36.320 just because you're using light. Typically you'd be relying on 0:12:36.440 --> 0:12:41.640 light with wavelengths of between four hundred and seven hundred nanometers. 0:12:41.640 --> 0:12:46.520 But that's way larger than stuff like proteins or some viruses, 0:12:46.640 --> 0:12:50.160 and and way way larger than atoms. If you want 0:12:50.200 --> 0:12:53.719 to observe these smaller things, you got to shed your 0:12:53.760 --> 0:12:58.920 dependence on light. Back in nineteen twenty six, a German 0:12:59.000 --> 0:13:04.600 scientist named Hans Bush developed the first electromagnetic lens. This 0:13:04.800 --> 0:13:07.079 isn't the same sort of lens you would find in 0:13:07.240 --> 0:13:11.319 eyeglasses or a telescope or a microscope. Instead, it was 0:13:11.360 --> 0:13:14.960 a couple of electro magnets which could generate a magnetic 0:13:15.080 --> 0:13:19.680 field sufficient to direct a beam of magnetically charged particles. 0:13:20.280 --> 0:13:23.920 This is the same sort of idea used in particle accelerators. 0:13:23.960 --> 0:13:27.320 In a particle accelerator, you've got these big, powerful magnets 0:13:27.360 --> 0:13:31.760 that create an extremely narrow channel through which charged particles 0:13:31.800 --> 0:13:34.360 can travel. They can't go outside of it because of 0:13:34.360 --> 0:13:38.240 these magnetic forces, and it guides the particles around a 0:13:38.320 --> 0:13:41.400 pathway so that they can collide with something else, such 0:13:41.400 --> 0:13:44.240 as a beam of charged particles that are traveling in 0:13:44.280 --> 0:13:48.760 the opposite direction. Now, Bush proposed using the lens to 0:13:48.880 --> 0:13:52.360 make a microscope that would use electrons rather than light, 0:13:52.840 --> 0:13:56.640 and electromagnetic coils rather than a glass lens. He even 0:13:56.679 --> 0:14:00.400 patented a design, but he never constructed the electron micro scope. 0:14:01.000 --> 0:14:05.280 Max Knell, an electrical engineer, and Ernst risk a physicist, 0:14:05.720 --> 0:14:09.600 did build one in ninety one, though this early version 0:14:10.120 --> 0:14:12.720 wasn't able to produce an image that was at a 0:14:12.800 --> 0:14:16.199 higher resolution than what you could achieve with an optical microscope. 0:14:16.559 --> 0:14:20.240 Those would come not that much longer down the road. However, 0:14:20.960 --> 0:14:23.880 the sample that you're looking at has to be inside 0:14:23.960 --> 0:14:27.640 a vacuum chamber, because air molecules would be like giant 0:14:27.680 --> 0:14:31.280 obstacles to an electron beam, and you wouldn't look at 0:14:31.320 --> 0:14:33.840 it through an eyepiece, you know, it's not like that 0:14:33.880 --> 0:14:37.480 type of microscope. Instead, you would capture the interactions of 0:14:37.480 --> 0:14:40.920 the electron beam with the sample you're examining on either 0:14:41.040 --> 0:14:45.120 special photographic film or later on a monitor. So typically 0:14:45.160 --> 0:14:47.480 you would have a sensor and then the sensor would 0:14:47.480 --> 0:14:52.360 send data that you would then interpret visually through a monitor. 0:14:53.000 --> 0:14:55.520 By the time Feynman gave his presentation in the late 0:14:55.600 --> 0:14:59.400 nineteen fifties, electron microscopes could produce images at a much 0:14:59.440 --> 0:15:04.480 smaller scale than optical microscopes. What scientists had learned from 0:15:04.600 --> 0:15:08.440 mathematics was actually beginning to bear out through observation. So 0:15:08.480 --> 0:15:12.760 sometimes we discover stuff because mathematically we understand that it 0:15:12.840 --> 0:15:15.480 has to be a certain way, even if we can't 0:15:15.560 --> 0:15:18.760 directly observe that way. That was kind of what was 0:15:18.800 --> 0:15:21.400 going on. We had sort of sussed out how the 0:15:21.440 --> 0:15:24.040 world had to be at that scale, and now we 0:15:24.080 --> 0:15:26.960 could actually directly observe it and learn even more. We 0:15:27.040 --> 0:15:30.480 appeared to be on the cusp of another major breakthrough. 0:15:31.080 --> 0:15:34.359 The crux of Fineman's presentation was all about the manipulation 0:15:34.400 --> 0:15:37.560 and controlling of the world on the small scale. He 0:15:37.640 --> 0:15:40.720 started off by talking about the possibility of printing something 0:15:40.800 --> 0:15:43.640 like a full encyclopedia onto the head of a pen. 0:15:44.080 --> 0:15:47.600 Then he elaborated from there. He talked about the possibility 0:15:47.600 --> 0:15:51.760 of printing twenty four million books, which he estimated to 0:15:51.760 --> 0:15:55.240 be about the number of notable books ever written, and 0:15:55.480 --> 0:15:58.840 printing them onto the equivalent of thirty five sheets of 0:15:58.920 --> 0:16:02.880 paper by making the print that tiny. His point was 0:16:03.160 --> 0:16:05.840 all about scale, that the scale of things we deal 0:16:05.880 --> 0:16:09.280 with in our everyday lives is gargantuan compared to what 0:16:09.320 --> 0:16:11.840 we could study with the help of powerful technologies like 0:16:11.920 --> 0:16:16.240 electron microscopes. He went on to hypothesize that if we 0:16:16.240 --> 0:16:20.920 were to develop a means of manipulating single atoms, you 0:16:21.000 --> 0:16:25.440 could encode information using some form of simple system. He 0:16:25.520 --> 0:16:28.480 likened it to the dots and dashes in Morse code, 0:16:28.880 --> 0:16:31.240 and you could use it in a three dimensional space 0:16:31.360 --> 0:16:34.400 for each character, and it would measure five by five 0:16:34.440 --> 0:16:38.240 by five atoms to a bit of information, and even 0:16:38.280 --> 0:16:41.760 while using additional atoms for separation, you could print the 0:16:41.800 --> 0:16:45.240 equivalent of those twenty four million volumes on a particle 0:16:45.360 --> 0:16:49.360 the size of a moat of dust. Feynman then goes 0:16:49.400 --> 0:16:52.920 on to suggest even more radical ideas, including using evaporation 0:16:52.960 --> 0:16:56.600 to reduce materials down to their smallest components, before then 0:16:56.680 --> 0:17:00.640 depositing those materials onto a substrate to build out wires 0:17:00.720 --> 0:17:03.960 and insulation and entire circuits. This way. Now, this is 0:17:04.000 --> 0:17:06.840 pretty similar to how we would make stuff like computer 0:17:06.920 --> 0:17:10.880 chips in the future, once we got all those technologies 0:17:10.920 --> 0:17:13.919 down to work on the nano scale. Fineman goes on 0:17:14.000 --> 0:17:16.880 in his presentation to propose the possibility that we could 0:17:16.880 --> 0:17:21.160 build mechanical systems at the nano scale, using the example 0:17:21.240 --> 0:17:24.159 of an automobile, saying how would it be possible to 0:17:24.160 --> 0:17:27.879 build that on this very tiny scale? He argued that 0:17:28.000 --> 0:17:31.240 such a thing might be hypothetically possible, but it would 0:17:31.280 --> 0:17:34.679 require some big changes in automobile design, and a tiny 0:17:34.720 --> 0:17:39.280 scale heat would dissipate much faster than at the macro scale. Again, 0:17:39.280 --> 0:17:42.640 you've got an incredible amount of surface area compared to volume, 0:17:43.040 --> 0:17:46.280 so an internal combustion engine wouldn't work. You wouldn't be 0:17:46.280 --> 0:17:49.640 able to get combustion. You would need some other sort 0:17:49.680 --> 0:17:53.919 of reaction to provide the energy needed to do work. Ultimately, 0:17:54.240 --> 0:17:56.800 Feynman suggested we might find a way to build such 0:17:56.920 --> 0:18:00.399 small devices as to be able to assemble matter atom 0:18:00.560 --> 0:18:04.800 by atom, building with precision on an atomic level, and 0:18:05.280 --> 0:18:10.399 that could create countless possible applications, including being able to 0:18:10.440 --> 0:18:14.560 synthesize chemicals, which previously we had to do through chemical synthesis, 0:18:15.119 --> 0:18:20.120 which is not necessarily as precise and fascinating idea. I'll 0:18:20.160 --> 0:18:22.400 talk a bit more about it, but first let's take 0:18:22.600 --> 0:18:34.480 a quick break. Fineman imagined a macro sized tool that 0:18:34.520 --> 0:18:38.479 could make essentially the parts to replicate itself, but on 0:18:38.520 --> 0:18:42.000 a much smaller scale. So imagine using a tool like 0:18:42.160 --> 0:18:44.840 a lathe to cut out all the parts for a 0:18:44.920 --> 0:18:48.560 smaller version of the lathe. Then you use this smaller 0:18:48.680 --> 0:18:52.040 lathe to cut out even smaller parts for an even 0:18:52.080 --> 0:18:55.840 smaller lathe, and so on, and then using these tiny 0:18:55.920 --> 0:19:00.399 tools to produce what Fineman called tiny hands to a symbol, 0:19:00.680 --> 0:19:04.320 very small components. But then he said, we'd start to 0:19:04.400 --> 0:19:08.520 encounter some challenges that don't exist in any appreciable way 0:19:08.680 --> 0:19:11.920 on the macro scale. For example, once you get down 0:19:12.000 --> 0:19:15.400 to the molecular level, you begin to encounter forces that 0:19:15.520 --> 0:19:19.440 you just don't notice at larger scales, forces like the 0:19:19.520 --> 0:19:24.600 Vanderwall's forces. These are electric forces that attract neutral molecules 0:19:24.600 --> 0:19:28.320 to one another. They are pretty weak forces, but when 0:19:28.320 --> 0:19:31.199 you get down to the molecular level, the forces are 0:19:31.240 --> 0:19:34.960 strong enough to cause issues. So he said, if you 0:19:35.000 --> 0:19:38.000 were to create the equivalent of a nut and bolt 0:19:38.320 --> 0:19:41.720 at the nano scale, you would find the Vanderwall's force 0:19:41.880 --> 0:19:44.520 strong enough that you would have trouble turning the nut 0:19:45.080 --> 0:19:47.760 like it would be difficult to tighten or loosen it 0:19:47.840 --> 0:19:51.800 because it would be clinging to the bolt due to 0:19:51.840 --> 0:19:55.320 the Vanderwall's force between the two. Now that's just the beginning. 0:19:55.359 --> 0:19:57.399 Of course. When you get down to the nano scale, 0:19:57.440 --> 0:19:59.880 you start to enter into a world governed more by 0:20:00.040 --> 0:20:03.040 quantum mechanics then the classical physics that you and I 0:20:03.119 --> 0:20:07.320 encounter in our day to day lives. Weird stuff starts 0:20:07.359 --> 0:20:10.320 to happen. At least it's weird to us because we 0:20:10.400 --> 0:20:14.399 don't observe the world working in that way on our scale. So, 0:20:14.440 --> 0:20:19.159 for example, there's the truly weird phenomena of quantum tunneling. 0:20:19.720 --> 0:20:22.320 I'll try to explain this as best I can. So 0:20:22.400 --> 0:20:25.879 let's start with the classical world, because we generally have 0:20:25.920 --> 0:20:29.160 a pretty good handle on that. Imagine you have a 0:20:29.200 --> 0:20:32.200 toy car, like a little matchbox car, and you've set 0:20:32.280 --> 0:20:36.040 up a ramp, and you probably understand that unless you 0:20:36.200 --> 0:20:39.840 push the toy car hard enough, it's not gonna make 0:20:39.880 --> 0:20:42.720 it up that ramp. It's not going to spontaneously go 0:20:42.840 --> 0:20:46.160 forward and climb that ramp. If you push too soft, 0:20:46.680 --> 0:20:48.480 then it's going to start going up the ramp and 0:20:48.480 --> 0:20:51.919 then roll back down. So the potential energy of the 0:20:52.040 --> 0:20:55.680 ramp is a certain level. You have to give enough 0:20:55.760 --> 0:20:59.000 kinetic energy to the toy car so it can overcome 0:20:59.040 --> 0:21:02.159 the potential energy g represented by the height and the 0:21:02.640 --> 0:21:05.680 and the slope of the ramp. Now let's say we're 0:21:05.720 --> 0:21:10.240 doing something similar, except instead of a little toy car 0:21:10.440 --> 0:21:14.600 and a ramp, we've got an electron and an electrical field. 0:21:15.000 --> 0:21:18.080 If the energy of the electron is higher than the 0:21:18.200 --> 0:21:21.280 energy level of the electric field, the electron can pass 0:21:21.320 --> 0:21:24.399 through it. But if the electric fields energy is higher, 0:21:24.560 --> 0:21:27.280 the electron will be repelled, just as the toy car 0:21:27.600 --> 0:21:29.919 would roll backward down the ramp if you didn't give 0:21:29.960 --> 0:21:33.520 it a hard enough push. But there's a tiny little problem. 0:21:33.640 --> 0:21:36.920 You see, at the quantum level, we're not talking in absolutes, 0:21:37.440 --> 0:21:41.960 we're actually talking in probabilities. Heisenberg's uncertainty principle explained that 0:21:42.119 --> 0:21:45.080 we'll never know the precise position and momentum of a 0:21:45.119 --> 0:21:48.600 particle like an electron. We can only know a little 0:21:48.600 --> 0:21:50.800 bit about each and then we can work out the 0:21:50.800 --> 0:21:54.199 probability that a given sub atomic particle is in a 0:21:54.240 --> 0:21:57.560 certain position at any given time. So you can actually 0:21:57.640 --> 0:22:01.199 plot this out in a wave function. The peak of 0:22:01.240 --> 0:22:06.240 the wave corresponds with the most likely outcomes, the places 0:22:06.320 --> 0:22:09.800 where the electron is most probably going to be located 0:22:09.880 --> 0:22:13.160 at a given time, but there will be a small 0:22:13.280 --> 0:22:16.680 chance that the electron will appear somewhere else. And if 0:22:16.720 --> 0:22:20.199 the wave function can actually overlap the entirety of the 0:22:20.240 --> 0:22:23.560 electric field, that means that there's a tiny little amount 0:22:23.680 --> 0:22:26.520 of that probability wave on the opposite side of the 0:22:26.520 --> 0:22:30.560 electric field where the electron could exist. The probability is 0:22:30.640 --> 0:22:34.600 very small, but it is there, which means it is 0:22:34.680 --> 0:22:37.040 possible the electron is on the other side of the 0:22:37.040 --> 0:22:40.800 electric field. And if something is possible, then if you 0:22:41.320 --> 0:22:44.000 do that something enough times it means it will happen. 0:22:44.600 --> 0:22:48.479 It probably doesn't happen frequently. The probability tells us it won't, 0:22:49.040 --> 0:22:52.200 but if there is a chance it will happen sooner 0:22:52.280 --> 0:22:55.520 or later, it will. Now, there's a lot more to 0:22:55.600 --> 0:22:59.600 this stuff, like the discussion of evanescent waves, but while 0:22:59.600 --> 0:23:03.240 those make me wake up inside, they are also super 0:23:03.240 --> 0:23:07.000 tricky to explain without visual aids. The important thing for 0:23:07.080 --> 0:23:10.320 us to remember is that if there is a probability 0:23:10.359 --> 0:23:14.639 that something will happen, if you have enough instances, you 0:23:14.680 --> 0:23:17.840 will eventually encounter that. And if that something means an 0:23:17.840 --> 0:23:21.480 electron suddenly appears on the opposite side of a barrier 0:23:21.720 --> 0:23:24.760 where it's not supposed to be, you gotta deal with that. 0:23:25.200 --> 0:23:27.879 So what this means for us in practical terms is 0:23:27.920 --> 0:23:31.080 that if we build stuff down at the nanoscale, we 0:23:31.280 --> 0:23:35.200 have to worry about things like quantum tunneling. So imagine 0:23:35.200 --> 0:23:39.000 you've got an electric circuit with all the components small 0:23:39.119 --> 0:23:42.080 enough that the wave function of the electron means that 0:23:42.240 --> 0:23:44.760 sometimes the electron can be on the other side of 0:23:44.840 --> 0:23:48.439 gates or even in a totally different wire. Well that 0:23:48.640 --> 0:23:51.000 those gates and those wires are meant to control the 0:23:51.040 --> 0:23:54.320 flow of electrons. That's what circuits are. Circuits really are 0:23:54.600 --> 0:23:59.960 controlled pathways for electrical signals, and the important part there 0:24:00.080 --> 0:24:02.320 is the control. If it's uncontrolled, you might as well 0:24:02.320 --> 0:24:05.560 not even have a circuit. So if electrons can just 0:24:05.960 --> 0:24:08.520 appear on the other side of gates as if those 0:24:08.560 --> 0:24:11.440 gates were open, or jump from one wire to the next, 0:24:12.000 --> 0:24:15.240 you've got a problem. You can't actually control electricity in 0:24:15.240 --> 0:24:18.679 a reliable way, you'll start to get errors. Now this 0:24:18.760 --> 0:24:22.040 is something microchip manufacturers actually have to deal with today 0:24:22.080 --> 0:24:26.080 because they keep scaling down. The components on their chips 0:24:26.359 --> 0:24:30.960 were happidly in the five nanometer range at this point, 0:24:30.960 --> 0:24:33.520 which is smaller than I ever thought we would ever 0:24:33.600 --> 0:24:36.760 be able to go, and their talks about possibly getting 0:24:36.760 --> 0:24:40.080 as low as three nanometers or beyond. But we really 0:24:40.080 --> 0:24:43.680 have to answer some big questions about fundamental quantum mechanics 0:24:43.720 --> 0:24:46.359 problems in order to get there. So what the heck 0:24:46.359 --> 0:24:48.760 does this mean if we were to blow it out 0:24:48.760 --> 0:24:52.160 to macro scale. Well, in our example with the toy car, 0:24:52.760 --> 0:24:55.400 it would mean that sometimes, let's just say you give 0:24:55.440 --> 0:24:58.960 the toy car a gentle push, most of the time 0:24:59.320 --> 0:25:01.040 it would just go a little bit up the ramp 0:25:01.240 --> 0:25:04.480