WEBVTT 00:00:00.000 --> 00:00:03.923 align:middle line:84% [SQUEAKING] [RUSTLING] [CLICKING] 00:00:03.923 --> 00:00:11.760 align:middle line:90% 00:00:11.760 --> 00:00:15.760 align:middle line:84% SCOTT KEMP: A week ago, we left off halfway 00:00:15.760 --> 00:00:18.680 align:middle line:84% through this section on proliferation, 00:00:18.680 --> 00:00:25.680 align:middle line:84% and this is the last of the nuclear-specific externalities, 00:00:25.680 --> 00:00:28.480 align:middle line:84% of the two main externalities that nuclear 00:00:28.480 --> 00:00:34.500 align:middle line:84% has to deal with are safety and a special kind of safety. 00:00:34.500 --> 00:00:35.740 align:middle line:90% Everything has safety issues. 00:00:35.740 --> 00:00:41.200 align:middle line:84% But there's large accident, rare event kind of safety, 00:00:41.200 --> 00:00:44.360 align:middle line:84% which we talked about, and nuclear proliferation. 00:00:44.360 --> 00:00:49.360 align:middle line:84% And so we were talking about how might a state 00:00:49.360 --> 00:00:51.160 align:middle line:90% go about acquiring a bomb. 00:00:51.160 --> 00:00:54.400 align:middle line:84% And we went through the plutonium route. 00:00:54.400 --> 00:00:56.920 align:middle line:84% And remember, we talked about different weapon designs 00:00:56.920 --> 00:00:58.680 align:middle line:90% and how you get plutonium. 00:00:58.680 --> 00:01:00.940 align:middle line:84% And then we were about to get into the uranium route, 00:01:00.940 --> 00:01:03.800 align:middle line:90% and then we ran out of time. 00:01:03.800 --> 00:01:05.840 align:middle line:84% So here let's continue with that. 00:01:05.840 --> 00:01:08.340 align:middle line:84% This is just to remind you that natural uranium out 00:01:08.340 --> 00:01:11.740 align:middle line:84% of the ground only has 7/10 of a percent, 235, 00:01:11.740 --> 00:01:16.100 align:middle line:84% and you have to do this process of getting rid of the 238 atoms, 00:01:16.100 --> 00:01:18.300 align:middle line:84% which is what we call enrichment, 00:01:18.300 --> 00:01:22.180 align:middle line:84% to get to something that's weapons-usable. 00:01:22.180 --> 00:01:24.060 align:middle line:84% So this is our navigation graphic 00:01:24.060 --> 00:01:27.740 align:middle line:84% that told us the different pathways to a bomb. 00:01:27.740 --> 00:01:31.047 align:middle line:90% You could have some enrichment. 00:01:31.047 --> 00:01:32.880 align:middle line:84% You could take spent fuel and get plutonium. 00:01:32.880 --> 00:01:35.920 align:middle line:84% You could get plutonium from regularly reprocessed fuel. 00:01:35.920 --> 00:01:39.220 align:middle line:84% You could build a dedicated plutonium production facility, 00:01:39.220 --> 00:01:40.840 align:middle line:84% or you can use uranium enrichment. 00:01:40.840 --> 00:01:43.660 align:middle line:84% And you can either do it by taking it 00:01:43.660 --> 00:01:46.060 align:middle line:84% straight off the civilian fuel cycle 00:01:46.060 --> 00:01:49.140 align:middle line:90% or having a dedicated facility. 00:01:49.140 --> 00:01:51.260 align:middle line:84% Historically, we haven't really worried too much 00:01:51.260 --> 00:01:54.460 align:middle line:84% about uranium enrichment because this 00:01:54.460 --> 00:01:56.740 align:middle line:90% was perceived to be very hard. 00:01:56.740 --> 00:02:00.120 align:middle line:84% And we're going to talk a little bit about that now. 00:02:00.120 --> 00:02:04.760 align:middle line:84% So just to talk a little bit about how much enrichment you 00:02:04.760 --> 00:02:10.800 align:middle line:84% need, this is the critical mass for an explosive quantity 00:02:10.800 --> 00:02:15.240 align:middle line:84% for standard metal density uranium with 5-centimeter 00:02:15.240 --> 00:02:16.920 align:middle line:90% beryllium reflector. 00:02:16.920 --> 00:02:21.320 align:middle line:84% And you can see that anything basically 00:02:21.320 --> 00:02:23.440 align:middle line:84% in this red region, which is what 00:02:23.440 --> 00:02:27.640 align:middle line:84% we call HEU, everything above 20%, 00:02:27.640 --> 00:02:32.760 align:middle line:90% has reasonable critical masses. 00:02:32.760 --> 00:02:36.960 align:middle line:84% And at 20% uncompressed, it's about half a ton, 00:02:36.960 --> 00:02:42.240 align:middle line:84% but half a ton is that much uranium. 00:02:42.240 --> 00:02:44.640 align:middle line:84% And as you go to lower, yes, the number 00:02:44.640 --> 00:02:49.800 align:middle line:84% goes way up and reaches a ton of material. 00:02:49.800 --> 00:02:53.300 align:middle line:84% But a ton has 18-inch diameter sphere. 00:02:53.300 --> 00:02:57.510 align:middle line:84% So yeah, it does have this knee in the curve, 00:02:57.510 --> 00:03:03.510 align:middle line:84% but there's a bit of a debate as to exactly where this should be. 00:03:03.510 --> 00:03:06.490 align:middle line:84% Just for reference, most reactors, 00:03:06.490 --> 00:03:10.750 align:middle line:84% in case you're not a nuclear engineer, use under 5%. 00:03:10.750 --> 00:03:13.330 align:middle line:84% Way down here, this dashed line is 6%. 00:03:13.330 --> 00:03:16.910 align:middle line:84% There's the asymptote, the theoretical limit where 00:03:16.910 --> 00:03:18.910 align:middle line:90% you need an infinite quantity. 00:03:18.910 --> 00:03:22.630 align:middle line:84% But there are research reactors like the MIT reactor 00:03:22.630 --> 00:03:27.150 align:middle line:90% that operate up here at 90%. 00:03:27.150 --> 00:03:32.870 align:middle line:84% But the MIT reactor only has, I think, 12 kilograms in its core. 00:03:32.870 --> 00:03:35.430 align:middle line:84% So that's only about half of what 00:03:35.430 --> 00:03:40.670 align:middle line:84% you need for a critical mass at normal density. 00:03:40.670 --> 00:03:43.870 align:middle line:84% So the idea is ideally we don't want 00:03:43.870 --> 00:03:48.070 align:middle line:84% weapon quantities of this material sitting in one place. 00:03:48.070 --> 00:03:49.830 align:middle line:84% Of course, you could go and steal 00:03:49.830 --> 00:03:52.590 align:middle line:84% the fuel from MITR and another reactor, 00:03:52.590 --> 00:03:54.970 align:middle line:84% or while the MITR was being refueled. 00:03:54.970 --> 00:03:58.050 align:middle line:84% And the old fuel was there, and the new fuel was there. 00:03:58.050 --> 00:03:59.710 align:middle line:84% You could do various things like this. 00:03:59.710 --> 00:04:03.490 align:middle line:90% So the security isn't perfect. 00:04:03.490 --> 00:04:03.990 align:middle line:90% All right. 00:04:03.990 --> 00:04:06.290 align:middle line:90% So enrichment is in general-- 00:04:06.290 --> 00:04:09.010 align:middle line:84% in the abstract, you start with some materials, always 00:04:09.010 --> 00:04:10.650 align:middle line:90% this three-stream process. 00:04:10.650 --> 00:04:11.730 align:middle line:90% You have some. 00:04:11.730 --> 00:04:13.510 align:middle line:90% Here's a black box device. 00:04:13.510 --> 00:04:15.010 align:middle line:84% You don't really know how it works, 00:04:15.010 --> 00:04:18.450 align:middle line:84% but you put in some material that's like in this example, 1% 00:04:18.450 --> 00:04:19.630 align:middle line:90% uranium-235. 00:04:19.630 --> 00:04:22.450 align:middle line:84% And you get out a depleted stream, which we call the tails, 00:04:22.450 --> 00:04:25.330 align:middle line:84% and then an enriched stream, which we call the product. 00:04:25.330 --> 00:04:29.130 align:middle line:84% And the key thing to note here is that the changes 00:04:29.130 --> 00:04:30.650 align:middle line:90% are very subtle. 00:04:30.650 --> 00:04:32.410 align:middle line:84% Some technologies are better than others 00:04:32.410 --> 00:04:35.770 align:middle line:90% in producing larger changes. 00:04:35.770 --> 00:04:38.570 align:middle line:84% But there's a trade-off between the amount 00:04:38.570 --> 00:04:39.750 align:middle line:90% that you can flow through. 00:04:39.750 --> 00:04:42.850 align:middle line:84% And the amount of that enrichment change. 00:04:42.850 --> 00:04:46.410 align:middle line:84% And generally, though, you wind up 00:04:46.410 --> 00:04:49.370 align:middle line:84% having to have a whole bunch of these things connected together 00:04:49.370 --> 00:04:53.770 align:middle line:84% in a cascade, if you want to get to any meaningful level, 00:04:53.770 --> 00:04:55.350 align:middle line:90% including fuel levels. 00:04:55.350 --> 00:05:00.190 align:middle line:84% So what goes in this black box, or in this case, white box? 00:05:00.190 --> 00:05:02.410 align:middle line:84% Lots of technologies can be used. 00:05:02.410 --> 00:05:05.950 align:middle line:84% Here are some of the early ones, according 00:05:05.950 --> 00:05:10.910 align:middle line:84% to the history of the field, more or less in order 00:05:10.910 --> 00:05:12.670 align:middle line:90% of their arrival. 00:05:12.670 --> 00:05:15.430 align:middle line:84% And I just want to talk about the two that 00:05:15.430 --> 00:05:17.550 align:middle line:90% were mainstream for a while. 00:05:17.550 --> 00:05:20.870 align:middle line:84% And that is gaseous diffusion and gas centrifuge. 00:05:20.870 --> 00:05:24.850 align:middle line:84% So gas diffusion works like this. 00:05:24.850 --> 00:05:27.970 align:middle line:90% You have a tiny membrane. 00:05:27.970 --> 00:05:32.430 align:middle line:84% This is typically made by taking a very fine nickel powder 00:05:32.430 --> 00:05:33.650 align:middle line:90% and then heating it. 00:05:33.650 --> 00:05:36.590 align:middle line:84% So what we call centers-- so it kind of melts together 00:05:36.590 --> 00:05:42.430 align:middle line:84% but doesn't fully melt. And you get basically this sponge. 00:05:42.430 --> 00:05:44.090 align:middle line:90% You can see the scale there. 00:05:44.090 --> 00:05:47.310 align:middle line:90% 00:05:47.310 --> 00:05:50.390 align:middle line:90% Yeah, it's really hard to read. 00:05:50.390 --> 00:05:56.090 align:middle line:90% 00:05:56.090 --> 00:05:58.130 align:middle line:90% I think that scale is 1 micron. 00:05:58.130 --> 00:06:02.170 align:middle line:90% I don't know what. 00:06:02.170 --> 00:06:07.090 align:middle line:84% Oh, it's definitely not millimeters. 00:06:07.090 --> 00:06:08.750 align:middle line:90% But yeah. 00:06:08.750 --> 00:06:13.050 align:middle line:84% So the idea is that the size is basically small 00:06:13.050 --> 00:06:16.390 align:middle line:84% compared to the mean-free path of the atoms. 00:06:16.390 --> 00:06:18.650 align:middle line:90% So this is a uranium gas. 00:06:18.650 --> 00:06:21.770 align:middle line:84% Just in case you don't know how you turn uranium into a gas, 00:06:21.770 --> 00:06:25.790 align:middle line:90% you put six fluorines on it. 00:06:25.790 --> 00:06:33.650 align:middle line:90% 00:06:33.650 --> 00:06:35.890 align:middle line:90% You have six. 00:06:35.890 --> 00:06:43.330 align:middle line:84% And the fluorines basically create a shield around it 00:06:43.330 --> 00:06:46.150 align:middle line:84% so that there's very low Van der Waals, 00:06:46.150 --> 00:06:48.330 align:middle line:84% and so they don't stick together. 00:06:48.330 --> 00:06:51.390 align:middle line:84% Fluorine conveniently only has one naturally occurring 00:06:51.390 --> 00:06:54.550 align:middle line:84% isotope in nature, so the mass difference 00:06:54.550 --> 00:06:57.670 align:middle line:84% of the fluorinated molecule and the unfluorinated molecule 00:06:57.670 --> 00:06:58.810 align:middle line:90% is exactly the same. 00:06:58.810 --> 00:07:02.710 align:middle line:84% And that allows you to do the separation chemistry, 00:07:02.710 --> 00:07:04.050 align:middle line:90% do the separation process. 00:07:04.050 --> 00:07:06.170 align:middle line:84% If fluorine had different fluorine isotopes, 00:07:06.170 --> 00:07:09.110 align:middle line:84% that would totally screw up your attempts 00:07:09.110 --> 00:07:14.190 align:middle line:84% to do this enrichment because you do it by weight. 00:07:14.190 --> 00:07:18.350 align:middle line:84% Conveniently, fluorine has this property. 00:07:18.350 --> 00:07:24.350 align:middle line:84% So the idea is basically that the energy is 00:07:24.350 --> 00:07:25.970 align:middle line:90% equal to one-half mv squared. 00:07:25.970 --> 00:07:34.150 align:middle line:84% And so v is proportional to 1 over the mass, 00:07:34.150 --> 00:07:35.790 align:middle line:90% root of 1 over the mass. 00:07:35.790 --> 00:07:41.670 align:middle line:84% And that means that provided that everything 00:07:41.670 --> 00:07:45.390 align:middle line:84% has the same energy, which it does because it's constantly 00:07:45.390 --> 00:07:48.170 align:middle line:84% colliding and exchanging energy between atoms. 00:07:48.170 --> 00:07:51.170 align:middle line:84% So on average, everything is roughly the same energy. 00:07:51.170 --> 00:07:54.770 align:middle line:84% On average, the isotopes that are heavier 00:07:54.770 --> 00:07:59.210 align:middle line:84% will be traveling a little bit more slowly. 00:07:59.210 --> 00:08:01.890 align:middle line:90% And that's just a simple idea. 00:08:01.890 --> 00:08:04.810 align:middle line:84% So if they're traveling more slowly, 00:08:04.810 --> 00:08:08.755 align:middle line:84% that means they're going to run into this membrane less often, 00:08:08.755 --> 00:08:11.130 align:middle line:84% and the light isotopes are going to run into the membrane 00:08:11.130 --> 00:08:12.370 align:middle line:90% more often. 00:08:12.370 --> 00:08:14.490 align:middle line:90% And that's it. 00:08:14.490 --> 00:08:17.810 align:middle line:84% Preferentially, because they hit the membrane more often, 00:08:17.810 --> 00:08:20.930 align:middle line:84% the light isotopes preferentially diffuse 00:08:20.930 --> 00:08:24.010 align:middle line:84% because, on average, they're going just a little bit faster. 00:08:24.010 --> 00:08:28.730 align:middle line:84% And that's how you separate atom by atom the uranium 00:08:28.730 --> 00:08:30.250 align:middle line:90% and do your enrichment. 00:08:30.250 --> 00:08:33.770 align:middle line:84% So that's how gaseous diffusion worked. 00:08:33.770 --> 00:08:37.150 align:middle line:84% It produced these very large buildings. 00:08:37.150 --> 00:08:39.669 align:middle line:84% I think I showed you this last time, this K-25, 00:08:39.669 --> 00:08:42.289 align:middle line:84% which was built at the end of Manhattan Project, 00:08:42.289 --> 00:08:46.050 align:middle line:84% 44 acres in a single building employed 12,000 people and used 00:08:46.050 --> 00:08:49.340 align:middle line:84% in one building three times the electricity of Detroit, 00:08:49.340 --> 00:08:49.925 align:middle line:90% Michigan. 00:08:49.925 --> 00:08:52.300 align:middle line:84% When it opened, it was the largest building in the world, 00:08:52.300 --> 00:08:55.200 align:middle line:84% and it was capable of making only 20 nuclear weapons 00:08:55.200 --> 00:08:57.720 align:middle line:90% per year. 00:08:57.720 --> 00:09:00.680 align:middle line:84% The same capacity, 20 nuclear weapons per year, 00:09:00.680 --> 00:09:06.040 align:middle line:84% can now be had using roughly 2,500 of these over tubes, 00:09:06.040 --> 00:09:08.640 align:middle line:84% which are called gas centrifuges. 00:09:08.640 --> 00:09:15.400 align:middle line:84% This is a photo of Iran's prototype facility. 00:09:15.400 --> 00:09:20.320 align:middle line:84% Such a facility might take up 25,000ft, 00:09:20.320 --> 00:09:23.320 align:middle line:84% which is about the size of a high school gymnasium. 00:09:23.320 --> 00:09:25.980 align:middle line:84% And it might consume as much electricity as, say, 00:09:25.980 --> 00:09:29.000 align:middle line:84% a typical office building at the same size. 00:09:29.000 --> 00:09:31.160 align:middle line:90% This is not hard. 00:09:31.160 --> 00:09:34.280 align:middle line:84% This is not 44 acres and 12,000 people and three times 00:09:34.280 --> 00:09:35.640 align:middle line:90% electricity of Detroit. 00:09:35.640 --> 00:09:37.822 align:middle line:90% I dug for this photo last time. 00:09:37.822 --> 00:09:40.280 align:middle line:84% This is the diffusion process-- the old diffusion process-- 00:09:40.280 --> 00:09:41.880 align:middle line:84% and that's a person, just to give you 00:09:41.880 --> 00:09:46.900 align:middle line:84% a sense of the scale of the equipment in this facility. 00:09:46.900 --> 00:09:49.540 align:middle line:84% To do the same for centrifuges, there they are. 00:09:49.540 --> 00:09:51.100 align:middle line:90% And there's a person. 00:09:51.100 --> 00:09:54.940 align:middle line:84% This is Iran's former president, Ahmadinejad. 00:09:54.940 --> 00:09:58.520 align:middle line:84% And you can see that there's a lot of pipework. 00:09:58.520 --> 00:10:04.780 align:middle line:84% But beyond that, this is not all that difficult by comparison. 00:10:04.780 --> 00:10:07.280 align:middle line:84% So let's just look a little bit at how the centrifuge works. 00:10:07.280 --> 00:10:10.140 align:middle line:90% I explained how diffusion works. 00:10:10.140 --> 00:10:13.060 align:middle line:84% Centrifuge works in a very simple way 00:10:13.060 --> 00:10:17.800 align:middle line:84% that you already know that basically you have a gas. 00:10:17.800 --> 00:10:20.740 align:middle line:84% So it's again a three-stream machine. 00:10:20.740 --> 00:10:23.060 align:middle line:84% So three-stream machine-- everything 00:10:23.060 --> 00:10:25.040 align:middle line:90% has to have an inlet and outlet. 00:10:25.040 --> 00:10:28.040 align:middle line:90% So here's the inlet right here. 00:10:28.040 --> 00:10:30.220 align:middle line:84% Can you see there's a skinny tube inside 00:10:30.220 --> 00:10:32.300 align:middle line:90% of a bigger diameter tube? 00:10:32.300 --> 00:10:36.340 align:middle line:84% And then there are two outlets-- one 00:10:36.340 --> 00:10:39.860 align:middle line:84% that takes the enriched product and one that 00:10:39.860 --> 00:10:42.100 align:middle line:90% takes the depleted product. 00:10:42.100 --> 00:10:44.940 align:middle line:84% Let me just see if I can remember which way this goes. 00:10:44.940 --> 00:10:50.460 align:middle line:90% 00:10:50.460 --> 00:10:52.460 align:middle line:84% Yeah, so this is the enriched product down here. 00:10:52.460 --> 00:10:55.920 align:middle line:84% And this is depleted product up here. 00:10:55.920 --> 00:11:03.000 align:middle line:84% So the way this works is gas comes out of this tube. 00:11:03.000 --> 00:11:06.640 align:middle line:84% This inner tube is a rotor, this here, 00:11:06.640 --> 00:11:09.440 align:middle line:90% and it's spinning very fast. 00:11:09.440 --> 00:11:10.700 align:middle line:90% How fast is very fast? 00:11:10.700 --> 00:11:14.060 align:middle line:84% Well, the peripheral velocity of this tube, 00:11:14.060 --> 00:11:16.280 align:middle line:84% if you put it on the ground and let it roll, 00:11:16.280 --> 00:11:20.320 align:middle line:84% is typically up to about a kilometer per second. 00:11:20.320 --> 00:11:22.720 align:middle line:90% So that's very fast. 00:11:22.720 --> 00:11:25.720 align:middle line:84% In order to spin the tube that fast, 00:11:25.720 --> 00:11:28.620 align:middle line:84% you have to evacuate all the air from around it. 00:11:28.620 --> 00:11:31.120 align:middle line:84% Otherwise, the air friction would completely 00:11:31.120 --> 00:11:32.980 align:middle line:84% heat this thing up and be a problem. 00:11:32.980 --> 00:11:35.280 align:middle line:84% So that's why it sits in another tube. 00:11:35.280 --> 00:11:37.920 align:middle line:84% And even so, these little pipes on the outside 00:11:37.920 --> 00:11:40.640 align:middle line:84% that are wrapping around here, those pipes 00:11:40.640 --> 00:11:46.020 align:middle line:84% are there to cool off the outer casing basically 00:11:46.020 --> 00:11:49.620 align:middle line:84% through radiative coupling to control the thermal profile 00:11:49.620 --> 00:11:51.460 align:middle line:90% on the inner tube. 00:11:51.460 --> 00:11:54.300 align:middle line:84% Then what happens is the gas comes out. 00:11:54.300 --> 00:11:55.960 align:middle line:84% And because everything is spinning it, 00:11:55.960 --> 00:11:58.480 align:middle line:84% immediately, the gas is going to the wall, 00:11:58.480 --> 00:12:01.080 align:middle line:84% like when you're spinning water in a bucket around your head, 00:12:01.080 --> 00:12:02.260 align:middle line:90% centrifugal force. 00:12:02.260 --> 00:12:04.840 align:middle line:84% So initially, this enters in what is mostly a vacuum. 00:12:04.840 --> 00:12:09.220 align:middle line:84% And the gas basically in effect freezes as it comes out. 00:12:09.220 --> 00:12:13.780 align:middle line:84% And then it slams into the wall or into the very thin layer 00:12:13.780 --> 00:12:16.020 align:middle line:84% of gas, which is pressed against the wall. 00:12:16.020 --> 00:12:19.620 align:middle line:84% And because the heavier isotope, uranium-238 00:12:19.620 --> 00:12:22.580 align:middle line:84% is heavier, it gets pressed against the wall a little bit 00:12:22.580 --> 00:12:26.300 align:middle line:84% further, so it tends to collect in next to the wall. 00:12:26.300 --> 00:12:28.420 align:middle line:84% Now, if you just stop there, you can 00:12:28.420 --> 00:12:32.340 align:middle line:84% imagine having-- this is called a scoop, and it extends in, 00:12:32.340 --> 00:12:35.060 align:middle line:84% has a little arm, a little pipe that extends 00:12:35.060 --> 00:12:37.740 align:middle line:90% into the gas next to the wall. 00:12:37.740 --> 00:12:40.520 align:middle line:84% And you could imagine that you could scrape off 00:12:40.520 --> 00:12:44.840 align:middle line:84% a thin layer of the inside, and in principle, you could do that. 00:12:44.840 --> 00:12:46.440 align:middle line:84% But the problem is if you just did 00:12:46.440 --> 00:12:51.800 align:middle line:84% that, you would get very, very, very little separation, 00:12:51.800 --> 00:12:55.200 align:middle line:84% because the difference between just 00:12:55.200 --> 00:12:59.080 align:middle line:84% the gravitational separation from that is not a lot. 00:12:59.080 --> 00:13:04.500 align:middle line:84% So what we do is we take a trick that is used in chemistry. 00:13:04.500 --> 00:13:09.080 align:middle line:84% Does anyone remember doing distillation in chemistry 00:13:09.080 --> 00:13:18.760 align:middle line:84% where you had a glass tube, and it was full of little beads? 00:13:18.760 --> 00:13:20.840 align:middle line:90% Have you seen this? 00:13:20.840 --> 00:13:21.840 align:middle line:90% No? 00:13:21.840 --> 00:13:23.300 align:middle line:84% Or if you're a chemical engineer, 00:13:23.300 --> 00:13:26.480 align:middle line:84% you may know about a separation stage 00:13:26.480 --> 00:13:32.200 align:middle line:84% that has so-called plates that look like this. 00:13:32.200 --> 00:13:34.000 align:middle line:84% No one's a chemical engineer either. 00:13:34.000 --> 00:13:35.920 align:middle line:84% All right, so here's how it works. 00:13:35.920 --> 00:13:40.460 align:middle line:90% You boil some mix of chemicals. 00:13:40.460 --> 00:13:46.100 align:middle line:84% And let's say it's water and ethanol because we 00:13:46.100 --> 00:13:47.460 align:middle line:90% want to make alcohol. 00:13:47.460 --> 00:13:51.160 align:middle line:84% Well, some water boils off, and some ethanol boils off. 00:13:51.160 --> 00:13:52.760 align:middle line:84% Ethanol boils off preferentially. 00:13:52.760 --> 00:13:55.140 align:middle line:84% But you still got a lot of water in there too. 00:13:55.140 --> 00:13:58.080 align:middle line:84% But then it boils off into a gas, 00:13:58.080 --> 00:14:00.680 align:middle line:84% but then it condenses here and drips onto this plate. 00:14:00.680 --> 00:14:02.900 align:middle line:84% And now, you have a little liquid of mixture. 00:14:02.900 --> 00:14:06.100 align:middle line:90% And now this is also ETOH-- 00:14:06.100 --> 00:14:09.820 align:middle line:90% methanol plus water. 00:14:09.820 --> 00:14:12.672 align:middle line:84% But it's slightly enriched in alcohol 00:14:12.672 --> 00:14:14.380 align:middle line:84% because the vapor pressure of the alcohol 00:14:14.380 --> 00:14:15.780 align:middle line:90% is a little bit lower. 00:14:15.780 --> 00:14:19.208 align:middle line:84% And if you do it again, you boil off some stuff, 00:14:19.208 --> 00:14:20.500 align:middle line:90% and then it will condense here. 00:14:20.500 --> 00:14:22.042 align:middle line:84% And you'll have some sphere, and this 00:14:22.042 --> 00:14:25.300 align:middle line:84% will be even more enriched in ethanol, and so on and so forth 00:14:25.300 --> 00:14:27.100 align:middle line:90% as you go up the column. 00:14:27.100 --> 00:14:28.962 align:middle line:84% So this is called a fractionating column. 00:14:28.962 --> 00:14:30.420 align:middle line:84% You can do the same thing with just 00:14:30.420 --> 00:14:32.920 align:middle line:84% with beads of glass in glass tubes. 00:14:32.920 --> 00:14:34.540 align:middle line:90% You'll see that in chem labs. 00:14:34.540 --> 00:14:37.130 align:middle line:90% We do the exact same thing here. 00:14:37.130 --> 00:14:41.670 align:middle line:84% And the way it works is that we have-- where's my laser? 00:14:41.670 --> 00:14:45.290 align:middle line:90% 00:14:45.290 --> 00:14:51.810 align:middle line:84% This scoop is dragging into the gas. 00:14:51.810 --> 00:14:54.290 align:middle line:84% It's pressed against the wall, and therefore it's taking up 00:14:54.290 --> 00:14:56.090 align:middle line:90% some of the angular momentum. 00:14:56.090 --> 00:15:00.690 align:middle line:84% And that causes the gas to move inwards and up. 00:15:00.690 --> 00:15:02.850 align:middle line:90% It acts like a little pump. 00:15:02.850 --> 00:15:06.370 align:middle line:84% So it pumps the gas up on the edge, 00:15:06.370 --> 00:15:09.370 align:middle line:84% and then towards the center in the rarefied area, 00:15:09.370 --> 00:15:12.490 align:middle line:84% there's a countercurrent downward flow. 00:15:12.490 --> 00:15:16.330 align:middle line:84% So you have these two streams of gas passing each other. 00:15:16.330 --> 00:15:19.370 align:middle line:84% And everywhere they are adjacent, 00:15:19.370 --> 00:15:23.330 align:middle line:84% there's this horizontal transport because of gravity. 00:15:23.330 --> 00:15:28.070 align:middle line:84% And so it's effectively acting like that fractionating column. 00:15:28.070 --> 00:15:30.890 align:middle line:84% And so at every radial coordinate, 00:15:30.890 --> 00:15:32.730 align:middle line:90% there is an enrichment. 00:15:32.730 --> 00:15:35.890 align:middle line:84% But as you move axially along this device. 00:15:35.890 --> 00:15:37.850 align:middle line:84% The enrichment continues to increase, increase, 00:15:37.850 --> 00:15:39.150 align:middle line:90% increase, increase, increase. 00:15:39.150 --> 00:15:42.830 align:middle line:84% And so you get a massive multiplication of the enrichment 00:15:42.830 --> 00:15:44.070 align:middle line:90% through this process. 00:15:44.070 --> 00:15:49.070 align:middle line:84% And so this is why this little shield is here, 00:15:49.070 --> 00:15:51.190 align:middle line:84% to prevent the opposite effect from happening 00:15:51.190 --> 00:15:54.390 align:middle line:84% from the other scoop, which would send it down. 00:15:54.390 --> 00:16:00.750 align:middle line:90% So that's how these things work. 00:16:00.750 --> 00:16:03.710 align:middle line:84% So they have some nice features because 00:16:03.710 --> 00:16:05.810 align:middle line:84% of this countercurrent flow inside the machine. 00:16:05.810 --> 00:16:08.470 align:middle line:84% They have a very high separation factor, which 00:16:08.470 --> 00:16:10.470 align:middle line:90% means they need fewer stages. 00:16:10.470 --> 00:16:17.110 align:middle line:84% So to make a bomb using the old gaseous diffusion plant, 00:16:17.110 --> 00:16:22.030 align:middle line:84% this thing, you need about 1,000 of these giant things piped 00:16:22.030 --> 00:16:23.710 align:middle line:90% together in a row. 00:16:23.710 --> 00:16:28.910 align:middle line:84% To make a bomb using these, you only need order of 100 or less, 00:16:28.910 --> 00:16:30.950 align:middle line:90% 100 or fewer. 00:16:30.950 --> 00:16:32.810 align:middle line:84% They also have very low inventory. 00:16:32.810 --> 00:16:34.850 align:middle line:84% There's not a lot of gas in here. 00:16:34.850 --> 00:16:37.690 align:middle line:84% So if you have this system running 00:16:37.690 --> 00:16:40.325 align:middle line:84% and you want to change the enrichment level 00:16:40.325 --> 00:16:41.950 align:middle line:84% from low enrichment to high enrichment, 00:16:41.950 --> 00:16:45.370 align:middle line:84% you can flush it out and get the whole thing reestablished 00:16:45.370 --> 00:16:48.250 align:middle line:84% at a new enrichment level in about a day. 00:16:48.250 --> 00:16:50.870 align:middle line:84% If you change the enrichment level in this system, 00:16:50.870 --> 00:16:54.010 align:middle line:84% it takes about a year to change the gas out. 00:16:54.010 --> 00:16:56.230 align:middle line:84% And that means that once this thing is going, 00:16:56.230 --> 00:16:59.190 align:middle line:84% you're not making LEU for reactors. 00:16:59.190 --> 00:17:02.130 align:middle line:84% You're not going to turn it into a bomb factory. 00:17:02.130 --> 00:17:03.690 align:middle line:90% It just takes too long. 00:17:03.690 --> 00:17:05.609 align:middle line:84% Whereas this you can just after a day, 00:17:05.609 --> 00:17:08.930 align:middle line:84% you've converted the whole plant to making bombs. 00:17:08.930 --> 00:17:12.230 align:middle line:90% So it also is modular. 00:17:12.230 --> 00:17:15.170 align:middle line:84% You have a bunch of these that are all identical. 00:17:15.170 --> 00:17:16.250 align:middle line:90% Yep. 00:17:16.250 --> 00:17:19.810 align:middle line:84% AUDIENCE: If you can use it for high energy [INAUDIBLE], 00:17:19.810 --> 00:17:23.890 align:middle line:84% what [INAUDIBLE] used to produce uranium for civil reactors, 00:17:23.890 --> 00:17:24.490 align:middle line:90% then? 00:17:24.490 --> 00:17:26.738 align:middle line:90% SCOTT KEMP: It also uses this. 00:17:26.738 --> 00:17:28.530 align:middle line:84% AUDIENCE: Doesn't that pose more of a risk? 00:17:28.530 --> 00:17:29.570 align:middle line:90% SCOTT KEMP: Yes. 00:17:29.570 --> 00:17:32.570 align:middle line:84% AUDIENCE: Is there a similar device that's is as effective, 00:17:32.570 --> 00:17:35.050 align:middle line:84% but can also be used for highly enriched uranium? 00:17:35.050 --> 00:17:37.670 align:middle line:90% SCOTT KEMP: No. 00:17:37.670 --> 00:17:41.230 align:middle line:84% Basically, if it can physically separate isotopes, 00:17:41.230 --> 00:17:46.190 align:middle line:84% it really doesn't care what the enrichment level is. 00:17:46.190 --> 00:17:49.470 align:middle line:84% And you can think of it this way. 00:17:49.470 --> 00:17:53.550 align:middle line:84% Low-enriched uranium, which is to say 00:17:53.550 --> 00:17:57.830 align:middle line:84% uranium that has very low uranium-235, 00:17:57.830 --> 00:18:01.030 align:middle line:90% is high-enriched uranium-238. 00:18:01.030 --> 00:18:04.550 align:middle line:90% The same thing has high levels. 00:18:04.550 --> 00:18:07.750 align:middle line:84% So you're starting off at one extreme, 00:18:07.750 --> 00:18:10.590 align:middle line:84% and it can get you to the other extreme. 00:18:10.590 --> 00:18:13.783 align:middle line:90% So yeah, it always works. 00:18:13.783 --> 00:18:15.450 align:middle line:84% No matter what, you would take one side. 00:18:15.450 --> 00:18:18.390 align:middle line:90% You would take the other side. 00:18:18.390 --> 00:18:19.270 align:middle line:90% So yeah. 00:18:19.270 --> 00:18:20.210 align:middle line:90% Yep. 00:18:20.210 --> 00:18:23.430 align:middle line:84% AUDIENCE: Does this also apply to the separation 00:18:23.430 --> 00:18:26.870 align:middle line:84% of other isotopes like people are talking about [INAUDIBLE] 00:18:26.870 --> 00:18:30.770 align:middle line:84% and like lithium enrichment [INAUDIBLE] 00:18:30.770 --> 00:18:35.350 align:middle line:84% also then turn that into uranium enrichment technology? 00:18:35.350 --> 00:18:37.090 align:middle line:84% Or are they trying to [INAUDIBLE]? 00:18:37.090 --> 00:18:40.110 align:middle line:84% SCOTT KEMP: So sometimes yes and sometimes no. 00:18:40.110 --> 00:18:45.250 align:middle line:90% 00:18:45.250 --> 00:18:47.110 align:middle line:90% The short answer is yes. 00:18:47.110 --> 00:18:49.110 align:middle line:84% You can always you can always do it, 00:18:49.110 --> 00:18:51.250 align:middle line:84% but it may not be the optimal technology. 00:18:51.250 --> 00:18:54.290 align:middle line:84% And so the reason is that when you're 00:18:54.290 --> 00:18:57.570 align:middle line:84% looking at enriching uranium, the difference 00:18:57.570 --> 00:19:05.170 align:middle line:90% between 238 and 235 is 3. 00:19:05.170 --> 00:19:10.290 align:middle line:84% And so it's three parts out of a roughly 238 parts. 00:19:10.290 --> 00:19:13.290 align:middle line:84% And that's a very small fraction. 00:19:13.290 --> 00:19:19.650 align:middle line:84% Now, the difference between 6-lithium and 7-lithium is 1 out 00:19:19.650 --> 00:19:23.370 align:middle line:90% of six, or 6 and 1/2. 00:19:23.370 --> 00:19:26.770 align:middle line:84% So this is a much larger fractional mass change. 00:19:26.770 --> 00:19:33.350 align:middle line:84% So technologies that are crummier and don't work as well 00:19:33.350 --> 00:19:35.270 align:middle line:90% are fine for doing lithium. 00:19:35.270 --> 00:19:38.610 align:middle line:84% But if you try to do it with uranium, 00:19:38.610 --> 00:19:40.750 align:middle line:90% it would just be very expensive. 00:19:40.750 --> 00:19:41.570 align:middle line:90% You could do it. 00:19:41.570 --> 00:19:45.270 align:middle line:84% It will work, but it's very inefficient. 00:19:45.270 --> 00:19:47.470 align:middle line:84% So the most efficient technologies 00:19:47.470 --> 00:19:49.450 align:middle line:84% are preferred when doing uranium. 00:19:49.450 --> 00:19:51.790 align:middle line:84% And that's why it's all very fancy. 00:19:51.790 --> 00:19:54.850 align:middle line:84% So with this, you can do crown ethers. 00:19:54.850 --> 00:19:56.350 align:middle line:84% You can do chemical separation where 00:19:56.350 --> 00:19:58.510 align:middle line:84% there's a chemical reaction, which preferentially 00:19:58.510 --> 00:19:59.850 align:middle line:90% reacts with one or the other. 00:19:59.850 --> 00:20:01.392 align:middle line:84% You can do all that with uranium too. 00:20:01.392 --> 00:20:02.910 align:middle line:84% It's just the amount of chemistry 00:20:02.910 --> 00:20:07.550 align:middle line:84% you need is just too much to make it economic. 00:20:07.550 --> 00:20:10.670 align:middle line:84% AUDIENCE: 2.1505 would be very expensive. 00:20:10.670 --> 00:20:11.570 align:middle line:90% SCOTT KEMP: Yeah. 00:20:11.570 --> 00:20:12.830 align:middle line:90% AUDIENCE: [INAUDIBLE] lithium. 00:20:12.830 --> 00:20:16.490 align:middle line:84% If you use this lithium, then would it be two stages? 00:20:16.490 --> 00:20:19.990 align:middle line:84% You're done super cheap, or would it still be-- 00:20:19.990 --> 00:20:20.950 align:middle line:90% SCOTT KEMP: It would. 00:20:20.950 --> 00:20:23.990 align:middle line:84% They've presumably worked it all out. 00:20:23.990 --> 00:20:27.050 align:middle line:84% So to do this, you need to have the material in a gas. 00:20:27.050 --> 00:20:30.180 align:middle line:84% Most things can be converted to gas-- 00:20:30.180 --> 00:20:32.460 align:middle line:84% not everything, but I don't know if lithium 00:20:32.460 --> 00:20:34.140 align:middle line:90% can be converted to a gas. 00:20:34.140 --> 00:20:38.060 align:middle line:90% Probably, but I'm not sure. 00:20:38.060 --> 00:20:38.560 align:middle line:90% Yep. 00:20:38.560 --> 00:20:40.810 align:middle line:84% AUDIENCE: I think you have the same separation if they 00:20:40.810 --> 00:20:41.900 align:middle line:90% were in liquid fuels. 00:20:41.900 --> 00:20:45.580 align:middle line:84% You're just exerting a centrifugal force. 00:20:45.580 --> 00:20:48.740 align:middle line:84% SCOTT KEMP: Could you do it in a liquid? 00:20:48.740 --> 00:20:51.780 align:middle line:90% So let me think. 00:20:51.780 --> 00:20:55.060 align:middle line:90% In principle, you could. 00:20:55.060 --> 00:20:57.460 align:middle line:84% They're both fluids, but you would 00:20:57.460 --> 00:21:03.060 align:middle line:84% be dealing with a lot more heating and energy 00:21:03.060 --> 00:21:07.380 align:middle line:84% from viscosity, which might make it impossible to spin it 00:21:07.380 --> 00:21:09.500 align:middle line:90% as fast as you need to. 00:21:09.500 --> 00:21:12.220 align:middle line:90% Yeah. 00:21:12.220 --> 00:21:12.900 align:middle line:90% Yep. 00:21:12.900 --> 00:21:14.525 align:middle line:84% AUDIENCE: Yeah, so I just looked it up. 00:21:14.525 --> 00:21:17.180 align:middle line:84% So lithium enrichment is done with chemical systems. 00:21:17.180 --> 00:21:20.060 align:middle line:84% SCOTT KEMP: Yeah, it's usually with crown ether, 00:21:20.060 --> 00:21:26.900 align:middle line:90% or there's a cyanide process. 00:21:26.900 --> 00:21:28.440 align:middle line:90% I forget. 00:21:28.440 --> 00:21:33.220 align:middle line:84% Yeah, generally, when you're talking low-numbered isotopes, 00:21:33.220 --> 00:21:36.460 align:middle line:84% you generally favor the chemical processes. 00:21:36.460 --> 00:21:40.580 align:middle line:84% There are chemical uranium enrichment. 00:21:40.580 --> 00:21:43.400 align:middle line:84% The Japanese were building one for a while, 00:21:43.400 --> 00:21:46.220 align:middle line:84% but they have very, very long equilibrium times 00:21:46.220 --> 00:21:49.720 align:middle line:84% because the separation stages are so weak. 00:21:49.720 --> 00:21:51.360 align:middle line:84% They take years and years and years 00:21:51.360 --> 00:21:53.840 align:middle line:84% to get those plants into equilibrium. 00:21:53.840 --> 00:21:55.760 align:middle line:84% And people have actually proposed that 00:21:55.760 --> 00:22:00.520 align:middle line:84% as a way of making uranium enrichment 00:22:00.520 --> 00:22:02.600 align:middle line:84% proliferation-resistant because it 00:22:02.600 --> 00:22:07.407 align:middle line:84% would be so inefficient to get to HEU levels. 00:22:07.407 --> 00:22:08.740 align:middle line:90% And I've always liked that idea. 00:22:08.740 --> 00:22:11.720 align:middle line:84% But the problem is that this is cheaper, and cheaper 00:22:11.720 --> 00:22:13.480 align:middle line:90% always wins. 00:22:13.480 --> 00:22:16.100 align:middle line:84% It's the same problem with wind and solar. 00:22:16.100 --> 00:22:21.080 align:middle line:84% Cheaper always wins, no matter how good your technology is. 00:22:21.080 --> 00:22:23.900 align:middle line:84% So modular, so you can crank these things out by you 00:22:23.900 --> 00:22:28.420 align:middle line:84% have just a factory mass production of these things. 00:22:28.420 --> 00:22:30.940 align:middle line:84% They don't really produce any emissions. 00:22:30.940 --> 00:22:35.780 align:middle line:84% They actually operate below atmospheric pressure. 00:22:35.780 --> 00:22:36.940 align:middle line:90% So oops. 00:22:36.940 --> 00:22:42.220 align:middle line:84% So if there's a leak anywhere in the system, air leaks into it 00:22:42.220 --> 00:22:45.320 align:middle line:84% rather than uranium gas leaking out. 00:22:45.320 --> 00:22:47.340 align:middle line:84% So it is almost nothing that you could 00:22:47.340 --> 00:22:49.160 align:middle line:90% sniff around and try to detect. 00:22:49.160 --> 00:22:51.900 align:middle line:90% 00:22:51.900 --> 00:22:54.120 align:middle line:84% Like I said, you don't need a special building. 00:22:54.120 --> 00:22:56.340 align:middle line:84% You put this in a high school gymnasium. 00:22:56.340 --> 00:22:58.780 align:middle line:84% And so there's not really any signatures 00:22:58.780 --> 00:23:02.140 align:middle line:84% that you can use to detect it from satellite imagery 00:23:02.140 --> 00:23:03.980 align:middle line:90% or something-- 00:23:03.980 --> 00:23:05.548 align:middle line:90% very low energy consumption. 00:23:05.548 --> 00:23:07.340 align:middle line:84% So again, it looks like an office building, 00:23:07.340 --> 00:23:09.700 align:middle line:84% doesn't look like a building that for some reason 00:23:09.700 --> 00:23:13.780 align:middle line:84% is using three times the electricity of Detroit. 00:23:13.780 --> 00:23:16.420 align:middle line:84% Nothing suspicious in the infrared thermal signature 00:23:16.420 --> 00:23:18.660 align:middle line:90% because of that. 00:23:18.660 --> 00:23:22.400 align:middle line:84% And so it's almost impossible to detect by technical means. 00:23:22.400 --> 00:23:25.360 align:middle line:90% 00:23:25.360 --> 00:23:28.920 align:middle line:84% The intelligence history for clandestine centrifuge programs 00:23:28.920 --> 00:23:30.360 align:middle line:90% bears that out. 00:23:30.360 --> 00:23:35.040 align:middle line:90% So here is the list. 00:23:35.040 --> 00:23:37.480 align:middle line:90% Iraq had a program. 00:23:37.480 --> 00:23:40.520 align:middle line:84% They accidentally spilled the beans to us. 00:23:40.520 --> 00:23:41.900 align:middle line:90% It went for years undetected. 00:23:41.900 --> 00:23:43.200 align:middle line:90% We didn't know about it. 00:23:43.200 --> 00:23:46.280 align:middle line:90% Iran's program started in '85. 00:23:46.280 --> 00:23:51.200 align:middle line:84% We found out around 1991, and we only 00:23:51.200 --> 00:23:56.520 align:middle line:84% did so because people involved in this program and Pakistan's 00:23:56.520 --> 00:23:58.400 align:middle line:90% program were being watched. 00:23:58.400 --> 00:24:01.000 align:middle line:84% And they led us to the Iranian program. 00:24:01.000 --> 00:24:06.360 align:middle line:84% But we didn't detect Iran's centrifuges directly. 00:24:06.360 --> 00:24:08.002 align:middle line:84% Same is true of the Libyan program. 00:24:08.002 --> 00:24:09.460 align:middle line:84% We were also watching these people. 00:24:09.460 --> 00:24:11.760 align:middle line:84% Despite that, it went undetected for 16 years 00:24:11.760 --> 00:24:16.000 align:middle line:84% because the program was operating at a very low cadence. 00:24:16.000 --> 00:24:19.000 align:middle line:84% They didn't have a lot of interactions that tipped us off. 00:24:19.000 --> 00:24:22.940 align:middle line:84% China had a program that went undetected for roughly 21 years 00:24:22.940 --> 00:24:25.100 align:middle line:84% by the United States intelligence community. 00:24:25.100 --> 00:24:28.920 align:middle line:90% The Soviet Union's program-- 00:24:28.920 --> 00:24:32.060 align:middle line:84% there's a lot of barriers between the US and Soviet Union 00:24:32.060 --> 00:24:34.580 align:middle line:90% during the Cold War. 00:24:34.580 --> 00:24:36.960 align:middle line:90% It was not easy to watch them. 00:24:36.960 --> 00:24:39.460 align:middle line:84% At the same time, we watched no country more 00:24:39.460 --> 00:24:42.060 align:middle line:84% assiduously than we watched the Soviet Union, 00:24:42.060 --> 00:24:44.202 align:middle line:84% and we had all kinds of intelligence assessments 00:24:44.202 --> 00:24:45.660 align:middle line:84% about their nuclear weapons program 00:24:45.660 --> 00:24:47.620 align:middle line:84% because that's what we were obsessed about. 00:24:47.620 --> 00:24:51.140 align:middle line:84% Despite that, the official conclusion 00:24:51.140 --> 00:24:55.500 align:middle line:84% that the Soviets had centrifuges came 34 years after they 00:24:55.500 --> 00:24:57.890 align:middle line:90% built their first plant. 00:24:57.890 --> 00:25:00.140 align:middle line:84% And that's because there was a lot of people in the US 00:25:00.140 --> 00:25:02.140 align:middle line:84% who were just like, now the Soviets 00:25:02.140 --> 00:25:04.580 align:middle line:90% do gaseous diffusion like we do. 00:25:04.580 --> 00:25:07.840 align:middle line:84% And there were individual analysts who were like, well, 00:25:07.840 --> 00:25:09.380 align:middle line:90% maybe they switched. 00:25:09.380 --> 00:25:14.660 align:middle line:84% But the overall community just couldn't believe it. 00:25:14.660 --> 00:25:19.320 align:middle line:84% North Korea also started their program tinkering around 1984, 00:25:19.320 --> 00:25:22.720 align:middle line:84% and they basically revealed it to us 00:25:22.720 --> 00:25:27.640 align:middle line:84% as part of a diplomatic process 26 years later. 00:25:27.640 --> 00:25:32.060 align:middle line:84% So easy to hide these things if you're trying to hide them. 00:25:32.060 --> 00:25:34.440 align:middle line:90% And the evidence is there. 00:25:34.440 --> 00:25:38.480 align:middle line:84% Compare these multi-decade time scales for detection 00:25:38.480 --> 00:25:41.720 align:middle line:84% to the time it has taken countries to build a facility. 00:25:41.720 --> 00:25:43.920 align:middle line:84% So these are the historical R&D times 00:25:43.920 --> 00:25:50.880 align:middle line:84% to go from nothing to a technology suitable for turning 00:25:50.880 --> 00:25:53.900 align:middle line:84% into a plant-- not building the whole plant, 00:25:53.900 --> 00:25:56.360 align:middle line:84% but that you've done all the R&D and you're 00:25:56.360 --> 00:25:59.780 align:middle line:84% ready to do the plant-- a plant good enough to make a bomb, 00:25:59.780 --> 00:26:03.400 align:middle line:84% not necessarily good enough to sell commercially viable uranium 00:26:03.400 --> 00:26:05.400 align:middle line:90% enrichment. 00:26:05.400 --> 00:26:08.040 align:middle line:84% And so you can see the average development time, 00:26:08.040 --> 00:26:10.840 align:middle line:84% if you look at them, is about 24 months, 00:26:10.840 --> 00:26:12.260 align:middle line:90% which is frankly not that long. 00:26:12.260 --> 00:26:14.720 align:middle line:84% It's faster than building a research reactor. 00:26:14.720 --> 00:26:21.670 align:middle line:84% And the Australian program is one I like to focus on here. 00:26:21.670 --> 00:26:23.270 align:middle line:90% Does this have a [INAUDIBLE]? 00:26:23.270 --> 00:26:25.663 align:middle line:90% No, It doesn't. 00:26:25.663 --> 00:26:27.830 align:middle line:84% The reason I like to focus on the Australian program 00:26:27.830 --> 00:26:31.630 align:middle line:84% is because it's the slowest program on record. 00:26:31.630 --> 00:26:33.910 align:middle line:84% The individuals who were involved 00:26:33.910 --> 00:26:37.470 align:middle line:84% had no prior experience in any of the relevant fields. 00:26:37.470 --> 00:26:40.230 align:middle line:84% They were not turbomachinery experts. 00:26:40.230 --> 00:26:44.290 align:middle line:84% They were not experts on uranium enrichment. 00:26:44.290 --> 00:26:48.190 align:middle line:84% They were just engineers who had general mechanical experience 00:26:48.190 --> 00:26:52.670 align:middle line:84% and were asked to build this plant. 00:26:52.670 --> 00:26:55.430 align:middle line:84% At no point in the history of the program 00:26:55.430 --> 00:26:58.710 align:middle line:84% did they ever employ more than six people. 00:26:58.710 --> 00:27:02.030 align:middle line:84% And even so, it took them about six years 00:27:02.030 --> 00:27:04.190 align:middle line:90% to make a working technology-- 00:27:04.190 --> 00:27:05.870 align:middle line:90% very tiny program. 00:27:05.870 --> 00:27:07.630 align:middle line:84% So the way I like to think about this 00:27:07.630 --> 00:27:11.970 align:middle line:84% is to engineer the machine, the centrifuge machine, 00:27:11.970 --> 00:27:15.190 align:middle line:84% it's like three or four PhD dissertations, 00:27:15.190 --> 00:27:18.930 align:middle line:84% and we could easily do it multiple times in this room 00:27:18.930 --> 00:27:20.170 align:middle line:90% if you wanted to do it. 00:27:20.170 --> 00:27:22.810 align:middle line:84% So it really should come as no surprise to us 00:27:22.810 --> 00:27:27.490 align:middle line:84% that countries like North Korea were able to pull this off. 00:27:27.490 --> 00:27:31.910 align:middle line:84% So many countries have shown have pursued this technology. 00:27:31.910 --> 00:27:35.370 align:middle line:84% Not all of them continue to operate it. 00:27:35.370 --> 00:27:37.630 align:middle line:84% So here is the worldwide enrichment capability. 00:27:37.630 --> 00:27:41.370 align:middle line:84% All of this is centrifuges with actually 00:27:41.370 --> 00:27:44.610 align:middle line:90% with the exception of Argentina. 00:27:44.610 --> 00:27:48.610 align:middle line:84% And you can just see here these in blue 00:27:48.610 --> 00:27:51.650 align:middle line:84% have very large capacities, and these in red 00:27:51.650 --> 00:27:53.970 align:middle line:90% have very small capacities. 00:27:53.970 --> 00:27:57.050 align:middle line:84% These in blue operate their programs 00:27:57.050 --> 00:27:59.250 align:middle line:90% for commercial purposes. 00:27:59.250 --> 00:28:03.610 align:middle line:84% And they sell enriched uranium to make reactor fuel. 00:28:03.610 --> 00:28:06.050 align:middle line:84% What are these guys in red doing? 00:28:06.050 --> 00:28:07.550 align:middle line:90% Their programs are really tiny. 00:28:07.550 --> 00:28:11.050 align:middle line:84% So here the number, the size of the enrichment program 00:28:11.050 --> 00:28:16.670 align:middle line:84% is specified in bombs per year just to make it easy. 00:28:16.670 --> 00:28:19.870 align:middle line:84% And you can see here, these countries basically 00:28:19.870 --> 00:28:24.670 align:middle line:84% operate programs that are suitable for making 00:28:24.670 --> 00:28:25.930 align:middle line:90% a few weapons. 00:28:25.930 --> 00:28:30.150 align:middle line:90% But that's about it. 00:28:30.150 --> 00:28:34.070 align:middle line:84% If they were really serious about supplying fuel 00:28:34.070 --> 00:28:37.310 align:middle line:84% for their reactor fleet, they would have much larger programs. 00:28:37.310 --> 00:28:39.230 align:middle line:90% But they don't. 00:28:39.230 --> 00:28:40.090 align:middle line:90% And you know what? 00:28:40.090 --> 00:28:42.870 align:middle line:84% We generally don't really complain about it. 00:28:42.870 --> 00:28:46.250 align:middle line:84% Why don't we complain that Japan can make 15 bombs a year? 00:28:46.250 --> 00:28:48.830 align:middle line:84% We complain about Iran can make one bomb a year. 00:28:48.830 --> 00:28:51.450 align:middle line:90% Because Japan is our friend. 00:28:51.450 --> 00:28:51.950 align:middle line:90% Yep. 00:28:51.950 --> 00:28:54.130 align:middle line:84% AUDIENCE: Why is Japan [INAUDIBLE]? 00:28:54.130 --> 00:28:55.370 align:middle line:90% Because they shut stuff down? 00:28:55.370 --> 00:28:57.310 align:middle line:90% SCOTT KEMP: Yeah, no. 00:28:57.310 --> 00:29:02.790 align:middle line:84% They are bad engineers when it comes to making the centrifuges. 00:29:02.790 --> 00:29:06.530 align:middle line:84% They keep making a mistake that they haven't gotten past yet. 00:29:06.530 --> 00:29:08.390 align:middle line:84% The Brazilians are in the same camp. 00:29:08.390 --> 00:29:12.090 align:middle line:84% There's a lot of you can make it work, 00:29:12.090 --> 00:29:16.010 align:middle line:84% but if you want to make it last a long time, 00:29:16.010 --> 00:29:18.410 align:middle line:84% there are some things you have to know. 00:29:18.410 --> 00:29:20.790 align:middle line:84% And the Japanese have not figured that out. 00:29:20.790 --> 00:29:24.050 align:middle line:84% And so they built this large capacity, and almost all of it 00:29:24.050 --> 00:29:26.130 align:middle line:90% broke after a while. 00:29:26.130 --> 00:29:29.230 align:middle line:84% So I haven't updated this number recently, 00:29:29.230 --> 00:29:34.330 align:middle line:84% so I really am not sure exactly how big their number is today. 00:29:34.330 --> 00:29:36.330 align:middle line:84% So sometimes I get asked, well, couldn't we 00:29:36.330 --> 00:29:41.370 align:middle line:84% just build enough enrichment to support a reactor 00:29:41.370 --> 00:29:43.210 align:middle line:90% but not build a bomb? 00:29:43.210 --> 00:29:44.750 align:middle line:90% And it's the answer is no. 00:29:44.750 --> 00:29:45.953 align:middle line:90% It's the opposite. 00:29:45.953 --> 00:29:47.370 align:middle line:84% So here's the amount of enrichment 00:29:47.370 --> 00:29:49.188 align:middle line:90% you need for a bomb per year. 00:29:49.188 --> 00:29:50.730 align:middle line:84% That's the amount you need to support 00:29:50.730 --> 00:29:52.390 align:middle line:90% one gigawatt reactor per year. 00:29:52.390 --> 00:29:55.570 align:middle line:84% And this is a typical commercial plant. 00:29:55.570 --> 00:29:58.010 align:middle line:90% And you can just see that. 00:29:58.010 --> 00:30:03.970 align:middle line:84% AUDIENCE: [INAUDIBLE] capacity of the program [INAUDIBLE] 00:30:03.970 --> 00:30:06.890 align:middle line:84% to leave the country has nuclear or not? 00:30:06.890 --> 00:30:08.890 align:middle line:84% SCOTT KEMP: Is there a link between the capacity 00:30:08.890 --> 00:30:10.190 align:middle line:90% of the program? 00:30:10.190 --> 00:30:12.590 align:middle line:90% AUDIENCE: [INAUDIBLE] 00:30:12.590 --> 00:30:13.790 align:middle line:90% SCOTT KEMP: No. 00:30:13.790 --> 00:30:16.990 align:middle line:84% AUDIENCE: It has nuclear submarines and carriers? 00:30:16.990 --> 00:30:20.830 align:middle line:84% SCOTT KEMP: No, because what has emerged 00:30:20.830 --> 00:30:32.830 align:middle line:84% is that some countries, in particular Holland and Germany, 00:30:32.830 --> 00:30:35.190 align:middle line:84% have a very, very good technology, 00:30:35.190 --> 00:30:40.470 align:middle line:84% and they are able to sell this at a good price. 00:30:40.470 --> 00:30:41.870 align:middle line:90% And they make money. 00:30:41.870 --> 00:30:43.250 align:middle line:90% And so they do this. 00:30:43.250 --> 00:30:45.550 align:middle line:84% They export all this stuff despite the fact 00:30:45.550 --> 00:30:50.030 align:middle line:84% that they don't have significant nuclear establishment. 00:30:50.030 --> 00:30:52.290 align:middle line:84% Yeah, Germany used to, but they got rid of it. 00:30:52.290 --> 00:30:54.030 align:middle line:90% Holland never really did. 00:30:54.030 --> 00:30:55.410 align:middle line:90% So yeah. 00:30:55.410 --> 00:30:58.430 align:middle line:90% So it's just a business. 00:30:58.430 --> 00:31:01.270 align:middle line:84% And that's where the technology is. 00:31:01.270 --> 00:31:04.030 align:middle line:84% The Russians just keep building capacity 00:31:04.030 --> 00:31:06.350 align:middle line:90% that they can't even use. 00:31:06.350 --> 00:31:08.190 align:middle line:84% And it seems like their strategy is 00:31:08.190 --> 00:31:11.890 align:middle line:84% to try to put these people out of business 00:31:11.890 --> 00:31:16.387 align:middle line:84% because the Russian technology is so cheap for them to build. 00:31:16.387 --> 00:31:18.470 align:middle line:84% Seems like they're hoping to take over the market, 00:31:18.470 --> 00:31:20.428 align:middle line:84% but then we put sanctions against the Russians. 00:31:20.428 --> 00:31:25.010 align:middle line:90% And so we'll see what happens. 00:31:25.010 --> 00:31:28.570 align:middle line:84% Yeah, so basically, a full-size plant, 00:31:28.570 --> 00:31:31.690 align:middle line:84% like the ones that they operate in Holland, 00:31:31.690 --> 00:31:34.970 align:middle line:84% typically supply about 40 reactors, 00:31:34.970 --> 00:31:38.290 align:middle line:84% and they could make 800 bombs in a year. 00:31:38.290 --> 00:31:40.850 align:middle line:84% So you can just do the math in your head. 00:31:40.850 --> 00:31:42.928 align:middle line:90% You can make 800 bombs a year. 00:31:42.928 --> 00:31:44.470 align:middle line:84% How many bombs can you make in a day? 00:31:44.470 --> 00:31:46.290 align:middle line:90% How fast could you make a bomb? 00:31:46.290 --> 00:31:48.730 align:middle line:84% It takes about a day to convert the plant or so. 00:31:48.730 --> 00:31:51.810 align:middle line:84% But you're making a bomb within about a day and a half. 00:31:51.810 --> 00:31:54.170 align:middle line:90% Your first weapon quantity. 00:31:54.170 --> 00:31:57.810 align:middle line:84% So you could have IAEA inspectors and whatever. 00:31:57.810 --> 00:31:59.810 align:middle line:84% And they could be there, and they could actually 00:31:59.810 --> 00:32:00.870 align:middle line:90% watch all this happen. 00:32:00.870 --> 00:32:03.170 align:middle line:84% And then they could go to the United Nations and say, 00:32:03.170 --> 00:32:04.870 align:middle line:84% Holland is making a nuclear weapon. 00:32:04.870 --> 00:32:08.400 align:middle line:84% And by the time everyone showed up for the meeting, 00:32:08.400 --> 00:32:11.360 align:middle line:84% they would already have their bomb. 00:32:11.360 --> 00:32:13.600 align:middle line:84% And this is basically the problem 00:32:13.600 --> 00:32:15.840 align:middle line:84% with these large-scale facilities 00:32:15.840 --> 00:32:20.400 align:middle line:84% is you can have safeguards, but it's not clear 00:32:20.400 --> 00:32:23.000 align:middle line:90% that it really means anything. 00:32:23.000 --> 00:32:24.620 align:middle line:90% It doesn't stop anyone. 00:32:24.620 --> 00:32:26.160 align:middle line:90% Yep. 00:32:26.160 --> 00:32:29.700 align:middle line:84% AUDIENCE: Because [INAUDIBLE] so you can make a bomb so quickly. 00:32:29.700 --> 00:32:32.280 align:middle line:84% And it does seem like the centrifuges and enrichment 00:32:32.280 --> 00:32:38.440 align:middle line:84% process is not necessarily difficult to accomplish. 00:32:38.440 --> 00:32:40.120 align:middle line:84% Is there an estimation on nations 00:32:40.120 --> 00:32:43.480 align:middle line:84% that we say do not have a bomb, that probably do? 00:32:43.480 --> 00:32:49.760 align:middle line:84% They just don't have the delivery of it effectively? 00:32:49.760 --> 00:32:52.760 align:middle line:84% SCOTT KEMP: So the general view is we know who has 00:32:52.760 --> 00:32:56.600 align:middle line:84% and who doesn't have nuclear weapons. 00:32:56.600 --> 00:32:58.800 align:middle line:84% And we generally think we know who has and doesn't 00:32:58.800 --> 00:33:02.080 align:middle line:90% have uranium enrichment. 00:33:02.080 --> 00:33:09.500 align:middle line:84% But in theory, it's possible that someone could have done it. 00:33:09.500 --> 00:33:11.540 align:middle line:84% Secrets tend to eventually leak-- 00:33:11.540 --> 00:33:14.840 align:middle line:84% not because they get detected technically, 00:33:14.840 --> 00:33:22.020 align:middle line:84% but because people's phones are tapped and things like this. 00:33:22.020 --> 00:33:26.500 align:middle line:84% So I think the general view in the intelligence community 00:33:26.500 --> 00:33:28.080 align:middle line:90% is that we feel confident. 00:33:28.080 --> 00:33:33.380 align:middle line:84% We know that there are no clandestine weapons out there. 00:33:33.380 --> 00:33:36.720 align:middle line:84% Maybe there's a couple of cases where things are a little fuzzy, 00:33:36.720 --> 00:33:40.940 align:middle line:84% like South Africa, which has a uranium enrichment program, 00:33:40.940 --> 00:33:44.450 align:middle line:84% previously had nuclear weapons, supposedly got rid 00:33:44.450 --> 00:33:46.700 align:middle line:84% of their nuclear weapons, but not a lot of information 00:33:46.700 --> 00:33:50.260 align:middle line:84% was forthcoming about where everything went. 00:33:50.260 --> 00:33:50.900 align:middle line:90% Yeah. 00:33:50.900 --> 00:33:54.100 align:middle line:90% 00:33:54.100 --> 00:33:57.560 align:middle line:84% AUDIENCE: All of this, [INAUDIBLE] for example, 00:33:57.560 --> 00:33:59.700 align:middle line:84% they produced 5% enriched uranium. 00:33:59.700 --> 00:34:01.380 align:middle line:90% SCOTT KEMP: Yes, correct. 00:34:01.380 --> 00:34:03.460 align:middle line:84% AUDIENCE: So why do you assume for them 00:34:03.460 --> 00:34:06.760 align:middle line:84% to be converted into facilities that can use bombs? 00:34:06.760 --> 00:34:11.800 align:middle line:84% Is it 90% of each bomb, or is it just about staggering the gas 00:34:11.800 --> 00:34:13.820 align:middle line:90% centrifuges [INAUDIBLE]? 00:34:13.820 --> 00:34:16.179 align:middle line:90% 00:34:16.179 --> 00:34:19.760 align:middle line:84% SCOTT KEMP: Yeah, so I was going to give a special lecture 00:34:19.760 --> 00:34:22.980 align:middle line:84% on the 19th, but now because of the projector issue, 00:34:22.980 --> 00:34:24.980 align:middle line:84% I'm not sure if I'll be able to do it. 00:34:24.980 --> 00:34:27.560 align:middle line:84% But basically, what you do is you 00:34:27.560 --> 00:34:30.120 align:middle line:84% connect the output of one cascade, 00:34:30.120 --> 00:34:35.480 align:middle line:84% and you feed it into the input of another cascade of machines. 00:34:35.480 --> 00:34:38.159 align:middle line:84% And when you do that, you run the cascade outside 00:34:38.159 --> 00:34:39.300 align:middle line:90% of its design point. 00:34:39.300 --> 00:34:41.719 align:middle line:84% So there's some inefficiency in that process, 00:34:41.719 --> 00:34:44.540 align:middle line:84% but you can somewhat compensate for it. 00:34:44.540 --> 00:34:47.840 align:middle line:84% And you still eventually get there. 00:34:47.840 --> 00:34:50.480 align:middle line:84% You basically connect all these cascades together. 00:34:50.480 --> 00:34:51.900 align:middle line:90% I didn't really tell you. 00:34:51.900 --> 00:34:53.980 align:middle line:84% I said you needed to put a bunch of units. 00:34:53.980 --> 00:34:56.719 align:middle line:90% 00:34:56.719 --> 00:34:59.980 align:middle line:84% The real way it works is that you have a bunch of units, 00:34:59.980 --> 00:35:02.960 align:middle line:90% both in parallel and in series. 00:35:02.960 --> 00:35:07.580 align:middle line:90% 00:35:07.580 --> 00:35:09.300 align:middle line:90% These are all centrifuges. 00:35:09.300 --> 00:35:11.240 align:middle line:84% These are all operating in parallel. 00:35:11.240 --> 00:35:13.740 align:middle line:84% So you can feed material in here. 00:35:13.740 --> 00:35:17.900 align:middle line:84% And then, because of conservation of mass, 00:35:17.900 --> 00:35:21.840 align:middle line:84% if you feed some material in here, let's say one unit, 00:35:21.840 --> 00:35:24.860 align:middle line:84% then you're going to get like 0.4 out here 00:35:24.860 --> 00:35:30.780 align:middle line:90% and/or maybe 0.6 and 0.4 here. 00:35:30.780 --> 00:35:32.380 align:middle line:84% About half the material goes one way, 00:35:32.380 --> 00:35:35.060 align:middle line:84% and half goes the other way, roughly. 00:35:35.060 --> 00:35:39.380 align:middle line:84% So this stage is only dealing with roughly half the total flow 00:35:39.380 --> 00:35:41.520 align:middle line:90% of this stage, as you know. 00:35:41.520 --> 00:35:45.980 align:middle line:84% And all the outputs like this are piped together 00:35:45.980 --> 00:35:50.960 align:middle line:84% and go to the inputs for these machines, and the same here. 00:35:50.960 --> 00:35:53.980 align:middle line:90% 00:35:53.980 --> 00:35:58.857 align:middle line:84% So you have fewer and fewer machines with every stage. 00:35:58.857 --> 00:36:00.440 align:middle line:84% If you just threw away all your stuff, 00:36:00.440 --> 00:36:01.700 align:middle line:90% you would have a lot of waste. 00:36:01.700 --> 00:36:05.420 align:middle line:84% So the way this works is that we have a stage here, 00:36:05.420 --> 00:36:08.080 align:middle line:84% which has outputs going this way, 00:36:08.080 --> 00:36:13.000 align:middle line:84% and we have the tails, the depleted uranium, 00:36:13.000 --> 00:36:19.000 align:middle line:84% slightly depleted coming out as waste from this stage. 00:36:19.000 --> 00:36:22.680 align:middle line:84% But this stuff, even though it's depleted for this stage, 00:36:22.680 --> 00:36:24.780 align:middle line:84% it's still good relative to this stage. 00:36:24.780 --> 00:36:29.000 align:middle line:84% So what we do is we pipe this back in up here like that. 00:36:29.000 --> 00:36:31.200 align:middle line:90% And so it cycles back through. 00:36:31.200 --> 00:36:36.120 align:middle line:84% And so this is how cascades are built. 00:36:36.120 --> 00:36:43.040 align:middle line:84% If you do the solution, so here's a stage, 00:36:43.040 --> 00:36:47.400 align:middle line:84% and you have some product coming in. 00:36:47.400 --> 00:36:48.280 align:middle line:90% Here's the n stage. 00:36:48.280 --> 00:36:51.920 align:middle line:84% From the n minus 1 stage, you have some product 00:36:51.920 --> 00:36:58.180 align:middle line:84% which is enriched, from the n plus 1 stage, some product. 00:36:58.180 --> 00:37:02.540 align:middle line:90% 00:37:02.540 --> 00:37:08.940 align:middle line:84% You have the tails of the waste coming in like that. 00:37:08.940 --> 00:37:11.020 align:middle line:84% Ideally, you want this enrichment 00:37:11.020 --> 00:37:14.260 align:middle line:84% coming out of this stage to be exactly the same as this waste 00:37:14.260 --> 00:37:14.760 align:middle line:90% stream. 00:37:14.760 --> 00:37:18.380 align:middle line:84% Because if they are not the same isotopic ratio, 00:37:18.380 --> 00:37:21.060 align:middle line:84% then what happens is you have a mixture 00:37:21.060 --> 00:37:23.820 align:middle line:84% two different isotopes, fractions, which 00:37:23.820 --> 00:37:26.420 align:middle line:90% is the opposite of separation. 00:37:26.420 --> 00:37:28.340 align:middle line:90% And so it's anti-enrichment. 00:37:28.340 --> 00:37:31.700 align:middle line:84% So to make this entropy process go to 0, 00:37:31.700 --> 00:37:35.020 align:middle line:84% you design the system so that this 00:37:35.020 --> 00:37:38.540 align:middle line:84% has the exact same enrichment as this. 00:37:38.540 --> 00:37:43.420 align:middle line:84% When you switch to cascading multiples of these 00:37:43.420 --> 00:37:47.260 align:middle line:84% together to get to HEU, you lose this optimality. 00:37:47.260 --> 00:37:50.980 align:middle line:84% And so you have some losses, and you have to calculate for that. 00:37:50.980 --> 00:37:54.420 align:middle line:90% But you can still do it. 00:37:54.420 --> 00:37:57.580 align:middle line:84% Yeah, it depends on how you build your plant. 00:37:57.580 --> 00:38:00.690 align:middle line:84% The one day is basically just flushing the gas out 00:38:00.690 --> 00:38:05.550 align:middle line:84% and moving the pipes, the output of this to something else. 00:38:05.550 --> 00:38:09.490 align:middle line:84% But yeah, I think a day is a good estimate 00:38:09.490 --> 00:38:10.970 align:middle line:90% if you've already planned it. 00:38:10.970 --> 00:38:13.490 align:middle line:84% Like you were, OK, here's what we're going to do, 00:38:13.490 --> 00:38:18.910 align:middle line:84% then a day is reasonable for the switchover. 00:38:18.910 --> 00:38:22.990 align:middle line:84% And then probably some time to get your first bomb after that. 00:38:22.990 --> 00:38:28.610 align:middle line:90% 00:38:28.610 --> 00:38:29.530 align:middle line:90% OK. 00:38:29.530 --> 00:38:35.410 align:middle line:84% So that's uranium enrichment and bomb-making 101. 00:38:35.410 --> 00:38:40.930 align:middle line:84% So let me just review some of these pathways. 00:38:40.930 --> 00:38:42.890 align:middle line:84% So we started talking about spent fuel, 00:38:42.890 --> 00:38:44.790 align:middle line:90% and we said spent fuel. 00:38:44.790 --> 00:38:47.270 align:middle line:84% Even the stuff at regular reactors is weapons-usable. 00:38:47.270 --> 00:38:48.390 align:middle line:90% Remember? 00:38:48.390 --> 00:38:50.890 align:middle line:84% I read you this long paragraph from the Department of Energy 00:38:50.890 --> 00:38:55.650 align:middle line:84% that said reactor-grade plutonium can be used in devices 00:38:55.650 --> 00:38:58.030 align:middle line:84% of minimal sophistication by countries 00:38:58.030 --> 00:39:00.550 align:middle line:90% with minimal experience. 00:39:00.550 --> 00:39:02.870 align:middle line:90% And it is weapons-usable. 00:39:02.870 --> 00:39:06.750 align:middle line:84% What protects us is that you have to build this facility 00:39:06.750 --> 00:39:07.972 align:middle line:90% to do this operation. 00:39:07.972 --> 00:39:09.930 align:middle line:84% We talked a little bit about how hard that was. 00:39:09.930 --> 00:39:13.810 align:middle line:84% I talked about this quick and dirty reprocessing plant, 00:39:13.810 --> 00:39:15.270 align:middle line:84% where we use stainless steel drums 00:39:15.270 --> 00:39:18.610 align:middle line:84% that we order from the dairy industry and so on. 00:39:18.610 --> 00:39:21.110 align:middle line:84% So you can do this, but you still have to do it. 00:39:21.110 --> 00:39:24.750 align:middle line:84% You have to build a facility, test it, run it, and so on. 00:39:24.750 --> 00:39:27.690 align:middle line:84% Some countries do this as a matter of course, 00:39:27.690 --> 00:39:30.810 align:middle line:84% and they produce separated plutonium. 00:39:30.810 --> 00:39:33.310 align:middle line:84% And that's worrisome because that's directly diverted, 00:39:33.310 --> 00:39:34.790 align:middle line:90% directly weapons-usable. 00:39:34.790 --> 00:39:37.830 align:middle line:84% We talk about dedicated plutonium production. 00:39:37.830 --> 00:39:41.230 align:middle line:84% This is like the Iraq reactor, but you're talking from scratch. 00:39:41.230 --> 00:39:43.710 align:middle line:84% You can do this under the cover of I'm 00:39:43.710 --> 00:39:45.762 align:middle line:84% building this to train my nuclear engineers. 00:39:45.762 --> 00:39:47.970 align:middle line:84% And because I eventually won a nuclear power program, 00:39:47.970 --> 00:39:49.762 align:middle line:84% and I need to know how to operate reactors, 00:39:49.762 --> 00:39:52.150 align:middle line:84% and I need to test materials to design reactors. 00:39:52.150 --> 00:39:55.250 align:middle line:84% And so I build my own dedicated small plant, 00:39:55.250 --> 00:39:59.370 align:middle line:84% but you're still talking 5 to 10 years to get this thing running. 00:39:59.370 --> 00:40:01.630 align:middle line:84% And so this is not exactly a fast way, 00:40:01.630 --> 00:40:05.290 align:middle line:84% but this is how most weapons programs eventually 00:40:05.290 --> 00:40:10.690 align:middle line:84% operate their weapons production material or material production 00:40:10.690 --> 00:40:13.290 align:middle line:84% because they don't want to have it connected 00:40:13.290 --> 00:40:16.050 align:middle line:90% to their civilian fuel cycle. 00:40:16.050 --> 00:40:17.810 align:middle line:84% And then we talked about enrichment. 00:40:17.810 --> 00:40:21.650 align:middle line:84% And the nice thing is in the light-water reactors, 00:40:21.650 --> 00:40:26.090 align:middle line:84% which run off less than 5%, this is not directly weapons-usable. 00:40:26.090 --> 00:40:30.490 align:middle line:84% So you'd have to have additional enrichment facilities. 00:40:30.490 --> 00:40:34.810 align:middle line:84% And if a country doesn't have that, then they can't do it. 00:40:34.810 --> 00:40:37.530 align:middle line:84% Or you could just build a dedicated facility. 00:40:37.530 --> 00:40:40.750 align:middle line:84% So here are some high-level conclusions. 00:40:40.750 --> 00:40:44.330 align:middle line:84% Any country with a power reactor can make plutonium weapons-- 00:40:44.330 --> 00:40:46.050 align:middle line:90% period. 00:40:46.050 --> 00:40:48.210 align:middle line:84% Doesn't matter what kind of power reactor. 00:40:48.210 --> 00:40:52.310 align:middle line:84% As long as it uses uranium fuel, it's going to make plutonium. 00:40:52.310 --> 00:40:54.630 align:middle line:84% That plutonium can be used in weapons, 00:40:54.630 --> 00:40:59.070 align:middle line:84% and some people will claim it can't be, but it can. 00:40:59.070 --> 00:41:02.830 align:middle line:84% There are, however, radiological barriers that make it hard. 00:41:02.830 --> 00:41:05.270 align:middle line:84% These are eliminated when you do the recycling 00:41:05.270 --> 00:41:08.110 align:middle line:90% and you separate the fuel. 00:41:08.110 --> 00:41:10.010 align:middle line:84% Dedicated plutonium production is ideal, 00:41:10.010 --> 00:41:11.945 align:middle line:90% but it's a decades-long process. 00:41:11.945 --> 00:41:13.570 align:middle line:84% However, if you want to take that path, 00:41:13.570 --> 00:41:16.230 align:middle line:84% it's easier to justify building these dedicated facilities 00:41:16.230 --> 00:41:19.310 align:middle line:84% if you have a nuclear power program. 00:41:19.310 --> 00:41:22.415 align:middle line:84% So you might acquire a little bit of nuclear power 00:41:22.415 --> 00:41:23.790 align:middle line:84% with the idea that you would then 00:41:23.790 --> 00:41:27.550 align:middle line:84% go build a dedicated program for producing plutonium 00:41:27.550 --> 00:41:31.550 align:middle line:84% and say it was part of our nuclear power research program. 00:41:31.550 --> 00:41:33.190 align:middle line:90% Iran did this. 00:41:33.190 --> 00:41:34.290 align:middle line:90% Algeria did this. 00:41:34.290 --> 00:41:36.550 align:middle line:84% A bunch of other countries did this. 00:41:36.550 --> 00:41:38.750 align:middle line:84% And then production of highly enriched uranium 00:41:38.750 --> 00:41:40.990 align:middle line:84% is becoming a proliferation pathway 00:41:40.990 --> 00:41:42.170 align:middle line:90% of choice for new states. 00:41:42.170 --> 00:41:48.230 align:middle line:84% So Iraq, Iran, Libya, Pakistan, North Korea 00:41:48.230 --> 00:41:50.670 align:middle line:84% are all countries that have shifted away from plutonium 00:41:50.670 --> 00:41:53.010 align:middle line:90% to uranium enrichment. 00:41:53.010 --> 00:41:56.250 align:middle line:84% But it's hard to justify building these facilities 00:41:56.250 --> 00:41:59.950 align:middle line:84% unless you have a nuclear power program. 00:41:59.950 --> 00:42:03.490 align:middle line:84% So the way Iran operated its program, 00:42:03.490 --> 00:42:09.650 align:middle line:84% and when we looked at this, the situation with Iran 00:42:09.650 --> 00:42:10.870 align:middle line:90% is still not settled. 00:42:10.870 --> 00:42:14.730 align:middle line:84% But remember, they started in 1985. 00:42:14.730 --> 00:42:16.210 align:middle line:90% That's a long time ago. 00:42:16.210 --> 00:42:17.870 align:middle line:84% Why are we still talking to Iran? 00:42:17.870 --> 00:42:21.690 align:middle line:84% Well, part of the reason is Iran has a nuclear reactor, 00:42:21.690 --> 00:42:24.770 align:middle line:84% and they assert that their enrichment 00:42:24.770 --> 00:42:29.570 align:middle line:84% program is to make peaceful civilian fuel for their reactor. 00:42:29.570 --> 00:42:31.950 align:middle line:84% And that just complicates everything, 00:42:31.950 --> 00:42:36.640 align:middle line:84% because you can't just say no, it's not and bomb it. 00:42:36.640 --> 00:42:38.890 align:middle line:84% The international community says you have to give them 00:42:38.890 --> 00:42:40.110 align:middle line:90% the benefit of the doubt. 00:42:40.110 --> 00:42:42.350 align:middle line:84% There's a bunch of debate in the United Nations, 00:42:42.350 --> 00:42:44.730 align:middle line:84% and things drag on and on and on and on. 00:42:44.730 --> 00:42:51.680 align:middle line:84% And so this connection has enabled Iran to pull this off. 00:42:51.680 --> 00:42:57.680 align:middle line:84% So those factors in blue here are 00:42:57.680 --> 00:43:03.160 align:middle line:84% where technology affects the pathways between nuclear power 00:43:03.160 --> 00:43:04.720 align:middle line:90% and nuclear weapons. 00:43:04.720 --> 00:43:10.160 align:middle line:84% These factors in red relate to the political decisions 00:43:10.160 --> 00:43:11.503 align:middle line:90% to pursue nuclear weapons. 00:43:11.503 --> 00:43:13.920 align:middle line:84% And you can see that there's an intimate connection, which 00:43:13.920 --> 00:43:16.360 align:middle line:84% is where I started out in the previous lecture, 00:43:16.360 --> 00:43:19.360 align:middle line:84% that, yes, the decision to acquire nuclear weapons 00:43:19.360 --> 00:43:22.880 align:middle line:84% is a political choice, but the political choice 00:43:22.880 --> 00:43:26.280 align:middle line:84% is intimately dependent on the technology 00:43:26.280 --> 00:43:28.280 align:middle line:90% that you have on hand. 00:43:28.280 --> 00:43:33.080 align:middle line:84% And that is changed by having nuclear power. 00:43:33.080 --> 00:43:36.160 align:middle line:84% So some high-level conclusions is 00:43:36.160 --> 00:43:39.560 align:middle line:84% that nuclear power will, of course, ease access 00:43:39.560 --> 00:43:43.360 align:middle line:84% to weapons materials, plutonium, or highly enriched uranium. 00:43:43.360 --> 00:43:48.260 align:middle line:84% But you can change this a little bit 00:43:48.260 --> 00:43:49.800 align:middle line:84% by trying to change the technology, 00:43:49.800 --> 00:43:53.060 align:middle line:84% but it turns out virtually all the technologies we 00:43:53.060 --> 00:43:55.820 align:middle line:84% talk about now going forward, whether they're microreactors 00:43:55.820 --> 00:43:58.460 align:middle line:84% that are fueled with HALEU or gen 00:43:58.460 --> 00:44:00.980 align:middle line:90% 4 reactor concepts, et cetera. 00:44:00.980 --> 00:44:03.380 align:middle line:84% All of these actually make it easier 00:44:03.380 --> 00:44:06.260 align:middle line:84% to get access to these weapons material relative to the ones 00:44:06.260 --> 00:44:09.340 align:middle line:84% through light-water reactor that we have today. 00:44:09.340 --> 00:44:13.420 align:middle line:84% There's almost no idea that has been put forward 00:44:13.420 --> 00:44:14.940 align:middle line:90% that makes it harder. 00:44:14.940 --> 00:44:17.460 align:middle line:84% The one thing that kind of makes it a little bit harder 00:44:17.460 --> 00:44:20.940 align:middle line:84% are TRISO particles are a little bit harder to process 00:44:20.940 --> 00:44:27.380 align:middle line:84% than uranium oxide pellets, but is not that much harder. 00:44:27.380 --> 00:44:30.480 align:middle line:84% You can take the TRISO, and you can put it in a blender, 00:44:30.480 --> 00:44:31.820 align:middle line:90% literally. 00:44:31.820 --> 00:44:34.500 align:middle line:84% And it breaks it up enough that you can then acid 00:44:34.500 --> 00:44:36.660 align:middle line:90% leach the uranium out of it. 00:44:36.660 --> 00:44:39.300 align:middle line:84% So OK, you have to add an extra step 00:44:39.300 --> 00:44:46.020 align:middle line:84% where you're blending this stuff, but it's not a big deal. 00:44:46.020 --> 00:44:51.920 align:middle line:84% Now, even if there were a hypothetical reactor that 00:44:51.920 --> 00:44:54.800 align:middle line:84% was more proliferation-resistant, 00:44:54.800 --> 00:44:57.960 align:middle line:90% could we build and sell that? 00:44:57.960 --> 00:45:00.160 align:middle line:90% And the answer is probably not. 00:45:00.160 --> 00:45:04.800 align:middle line:84% Because while we as reactor exporters 00:45:04.800 --> 00:45:11.720 align:middle line:84% may not care what that cost, the buyers will care. 00:45:11.720 --> 00:45:15.440 align:middle line:84% And they are not going to buy a reactor that 00:45:15.440 --> 00:45:18.460 align:middle line:84% is bad for proliferation if it costs more. 00:45:18.460 --> 00:45:22.550 align:middle line:84% And if it costs less, we would have already invented it. 00:45:22.550 --> 00:45:24.800 align:middle line:84% We spend a lot of time trying to make nuclear power as 00:45:24.800 --> 00:45:26.120 align:middle line:90% cheap as possible. 00:45:26.120 --> 00:45:30.600 align:middle line:84% So it's unlikely that any kind of proliferation-resistant 00:45:30.600 --> 00:45:34.040 align:middle line:84% reactors is forthcoming and competitive 00:45:34.040 --> 00:45:35.380 align:middle line:90% on the international stage. 00:45:35.380 --> 00:45:37.640 align:middle line:84% If we had a monopoly, we were the only suppliers 00:45:37.640 --> 00:45:39.280 align:middle line:84% of reactors to the world, then maybe we 00:45:39.280 --> 00:45:41.580 align:middle line:84% can force it down the throat of various countries. 00:45:41.580 --> 00:45:43.960 align:middle line:90% But we're not a monopoly. 00:45:43.960 --> 00:45:47.820 align:middle line:84% Buy reactors from Russia and from Korea, and others. 00:45:47.820 --> 00:45:52.060 align:middle line:90% So what about number two? 00:45:52.060 --> 00:45:56.220 align:middle line:84% Number two-- you can actually change the justification 00:45:56.220 --> 00:45:57.860 align:middle line:90% through policies. 00:45:57.860 --> 00:46:00.460 align:middle line:84% And let's just have a quick look. 00:46:00.460 --> 00:46:02.380 align:middle line:84% So it used to be people talked about doing 00:46:02.380 --> 00:46:05.180 align:middle line:84% this all the time-- closed fuel cycle, recycling. 00:46:05.180 --> 00:46:06.420 align:middle line:90% Sounds great. 00:46:06.420 --> 00:46:11.140 align:middle line:84% But under President Ford, the United States 00:46:11.140 --> 00:46:14.100 align:middle line:84% decided to make a declaratory policy that it would not 00:46:14.100 --> 00:46:18.740 align:middle line:84% allow reprocessing of its civilian uranium, 00:46:18.740 --> 00:46:22.700 align:middle line:84% or we're not supported rather, and would not 00:46:22.700 --> 00:46:27.540 align:middle line:84% allow foreign countries to reprocess US-origin fuels 00:46:27.540 --> 00:46:32.460 align:middle line:84% or fuels that touched US-origin technology, unless they 00:46:32.460 --> 00:46:34.673 align:middle line:84% had explicit permission from the United States. 00:46:34.673 --> 00:46:36.840 align:middle line:84% And only one country has been given that permission. 00:46:36.840 --> 00:46:37.940 align:middle line:90% That is Japan. 00:46:37.940 --> 00:46:39.940 align:middle line:84% Every other country that has a nuclear agreement 00:46:39.940 --> 00:46:42.460 align:middle line:84% with the United States, Nuclear Cooperation Agreement, 00:46:42.460 --> 00:46:47.200 align:middle line:84% has basically signed away their right to do this number two. 00:46:47.200 --> 00:46:53.360 align:middle line:84% And so that is a policy that helps prevent the ready access 00:46:53.360 --> 00:46:54.660 align:middle line:90% to direct-use materials. 00:46:54.660 --> 00:46:56.240 align:middle line:90% So people get stuck here. 00:46:56.240 --> 00:46:58.920 align:middle line:84% And then they have to build that dedicated facility 00:46:58.920 --> 00:47:00.920 align:middle line:90% to do the extraction. 00:47:00.920 --> 00:47:04.400 align:middle line:90% So that's good. 00:47:04.400 --> 00:47:07.220 align:middle line:84% You might ask why Japan got an got an exception. 00:47:07.220 --> 00:47:08.400 align:middle line:90% That was in 1977. 00:47:08.400 --> 00:47:12.200 align:middle line:84% Prime Minister Fukuda visited President Carter 00:47:12.200 --> 00:47:18.160 align:middle line:84% and personally asked the president for an exception, 00:47:18.160 --> 00:47:22.060 align:middle line:84% and the ambassador to Japan, Ambassador Mansfield, 00:47:22.060 --> 00:47:25.460 align:middle line:84% characterized Fukuda's plea as basically telling the president, 00:47:25.460 --> 00:47:29.240 align:middle line:84% this is a matter of life or death for Japan. 00:47:29.240 --> 00:47:34.440 align:middle line:84% And basically, that we need the ability to make a nuclear weapon 00:47:34.440 --> 00:47:39.720 align:middle line:84% in the event that our adversaries, the Soviet Union, 00:47:39.720 --> 00:47:40.820 align:middle line:90% come after us. 00:47:40.820 --> 00:47:45.940 align:middle line:84% And the US gave in and allowed Japan to do this. 00:47:45.940 --> 00:47:49.500 align:middle line:84% And that's why Japan operates the Rokkasho Reprocessing Plant 00:47:49.500 --> 00:47:52.820 align:middle line:84% today is they have that special exception. 00:47:52.820 --> 00:47:57.820 align:middle line:84% So now Iran says we just wish to be treated the same as Japan. 00:47:57.820 --> 00:48:00.100 align:middle line:84% And I've had the ambassador to Iran personally tell me 00:48:00.100 --> 00:48:02.820 align:middle line:84% this multiple times, and it's really hard 00:48:02.820 --> 00:48:05.040 align:middle line:84% to sit there and say, oh no, sorry, you're different. 00:48:05.040 --> 00:48:05.880 align:middle line:90% We don't trust you. 00:48:05.880 --> 00:48:12.100 align:middle line:84% So keeping these red lines in the international community 00:48:12.100 --> 00:48:13.140 align:middle line:90% are really hard. 00:48:13.140 --> 00:48:18.180 align:middle line:84% We recently produced a similar red line for-- 00:48:18.180 --> 00:48:19.760 align:middle line:90% there's our policy there. 00:48:19.760 --> 00:48:25.180 align:middle line:84% We recently produced a similar red line for enrichment. 00:48:25.180 --> 00:48:28.740 align:middle line:84% We did a deal with United Arab Emirates 00:48:28.740 --> 00:48:30.480 align:middle line:84% when they wanted to build their reactors. 00:48:30.480 --> 00:48:33.320 align:middle line:84% And we said, not only will you not reprocess, 00:48:33.320 --> 00:48:36.820 align:middle line:84% but you will also not have domestic enrichment. 00:48:36.820 --> 00:48:42.210 align:middle line:84% And they accepted that, and that has been called now 00:48:42.210 --> 00:48:43.810 align:middle line:90% the gold standard. 00:48:43.810 --> 00:48:49.370 align:middle line:84% But unfortunately, it seems like the gold standard was a one-off. 00:48:49.370 --> 00:48:52.970 align:middle line:84% Biden, his officials started to talk 00:48:52.970 --> 00:48:55.450 align:middle line:84% to Saudi Arabia, who wants a nuclear power 00:48:55.450 --> 00:48:56.910 align:middle line:90% program for whatever reason. 00:48:56.910 --> 00:48:59.210 align:middle line:84% It's hard to imagine why because there's 00:48:59.210 --> 00:49:02.810 align:middle line:90% a lot of sun in Saudi Arabia. 00:49:02.810 --> 00:49:07.810 align:middle line:84% And Saudis kept saying they would not accept this provision. 00:49:07.810 --> 00:49:13.370 align:middle line:84% And the industry was pushing the Biden government so forcefully 00:49:13.370 --> 00:49:17.810 align:middle line:84% to do a deal with Saudi so they could export reactors 00:49:17.810 --> 00:49:20.810 align:middle line:84% to Saudi, that Biden's administration was 00:49:20.810 --> 00:49:22.520 align:middle line:90% willing to look past it. 00:49:22.520 --> 00:49:24.770 align:middle line:84% Then the Saudis seem to have lobbied a bunch of people 00:49:24.770 --> 00:49:27.930 align:middle line:84% in Congress, and they are now expressing support 00:49:27.930 --> 00:49:29.530 align:middle line:84% for a Saudi deal that doesn't have 00:49:29.530 --> 00:49:31.090 align:middle line:90% a restriction on enrichment. 00:49:31.090 --> 00:49:32.890 align:middle line:84% And now the Trump administration is indeed 00:49:32.890 --> 00:49:34.930 align:middle line:84% negotiating that agreement right now. 00:49:34.930 --> 00:49:39.470 align:middle line:84% And from all publicly leaked information, 00:49:39.470 --> 00:49:42.350 align:middle line:84% it appears not to have any restriction limiting Saudi 00:49:42.350 --> 00:49:44.270 align:middle line:90% from doing this enrichment. 00:49:44.270 --> 00:49:48.030 align:middle line:84% So it's a hard thing to pull off. 00:49:48.030 --> 00:49:51.455 align:middle line:84% And oh, by the way, the Emiratis who did sign up, 00:49:51.455 --> 00:49:53.830 align:middle line:84% have an escape clause that says anyone else in the region 00:49:53.830 --> 00:49:54.850 align:middle line:90% gets to have enrichment. 00:49:54.850 --> 00:49:55.670 align:middle line:90% So do we. 00:49:55.670 --> 00:49:57.250 align:middle line:84% So as soon as we do the Saudi deal, 00:49:57.250 --> 00:50:00.010 align:middle line:84% we'll destroy the Emirati deal at the same time. 00:50:00.010 --> 00:50:02.590 align:middle line:90% 00:50:02.590 --> 00:50:04.730 align:middle line:90% So why is it hard? 00:50:04.730 --> 00:50:07.390 align:middle line:84% Well, I would say the real reason 00:50:07.390 --> 00:50:10.070 align:middle line:84% it's probably hard to get these agreements is 00:50:10.070 --> 00:50:18.390 align:middle line:84% that many of these countries want nuclear power because it 00:50:18.390 --> 00:50:22.790 align:middle line:84% gives them a nuclear weapons option. 00:50:22.790 --> 00:50:24.615 align:middle line:84% Emiratis probably wanted it principally 00:50:24.615 --> 00:50:25.990 align:middle line:84% as a matter of status, and that's 00:50:25.990 --> 00:50:29.690 align:middle line:84% why they were willing to sign that agreement. 00:50:29.690 --> 00:50:32.590 align:middle line:84% But I think a lot of countries are 00:50:32.590 --> 00:50:36.570 align:middle line:84% willing to spend the money because it's like, well, it's 00:50:36.570 --> 00:50:39.010 align:middle line:84% expensive, but we'll be able to make 00:50:39.010 --> 00:50:41.210 align:middle line:90% nuclear weapons if we need to. 00:50:41.210 --> 00:50:43.950 align:middle line:90% And that's worth it for them. 00:50:43.950 --> 00:50:46.850 align:middle line:84% But they're not going to sign away their right to do that. 00:50:46.850 --> 00:50:49.470 align:middle line:90% So I think that's in my opinion. 00:50:49.470 --> 00:50:52.250 align:middle line:84% And it's just an opinion-- the real reason why this is hard. 00:50:52.250 --> 00:50:55.530 align:middle line:84% But it's also logically, it should be hard 00:50:55.530 --> 00:51:00.290 align:middle line:84% because if you're going to buy a $10 billion nuclear reactor, 00:51:00.290 --> 00:51:04.770 align:middle line:84% you better have fuel to put in it because you've paid up front. 00:51:04.770 --> 00:51:06.810 align:middle line:84% And if you have to depend on a foreign country 00:51:06.810 --> 00:51:09.950 align:middle line:84% to provide that fuel to you, that is not energy security. 00:51:09.950 --> 00:51:11.890 align:middle line:84% That's the opposite of energy security. 00:51:11.890 --> 00:51:14.390 align:middle line:84% So people talk about nuclear being good for energy security, 00:51:14.390 --> 00:51:19.050 align:middle line:84% but it is not unless you have the full front end of the fuel 00:51:19.050 --> 00:51:21.330 align:middle line:84% cycle under your control, including 00:51:21.330 --> 00:51:23.690 align:middle line:84% the fabrication of the fuel and having 00:51:23.690 --> 00:51:27.110 align:middle line:84% validated the design of the fuel that you can fabricate. 00:51:27.110 --> 00:51:30.850 align:middle line:84% And you know that fuel is safe to put in the reactor and so on. 00:51:30.850 --> 00:51:32.990 align:middle line:90% And this is kind of the problem. 00:51:32.990 --> 00:51:36.350 align:middle line:84% This is where the evidence shows that this is not really what 00:51:36.350 --> 00:51:38.390 align:middle line:90% is behind people's motivations. 00:51:38.390 --> 00:51:42.590 align:middle line:84% Countries like Iran or Japan say we 00:51:42.590 --> 00:51:46.790 align:middle line:84% need to have this enrichment capacity because we 00:51:46.790 --> 00:51:48.710 align:middle line:90% need energy security-- 00:51:48.710 --> 00:51:49.790 align:middle line:90% sounds a valid argument. 00:51:49.790 --> 00:51:52.290 align:middle line:84% And then you say, and do you also have the fuel fabrication? 00:51:52.290 --> 00:51:54.150 align:middle line:90% They're like, oh no. 00:51:54.150 --> 00:51:55.290 align:middle line:90% And are you pursuing it? 00:51:55.290 --> 00:51:57.070 align:middle line:90% Well, not really. 00:51:57.070 --> 00:52:01.370 align:middle line:84% So it's kind of just a fake argument, 00:52:01.370 --> 00:52:02.930 align:middle line:90% but it is a legitimate argument. 00:52:02.930 --> 00:52:05.130 align:middle line:84% If you really care about energy security, 00:52:05.130 --> 00:52:07.010 align:middle line:90% you ought to try to have this. 00:52:07.010 --> 00:52:09.390 align:middle line:84% And so you have this issue where energy security 00:52:09.390 --> 00:52:11.150 align:middle line:84% and nonproliferation are directly 00:52:11.150 --> 00:52:12.990 align:middle line:90% at odds with each other. 00:52:12.990 --> 00:52:16.050 align:middle line:84% And if someone is advocating one for one, 00:52:16.050 --> 00:52:18.630 align:middle line:84% they must be advocating against the other. 00:52:18.630 --> 00:52:22.230 align:middle line:90% And this is a problem. 00:52:22.230 --> 00:52:22.730 align:middle line:90% All right. 00:52:22.730 --> 00:52:24.870 align:middle line:84% So the bottom line on technology is 00:52:24.870 --> 00:52:28.150 align:middle line:84% you can make it somewhat politically more expensive. 00:52:28.150 --> 00:52:29.750 align:middle line:84% But once the state has nuclear power, 00:52:29.750 --> 00:52:31.750 align:middle line:84% it's kind of too late because you will always be 00:52:31.750 --> 00:52:34.970 align:middle line:90% able to do this pathway here. 00:52:34.970 --> 00:52:38.770 align:middle line:84% And we saw that took about six months of planning 00:52:38.770 --> 00:52:40.550 align:middle line:90% and three weeks of execution. 00:52:40.550 --> 00:52:45.050 align:middle line:84% So it's maybe not the best way to have 00:52:45.050 --> 00:52:47.070 align:middle line:84% to deal with this reactor-grade plutonium, 00:52:47.070 --> 00:52:51.490 align:middle line:84% but it's always an option once you have that reactor on site. 00:52:51.490 --> 00:52:58.490 align:middle line:84% So what I have argued so far is that there 00:52:58.490 --> 00:53:03.410 align:middle line:84% could be a connection between nuclear power 00:53:03.410 --> 00:53:05.290 align:middle line:90% and nuclear proliferation. 00:53:05.290 --> 00:53:08.690 align:middle line:90% It's there for the taking. 00:53:08.690 --> 00:53:10.610 align:middle line:90% It's technologically there. 00:53:10.610 --> 00:53:13.250 align:middle line:84% But maybe we want to ask a different question, which 00:53:13.250 --> 00:53:17.930 align:middle line:84% is there in fact a connection historically? 00:53:17.930 --> 00:53:24.250 align:middle line:84% Is this the way countries actually pursue nuclear weapons? 00:53:24.250 --> 00:53:26.530 align:middle line:90% Bouncy, bouncy. 00:53:26.530 --> 00:53:32.990 align:middle line:84% So this is a tricky question to answer because it always 00:53:32.990 --> 00:53:36.190 align:middle line:84% depends on which country you choose to look at. 00:53:36.190 --> 00:53:42.260 align:middle line:84% And I mentioned, for example, that-- oh, it's 00:53:42.260 --> 00:53:43.010 align:middle line:90% not on this slide. 00:53:43.010 --> 00:53:44.093 align:middle line:90% It's a much earlier slide. 00:53:44.093 --> 00:53:48.870 align:middle line:84% There are countries like France and Sweden, 00:53:48.870 --> 00:53:53.030 align:middle line:84% who specifically acquired their civilian nuclear programs 00:53:53.030 --> 00:53:54.910 align:middle line:84% in a way that gave them the ability 00:53:54.910 --> 00:53:57.250 align:middle line:90% to make nuclear weapons-- 00:53:57.250 --> 00:54:00.430 align:middle line:90% Taiwan, another one. 00:54:00.430 --> 00:54:03.350 align:middle line:84% So here's how I propose we look at it. 00:54:03.350 --> 00:54:06.310 align:middle line:84% Here's a list of all the states that 00:54:06.310 --> 00:54:08.870 align:middle line:90% have pursued nuclear weapons. 00:54:08.870 --> 00:54:12.950 align:middle line:84% There are countries that have weapons, that had a weapon 00:54:12.950 --> 00:54:13.630 align:middle line:90% but gave it up-- 00:54:13.630 --> 00:54:15.390 align:middle line:90% South Africa, just one-- 00:54:15.390 --> 00:54:18.790 align:middle line:84% sought weapons, wanted weapons but didn't really build 00:54:18.790 --> 00:54:20.690 align:middle line:84% a program, and then didn't want weapons 00:54:20.690 --> 00:54:22.950 align:middle line:84% but wanted the option for weapons. 00:54:22.950 --> 00:54:25.790 align:middle line:84% So they wanted the capacity to build weapons, 00:54:25.790 --> 00:54:27.550 align:middle line:90% all these different countries. 00:54:27.550 --> 00:54:30.960 align:middle line:84% And here you can see their technology, 00:54:30.960 --> 00:54:35.660 align:middle line:84% whether it was dedicated to military purposes or civilian. 00:54:35.660 --> 00:54:36.843 align:middle line:90% They are experts. 00:54:36.843 --> 00:54:39.260 align:middle line:84% Where did the people come from-- from the military program 00:54:39.260 --> 00:54:41.660 align:middle line:90% or from a civilian program? 00:54:41.660 --> 00:54:44.620 align:middle line:84% And the facilities that they actually used. 00:54:44.620 --> 00:54:46.680 align:middle line:84% So they might have used a civilian technology, 00:54:46.680 --> 00:54:50.460 align:middle line:84% but they put it into a dedicated weapons facility. 00:54:50.460 --> 00:54:53.100 align:middle line:84% So they copied the technology, but they 00:54:53.100 --> 00:54:56.060 align:middle line:84% didn't use a civilian plant kind of a thing. 00:54:56.060 --> 00:54:59.720 align:middle line:84% So these are your configurations. 00:54:59.720 --> 00:55:02.140 align:middle line:84% And if I highlight in yellow all those that 00:55:02.140 --> 00:55:07.820 align:middle line:84% have the word civil somewhere, you see it's 70% of them. 00:55:07.820 --> 00:55:10.420 align:middle line:84% But really, we should remove these 00:55:10.420 --> 00:55:18.300 align:middle line:84% because the United States and Russia and China and United 00:55:18.300 --> 00:55:21.700 align:middle line:84% Kingdom had weapons before civilian nuclear power was 00:55:21.700 --> 00:55:23.340 align:middle line:90% a thing. 00:55:23.340 --> 00:55:28.760 align:middle line:84% So really, if you take those out, it's 80% of states. 00:55:28.760 --> 00:55:32.040 align:middle line:84% But this is slightly the wrong question. 00:55:32.040 --> 00:55:34.640 align:middle line:84% This is answering the number of countries 00:55:34.640 --> 00:55:36.100 align:middle line:90% that have pursued weapons. 00:55:36.100 --> 00:55:39.080 align:middle line:84% How many use civilian technology? 00:55:39.080 --> 00:55:41.320 align:middle line:84% We are interested in the opposite question. 00:55:41.320 --> 00:55:43.480 align:middle line:84% If you use civilian technology, what 00:55:43.480 --> 00:55:48.160 align:middle line:84% is the probability that you will pursue a weapon? 00:55:48.160 --> 00:55:50.040 align:middle line:84% That's what matters going forward. 00:55:50.040 --> 00:55:54.040 align:middle line:84% But thankfully, we have Bayes' formula. 00:55:54.040 --> 00:55:59.400 align:middle line:84% So the probability of a weapons program given nuclear power 00:55:59.400 --> 00:56:02.360 align:middle line:84% is the probability of a weapons program times the probability 00:56:02.360 --> 00:56:04.387 align:middle line:84% of nuclear power given a weapons program 00:56:04.387 --> 00:56:06.220 align:middle line:84% divided by the probability of nuclear power. 00:56:06.220 --> 00:56:08.080 align:middle line:90% And so you can do this. 00:56:08.080 --> 00:56:11.020 align:middle line:84% The problem is that we don't know how many weapons. 00:56:11.020 --> 00:56:13.740 align:middle line:90% 00:56:13.740 --> 00:56:17.280 align:middle line:84% We don't have this coefficient x, but it just cancels. 00:56:17.280 --> 00:56:23.520 align:middle line:84% And so there are 24 countries that have had weapons programs-- 00:56:23.520 --> 00:56:27.220 align:middle line:84% 25 I guess-- and 32 countries that have nuclear power 00:56:27.220 --> 00:56:28.900 align:middle line:90% programs in our sample. 00:56:28.900 --> 00:56:31.700 align:middle line:84% So I do this, and I basically estimate 00:56:31.700 --> 00:56:35.020 align:middle line:84% that there's a 62% chance that if you build 00:56:35.020 --> 00:56:37.940 align:middle line:84% civilian nuclear power, you'll eventually decide 00:56:37.940 --> 00:56:39.960 align:middle line:90% to use it for nuclear weapons. 00:56:39.960 --> 00:56:43.140 align:middle line:84% That's the idea behind this calculation. 00:56:43.140 --> 00:56:46.640 align:middle line:84% So let's put this in perspective. 00:56:46.640 --> 00:56:49.220 align:middle line:90% 00:56:49.220 --> 00:56:56.300 align:middle line:84% What is the arrival rate for nuclear weapon programs? 00:56:56.300 --> 00:56:58.560 align:middle line:84% Well, it depends on how you count. 00:56:58.560 --> 00:57:02.600 align:middle line:84% If you count since 1940, there's about 25 weapon programs. 00:57:02.600 --> 00:57:04.580 align:middle line:90% You get one every three years. 00:57:04.580 --> 00:57:07.440 align:middle line:84% Nonproliferation has gotten a little bit better. 00:57:07.440 --> 00:57:11.740 align:middle line:84% Ambitions of countries that want weapons have settled down. 00:57:11.740 --> 00:57:16.380 align:middle line:84% So if you look since 1990, it's about one every six years. 00:57:16.380 --> 00:57:19.820 align:middle line:84% That depends on the stability of the international system 00:57:19.820 --> 00:57:21.700 align:middle line:84% and a whole bunch of other things. 00:57:21.700 --> 00:57:24.860 align:middle line:84% Let's just say, since the international system is getting 00:57:24.860 --> 00:57:27.140 align:middle line:84% a little less stable and it makes the math easy, 00:57:27.140 --> 00:57:31.760 align:middle line:84% let's say one program every five years is what we might expect-- 00:57:31.760 --> 00:57:35.280 align:middle line:84% one weapons program to emerge every five years. 00:57:35.280 --> 00:57:38.100 align:middle line:90% So we have 0.2 cases per year. 00:57:38.100 --> 00:57:39.600 align:middle line:84% From the Bayes calculation, we could 00:57:39.600 --> 00:57:43.000 align:middle line:84% say 63% of those exist because of nuclear power. 00:57:43.000 --> 00:57:49.160 align:middle line:84% The others are not dependent in any way on nuclear power. 00:57:49.160 --> 00:57:50.680 align:middle line:84% So the contribution of nuclear power 00:57:50.680 --> 00:57:54.240 align:middle line:84% makes us about a tenth of a program per year, tenth 00:57:54.240 --> 00:57:57.240 align:middle line:90% of a weapons program per year. 00:57:57.240 --> 00:57:59.640 align:middle line:84% So what does this tenth of a weapons program per year 00:57:59.640 --> 00:58:00.880 align:middle line:90% cost us? 00:58:00.880 --> 00:58:02.920 align:middle line:90% Does everyone understand this? 00:58:02.920 --> 00:58:04.680 align:middle line:90% Yeah, all right. 00:58:04.680 --> 00:58:06.120 align:middle line:90% So I don't know. 00:58:06.120 --> 00:58:08.120 align:middle line:84% Here's three ways of trying to come up 00:58:08.120 --> 00:58:12.440 align:middle line:84% with a cost of proliferation activity. 00:58:12.440 --> 00:58:14.600 align:middle line:84% We could say, well, we're going to sanction them 00:58:14.600 --> 00:58:16.800 align:middle line:90% like we did with Iran. 00:58:16.800 --> 00:58:20.120 align:middle line:84% There's a certain cost with doing sanctions. 00:58:20.120 --> 00:58:23.280 align:middle line:84% We could go to war with them, like we did with Iraq. 00:58:23.280 --> 00:58:25.620 align:middle line:84% There's a cost associated with that. 00:58:25.620 --> 00:58:28.540 align:middle line:84% Or we could just let them have nuclear weapons 00:58:28.540 --> 00:58:30.260 align:middle line:90% and see what happens. 00:58:30.260 --> 00:58:33.560 align:middle line:84% And so let's try to put numbers on these. 00:58:33.560 --> 00:58:35.820 align:middle line:84% This is very, very, very, very rough. 00:58:35.820 --> 00:58:38.820 align:middle line:84% But we're just trying to get a sense of whether proliferation 00:58:38.820 --> 00:58:42.860 align:middle line:84% is a big thing, a big externality on nuclear power, 00:58:42.860 --> 00:58:46.780 align:middle line:84% or not really a big externality on nuclear power. 00:58:46.780 --> 00:58:50.140 align:middle line:84% We need some notion about whether we 00:58:50.140 --> 00:58:52.140 align:middle line:90% should be worried about this. 00:58:52.140 --> 00:58:52.640 align:middle line:90% All right. 00:58:52.640 --> 00:58:56.580 align:middle line:84% So here's the cost of Iran sanctions in 2025 dollars 00:58:56.580 --> 00:58:58.340 align:middle line:90% since they started. 00:58:58.340 --> 00:59:07.320 align:middle line:84% And it's estimated since 1995 to be about $360 billion. 00:59:07.320 --> 00:59:09.480 align:middle line:84% You can see most of it is exports, 00:59:09.480 --> 00:59:12.030 align:middle line:84% foreign direct investment, oil swaps. 00:59:12.030 --> 00:59:14.280 align:middle line:84% There are some things, which are too hard to estimate, 00:59:14.280 --> 00:59:16.540 align:middle line:90% not in here. 00:59:16.540 --> 00:59:22.640 align:middle line:84% So $360 billion per proliferator. 00:59:22.640 --> 00:59:25.390 align:middle line:84% Now, this actually will probably continue for a while, 00:59:25.390 --> 00:59:27.140 align:middle line:84% so this number will eventually get bigger. 00:59:27.140 --> 00:59:28.780 align:middle line:90% But this is where we are today. 00:59:28.780 --> 00:59:31.480 align:middle line:90% 00:59:31.480 --> 00:59:35.520 align:middle line:84% Let's look at the cost of option two-- cost of a war. 00:59:35.520 --> 00:59:37.920 align:middle line:84% So this is the cost of the Iraq War. 00:59:37.920 --> 00:59:44.600 align:middle line:84% Inflated it to 2025 dollars, and it cost us about $1.7 trillion, 00:59:44.600 --> 00:59:47.960 align:middle line:84% about $1 trillion direct spending to fight the war. 00:59:47.960 --> 00:59:51.440 align:middle line:84% And then the rest of this is debts, value of a life, 00:59:51.440 --> 00:59:54.440 align:middle line:84% and various other things, greenhouse gases 00:59:54.440 --> 00:59:59.960 align:middle line:84% from all the petroleum reused, things like that. 00:59:59.960 --> 01:00:02.280 align:middle line:90% All right. 01:00:02.280 --> 01:00:05.120 align:middle line:90% What about number three? 01:00:05.120 --> 01:00:09.000 align:middle line:84% What is the cost of letting countries have nuclear weapons? 01:00:09.000 --> 01:00:11.740 align:middle line:90% So this one is very tricky. 01:00:11.740 --> 01:00:16.120 align:middle line:90% 01:00:16.120 --> 01:00:18.360 align:middle line:84% Here's, I think, the upper bound. 01:00:18.360 --> 01:00:20.684 align:middle line:90% So has anyone seen this photo? 01:00:20.684 --> 01:00:21.726 align:middle line:90% AUDIENCE: It's Hiroshima. 01:00:21.726 --> 01:00:23.950 align:middle line:84% SCOTT KEMP: Yeah, it's Hiroshima. 01:00:23.950 --> 01:00:29.350 align:middle line:84% And I like to ask, did they really 01:00:29.350 --> 01:00:31.350 align:middle line:84% have just a few buildings like this? 01:00:31.350 --> 01:00:34.150 align:middle line:90% What is all this? 01:00:34.150 --> 01:00:38.750 align:middle line:84% It was made of wood, so it burned. 01:00:38.750 --> 01:00:41.350 align:middle line:84% And it created this huge megafire 01:00:41.350 --> 01:00:46.150 align:middle line:84% that injected a bunch of soot into the stratosphere, 01:00:46.150 --> 01:00:48.110 align:middle line:90% like a volcano going off. 01:00:48.110 --> 01:00:53.510 align:middle line:84% And as a result, it actually cooled the ground 01:00:53.510 --> 01:00:56.390 align:middle line:84% from a lack of sunlight hitting the ground. 01:00:56.390 --> 01:01:00.430 align:middle line:84% And this is the principal concern 01:01:00.430 --> 01:01:04.110 align:middle line:90% of a major nuclear conflict. 01:01:04.110 --> 01:01:07.830 align:middle line:84% So this is a very much debated issue 01:01:07.830 --> 01:01:09.090 align:middle line:90% in the scientific literature. 01:01:09.090 --> 01:01:13.150 align:middle line:84% I myself worked on this, and I believe that these claims 01:01:13.150 --> 01:01:16.550 align:middle line:90% are overblown personally. 01:01:16.550 --> 01:01:22.130 align:middle line:84% But I will give you the claim that others have made. 01:01:22.130 --> 01:01:24.950 align:middle line:90% So this is the soot. 01:01:24.950 --> 01:01:28.370 align:middle line:84% This is the consequences of injecting soot 01:01:28.370 --> 01:01:33.750 align:middle line:84% into the atmosphere from 100 Hiroshima-sized weapons. 01:01:33.750 --> 01:01:36.490 align:middle line:84% So you imagine India and Pakistan. 01:01:36.490 --> 01:01:38.770 align:middle line:90% They both have nuclear weapons. 01:01:38.770 --> 01:01:42.330 align:middle line:84% They're constantly at each other's throats. 01:01:42.330 --> 01:01:47.210 align:middle line:84% Something happens in Kashmir, causes a conflict to erupt. 01:01:47.210 --> 01:01:48.810 align:middle line:84% A small number of nuclear weapons 01:01:48.810 --> 01:01:53.130 align:middle line:90% are used, just 100 in total. 01:01:53.130 --> 01:01:55.170 align:middle line:90% What happens? 01:01:55.170 --> 01:01:58.410 align:middle line:84% This is the climate model prediction 01:01:58.410 --> 01:02:01.050 align:middle line:90% from these guys, Toon et Al. 01:02:01.050 --> 01:02:04.890 align:middle line:90% And they estimate cooling. 01:02:04.890 --> 01:02:06.450 align:middle line:84% And here they specify the cooling 01:02:06.450 --> 01:02:15.450 align:middle line:84% in the number of growing days reduced-- 01:02:15.450 --> 01:02:17.670 align:middle line:84% basically frost-free days where you can actually 01:02:17.670 --> 01:02:19.470 align:middle line:90% have a crop growing. 01:02:19.470 --> 01:02:25.390 align:middle line:84% And they estimate that it can be as low as, say, minus 40. 01:02:25.390 --> 01:02:29.010 align:middle line:84% This is blue here, minus 40, minus 50 days of growing. 01:02:29.010 --> 01:02:32.670 align:middle line:84% So you might lose a month of growing, especially 01:02:32.670 --> 01:02:36.350 align:middle line:84% if this happens after your crop has already sprouted, 01:02:36.350 --> 01:02:38.830 align:middle line:90% like let's say in April. 01:02:38.830 --> 01:02:40.745 align:middle line:90% And then you get back to frost. 01:02:40.745 --> 01:02:42.370 align:middle line:84% You're going to lose all of your crops, 01:02:42.370 --> 01:02:44.590 align:middle line:84% so you're not going to have crops for a year. 01:02:44.590 --> 01:02:50.430 align:middle line:84% So this is the concern that even a small nuclear skirmish 01:02:50.430 --> 01:02:55.030 align:middle line:84% might result in essentially a lost year of food production, 01:02:55.030 --> 01:02:58.670 align:middle line:84% in which case you're going to have massive global starvation 01:02:58.670 --> 01:03:01.990 align:middle line:90% and billions of people dying. 01:03:01.990 --> 01:03:06.470 align:middle line:84% Now, all this depends on a lot of parameters that are very 01:03:06.470 --> 01:03:08.590 align:middle line:90% unpredictable and sensitive. 01:03:08.590 --> 01:03:10.870 align:middle line:84% Most important one is probably the particle size 01:03:10.870 --> 01:03:14.670 align:middle line:84% of the soot particles, which determine their cross-sections 01:03:14.670 --> 01:03:16.450 align:middle line:90% with the photons. 01:03:16.450 --> 01:03:20.010 align:middle line:84% And none of that has been sufficiently studied. 01:03:20.010 --> 01:03:23.650 align:middle line:84% So I think there is some serious question 01:03:23.650 --> 01:03:29.350 align:middle line:84% as to whether this is actually a true result. But if it is, 01:03:29.350 --> 01:03:30.490 align:middle line:90% it's worrisome. 01:03:30.490 --> 01:03:33.130 align:middle line:84% There could be something very serious. 01:03:33.130 --> 01:03:37.090 align:middle line:90% So starving the planet-- 01:03:37.090 --> 01:03:40.770 align:middle line:90% I estimate the cost at infinity. 01:03:40.770 --> 01:03:43.930 align:middle line:90% And so there's your cost model. 01:03:43.930 --> 01:03:47.730 align:middle line:90% So here's our calculation. 01:03:47.730 --> 01:03:52.410 align:middle line:84% 0.13 programs per year times the cost per program. 01:03:52.410 --> 01:03:54.930 align:middle line:84% If you choose this pathway, if you choose this pathway, 01:03:54.930 --> 01:03:59.570 align:middle line:84% $47 billion per year or $220 billion per year 01:03:59.570 --> 01:04:02.230 align:middle line:84% is kind of a way to do the calculation. 01:04:02.230 --> 01:04:02.890 align:middle line:90% Yeah. 01:04:02.890 --> 01:04:04.973 align:middle line:84% AUDIENCE: What about the cost of bringing somebody 01:04:04.973 --> 01:04:06.647 align:middle line:90% under the US nuclear umbrella? 01:04:06.647 --> 01:04:08.730 align:middle line:84% SCOTT KEMP: Well, first, you have to convince them 01:04:08.730 --> 01:04:10.150 align:middle line:90% that it's a real thing. 01:04:10.150 --> 01:04:13.010 align:middle line:90% 01:04:13.010 --> 01:04:18.550 align:middle line:84% I have typically liked the idea of going to your adversaries 01:04:18.550 --> 01:04:22.410 align:middle line:84% and saying, OK, Iran, you shouldn't be doing this. 01:04:22.410 --> 01:04:23.970 align:middle line:84% We're going to pay you $2 trillion, 01:04:23.970 --> 01:04:25.505 align:middle line:84% which is like 30% of your economy, 01:04:25.505 --> 01:04:27.630 align:middle line:84% and we're going to give you a whole bunch of stuff. 01:04:27.630 --> 01:04:29.670 align:middle line:84% And you're just going to stop this. 01:04:29.670 --> 01:04:32.590 align:middle line:84% And that would have been cheaper. 01:04:32.590 --> 01:04:34.370 align:middle line:84% But we can't do that politically. 01:04:34.370 --> 01:04:36.310 align:middle line:90% We're somehow unable to do this. 01:04:36.310 --> 01:04:39.150 align:middle line:84% AUDIENCE: I think that Pakistan is actually 01:04:39.150 --> 01:04:41.770 align:middle line:90% a US ally in the area. 01:04:41.770 --> 01:04:44.290 align:middle line:90% 01:04:44.290 --> 01:04:46.150 align:middle line:90% SCOTT KEMP: Not strongly. 01:04:46.150 --> 01:04:48.455 align:middle line:84% It's really more of a Chinese alloy. 01:04:48.455 --> 01:04:49.830 align:middle line:84% AUDIENCE: Yeah, but historically, 01:04:49.830 --> 01:04:50.850 align:middle line:90% it has been an ally. 01:04:50.850 --> 01:04:51.410 align:middle line:90% I mean-- 01:04:51.410 --> 01:04:52.770 align:middle line:90% SCOTT KEMP: Pretty minimally. 01:04:52.770 --> 01:04:54.830 align:middle line:84% AUDIENCE: This is what I've read, actually. 01:04:54.830 --> 01:04:57.750 align:middle line:84% And then the US, I mean, I've also 01:04:57.750 --> 01:05:01.350 align:middle line:84% read that nuclear weapons could be 01:05:01.350 --> 01:05:05.310 align:middle line:84% used as a method of nuclear deterrence in some cases. 01:05:05.310 --> 01:05:07.003 align:middle line:84% So if both countries have weapons-- 01:05:07.003 --> 01:05:08.170 align:middle line:90% SCOTT KEMP: That's the idea. 01:05:08.170 --> 01:05:09.670 align:middle line:84% AUDIENCE: --then both countries know 01:05:09.670 --> 01:05:11.750 align:middle line:84% that the mutual self-destruction. 01:05:11.750 --> 01:05:13.990 align:middle line:90% So there was a-- 01:05:13.990 --> 01:05:17.050 align:middle line:84% SCOTT KEMP: Have I given my spiel on deterrence yet? 01:05:17.050 --> 01:05:20.850 align:middle line:84% Deterrence is an idea that works in theory 01:05:20.850 --> 01:05:22.990 align:middle line:84% but doesn't work perfectly in practice. 01:05:22.990 --> 01:05:26.970 align:middle line:84% There are all kinds of ways that it can break down. 01:05:26.970 --> 01:05:27.470 align:middle line:90% Yeah. 01:05:27.470 --> 01:05:30.552 align:middle line:84% AUDIENCE: I mean, choices are not always rational. 01:05:30.552 --> 01:05:32.010 align:middle line:84% SCOTT KEMP: It's more it's not only 01:05:32.010 --> 01:05:33.710 align:middle line:90% that it's not always rational. 01:05:33.710 --> 01:05:36.030 align:middle line:84% It's that for deterrence to work, 01:05:36.030 --> 01:05:40.050 align:middle line:84% you have to estimate the expectation value 01:05:40.050 --> 01:05:42.730 align:middle line:90% of your adversary's next step. 01:05:42.730 --> 01:05:45.490 align:middle line:84% You say they're going to do this escalatory path 01:05:45.490 --> 01:05:48.410 align:middle line:84% with certain probability and an expectation value that 01:05:48.410 --> 01:05:49.650 align:middle line:90% cost me x. 01:05:49.650 --> 01:05:53.010 align:middle line:84% And to me, if x is bigger than what I want y, 01:05:53.010 --> 01:05:54.490 align:middle line:90% then I'm going to stop. 01:05:54.490 --> 01:05:58.610 align:middle line:84% And the problem is you can't predict your adversary's 01:05:58.610 --> 01:05:59.350 align:middle line:90% next step. 01:05:59.350 --> 01:06:01.392 align:middle line:84% AUDIENCE: And there's no historical data as well. 01:06:01.392 --> 01:06:06.330 align:middle line:84% The closest would be the Cuban Missile Crisis in 1960. 01:06:06.330 --> 01:06:08.950 align:middle line:84% There's no historical data to calculate those probabilities. 01:06:08.950 --> 01:06:11.400 align:middle line:90% SCOTT KEMP: Not really. 01:06:11.400 --> 01:06:13.800 align:middle line:84% Did I tell you about the submarine people? 01:06:13.800 --> 01:06:17.000 align:middle line:84% AUDIENCE: That's what I was about to ask about, our mistakes 01:06:17.000 --> 01:06:17.500 align:middle line:90% as well. 01:06:17.500 --> 01:06:19.930 align:middle line:90% [INAUDIBLE] false alarms. 01:06:19.930 --> 01:06:22.160 align:middle line:84% SCOTT KEMP: Actions on false alarms. 01:06:22.160 --> 01:06:23.920 align:middle line:84% Have I not told the submarine story? 01:06:23.920 --> 01:06:25.680 align:middle line:90% All right. 01:06:25.680 --> 01:06:28.160 align:middle line:84% Everyone know what the Cuban Missile Crisis is? 01:06:28.160 --> 01:06:28.880 align:middle line:90% All right. 01:06:28.880 --> 01:06:29.580 align:middle line:90% So, yes. 01:06:29.580 --> 01:06:32.840 align:middle line:84% So the United States was putting missiles in Europe. 01:06:32.840 --> 01:06:33.983 align:middle line:90% Russians didn't like it. 01:06:33.983 --> 01:06:36.400 align:middle line:84% So they were like, OK, we're going to put missiles outside 01:06:36.400 --> 01:06:39.160 align:middle line:90% of the United States in Cuba. 01:06:39.160 --> 01:06:42.720 align:middle line:84% And so they had these boats bringing the missiles to Cuba. 01:06:42.720 --> 01:06:47.060 align:middle line:84% And the US had satellite imagery of these missiles being sites, 01:06:47.060 --> 01:06:49.560 align:middle line:84% missile launchers being built. And they freaked out, 01:06:49.560 --> 01:06:52.880 align:middle line:84% and they sent the US Navy to stop the missiles 01:06:52.880 --> 01:06:55.880 align:middle line:90% from being delivered to Cuba. 01:06:55.880 --> 01:06:58.760 align:middle line:84% So this is a blockade, which technically an act of war. 01:06:58.760 --> 01:07:04.660 align:middle line:84% And they were told, don't do anything to provoke these guys. 01:07:04.660 --> 01:07:07.120 align:middle line:84% Just make sure the missiles don't land in Cuba. 01:07:07.120 --> 01:07:08.920 align:middle line:90% Navy ships go there. 01:07:08.920 --> 01:07:13.500 align:middle line:84% Well, the Soviet ships that are carrying the missiles 01:07:13.500 --> 01:07:17.220 align:middle line:84% have an escort in the form of a submarine, 01:07:17.220 --> 01:07:19.580 align:middle line:84% and the submarines are prowling around 01:07:19.580 --> 01:07:22.260 align:middle line:84% as the surface ships have been stopped by the US Navy surface 01:07:22.260 --> 01:07:23.140 align:middle line:90% ships. 01:07:23.140 --> 01:07:25.780 align:middle line:84% And the US Navy knows that these submarines are down there 01:07:25.780 --> 01:07:28.740 align:middle line:84% and doesn't like it, because they can shoot torpedoes 01:07:28.740 --> 01:07:29.720 align:middle line:90% at their boats. 01:07:29.720 --> 01:07:31.780 align:middle line:90% So they want them to surface. 01:07:31.780 --> 01:07:36.460 align:middle line:84% So some guy has the bright idea of dropping a depth charge off 01:07:36.460 --> 01:07:41.460 align:middle line:84% of the US surface ship to try to get the Soviet submarines 01:07:41.460 --> 01:07:43.340 align:middle line:90% to surface. 01:07:43.340 --> 01:07:47.740 align:middle line:84% So what happens is people on the Soviet submarines 01:07:47.740 --> 01:07:52.460 align:middle line:84% interpret incorrectly these depth charges as above water 01:07:52.460 --> 01:07:54.780 align:middle line:90% nuclear detonations. 01:07:54.780 --> 01:07:57.580 align:middle line:84% They're like nuclear war has started. 01:07:57.580 --> 01:08:01.100 align:middle line:84% And the reason you have nuclear submarines 01:08:01.100 --> 01:08:02.780 align:middle line:84% is that they're survivable platforms 01:08:02.780 --> 01:08:06.380 align:middle line:90% in the event of a nuclear war. 01:08:06.380 --> 01:08:09.760 align:middle line:84% And you predelegate launch authority 01:08:09.760 --> 01:08:12.240 align:middle line:90% to the people on the submarine. 01:08:12.240 --> 01:08:15.600 align:middle line:84% And so the rule is that the captain 01:08:15.600 --> 01:08:19.920 align:middle line:84% and the executive officer have to agree, 01:08:19.920 --> 01:08:21.760 align:middle line:84% and they both have to put their keys in 01:08:21.760 --> 01:08:24.240 align:middle line:90% and turn and open the safes. 01:08:24.240 --> 01:08:26.359 align:middle line:84% And each one has a combination, and so on. 01:08:26.359 --> 01:08:29.620 align:middle line:84% So both the captain of the submarine and the first officer, 01:08:29.620 --> 01:08:33.040 align:middle line:84% the executive officer, agreed to launch nuclear weapons 01:08:33.040 --> 01:08:34.180 align:middle line:90% at the United States. 01:08:34.180 --> 01:08:38.840 align:middle line:90% 01:08:38.840 --> 01:08:43.080 align:middle line:84% And this guy named Admiral Arkhipov, who 01:08:43.080 --> 01:08:49.000 align:middle line:84% was not the captain of the boat, but was like a fleet commander, 01:08:49.000 --> 01:08:52.960 align:middle line:84% happened to be on a joyride on this particular submarine. 01:08:52.960 --> 01:08:54.987 align:middle line:84% And although he was not in charge of the boat 01:08:54.987 --> 01:08:56.779 align:middle line:84% because he was not the captain of the boat, 01:08:56.779 --> 01:09:00.399 align:middle line:84% but he technically kind of outranked the captain. 01:09:00.399 --> 01:09:02.899 align:middle line:84% And he was like, no, you shouldn't be launching this. 01:09:02.899 --> 01:09:05.112 align:middle line:84% And the captain's like, I am absolutely launching. 01:09:05.112 --> 01:09:05.779 align:middle line:90% This is my boat. 01:09:05.779 --> 01:09:06.979 align:middle line:90% Get out of the way. 01:09:06.979 --> 01:09:09.125 align:middle line:84% And apparently, it got into a very heated argument 01:09:09.125 --> 01:09:11.500 align:middle line:84% where they were shouting at each other about whether they 01:09:11.500 --> 01:09:13.020 align:middle line:90% should or should not launch. 01:09:13.020 --> 01:09:16.779 align:middle line:84% And eventually, the guy managed to convince the captain 01:09:16.779 --> 01:09:19.740 align:middle line:84% to surface and just check to see if nuclear war had actually 01:09:19.740 --> 01:09:21.180 align:middle line:90% started. 01:09:21.180 --> 01:09:23.840 align:middle line:84% And lo and behold, nuclear war had not started. 01:09:23.840 --> 01:09:26.220 align:middle line:84% But if this guy had not been on the joyride, 01:09:26.220 --> 01:09:31.540 align:middle line:84% or if he had not been so hesitant to retaliate, 01:09:31.540 --> 01:09:34.540 align:middle line:84% the Soviets would have launched, and probably the US 01:09:34.540 --> 01:09:35.660 align:middle line:90% would have retaliated. 01:09:35.660 --> 01:09:37.500 align:middle line:84% And we would have had nuclear war. 01:09:37.500 --> 01:09:39.580 align:middle line:84% And this was an era when we had tens of thousands 01:09:39.580 --> 01:09:41.939 align:middle line:84% of nuclear weapons pointed at each other. 01:09:41.939 --> 01:09:44.300 align:middle line:84% It's not hard to contrive scenarios 01:09:44.300 --> 01:09:47.340 align:middle line:90% where we would not be here. 01:09:47.340 --> 01:09:49.420 align:middle line:84% So it came down to a yelling match. 01:09:49.420 --> 01:09:53.319 align:middle line:84% So this is why nuclear deterrence is not the answer. 01:09:53.319 --> 01:09:56.180 align:middle line:90% 01:09:56.180 --> 01:09:58.160 align:middle line:90% So let me just finish this. 01:09:58.160 --> 01:10:01.620 align:middle line:90% 01:10:01.620 --> 01:10:07.640 align:middle line:84% How much does this cost per year to nuclear power? 01:10:07.640 --> 01:10:09.727 align:middle line:84% So what I did is I take this cost per year, 01:10:09.727 --> 01:10:11.560 align:middle line:84% and I divide it against either US production 01:10:11.560 --> 01:10:13.520 align:middle line:84% or global production of nuclear energy 01:10:13.520 --> 01:10:16.980 align:middle line:84% to get a cost per kilowatt-hour or cost per megawatt-hour. 01:10:16.980 --> 01:10:20.840 align:middle line:90% 01:10:20.840 --> 01:10:23.720 align:middle line:84% It goes as low as $21 a megawatt-hour. 01:10:23.720 --> 01:10:26.880 align:middle line:84% But when I did this calculation, I actually only assumed the cost 01:10:26.880 --> 01:10:28.520 align:middle line:90% to the United States. 01:10:28.520 --> 01:10:31.600 align:middle line:84% So really, you should only consider the cost 01:10:31.600 --> 01:10:34.860 align:middle line:90% against the US nuclear program. 01:10:34.860 --> 01:10:36.800 align:middle line:84% I didn't include the cost of sanctions 01:10:36.800 --> 01:10:39.200 align:middle line:90% to Europe, for example. 01:10:39.200 --> 01:10:42.940 align:middle line:84% So this is kind of not the correct math over here, 01:10:42.940 --> 01:10:45.600 align:middle line:84% but I put it there so you can see it. 01:10:45.600 --> 01:10:51.840 align:middle line:90% So $59 to $275 to infinity-- 01:10:51.840 --> 01:10:54.960 align:middle line:90% so this is not a small number. 01:10:54.960 --> 01:11:01.200 align:middle line:84% This is basically doubling the cost of nuclear power. 01:11:01.200 --> 01:11:04.700 align:middle line:84% $59 a megawatt-hour is the cost of wind and solar 01:11:04.700 --> 01:11:08.340 align:middle line:84% without overbuild or storage, a lot of storage. 01:11:08.340 --> 01:11:12.300 align:middle line:84% And preventative war-- this is the cost of the most expensive 01:11:12.300 --> 01:11:13.620 align:middle line:90% nuclear today. 01:11:13.620 --> 01:11:15.000 align:middle line:90% So it's basically double. 01:11:15.000 --> 01:11:17.780 align:middle line:90% 01:11:17.780 --> 01:11:20.660 align:middle line:84% So if you believe this calculation, 01:11:20.660 --> 01:11:23.585 align:middle line:84% it turns out proliferation is a big externality. 01:11:23.585 --> 01:11:25.460 align:middle line:84% AUDIENCE: Isn't this kind of a weird argument 01:11:25.460 --> 01:11:27.500 align:middle line:84% to be making for the US in particular, at least 01:11:27.500 --> 01:11:30.500 align:middle line:90% for US reactors? 01:11:30.500 --> 01:11:33.660 align:middle line:84% A US reactor does not necessarily mean that-- 01:11:33.660 --> 01:11:36.020 align:middle line:84% I don't know-- Brazil is going to get 01:11:36.020 --> 01:11:38.860 align:middle line:90% a nuclear weapon from that. 01:11:38.860 --> 01:11:40.600 align:middle line:90% We already have nuclear weapons. 01:11:40.600 --> 01:11:43.340 align:middle line:84% It's already a part of our nuclear fleet. 01:11:43.340 --> 01:11:46.620 align:middle line:84% SCOTT KEMP: Yeah, the argument is more like 01:11:46.620 --> 01:11:51.040 align:middle line:84% if we decide that nuclear power is something we want to have 01:11:51.040 --> 01:11:52.153 align:middle line:90% and we want to-- 01:11:52.153 --> 01:11:53.320 align:middle line:90% AUDIENCE: Or just in the US. 01:11:53.320 --> 01:11:57.780 align:middle line:84% SCOTT KEMP: Well, I'll get to that in a second. 01:11:57.780 --> 01:11:59.400 align:middle line:84% You're going to promote nuclear power. 01:11:59.400 --> 01:12:00.640 align:middle line:84% You have a nuclear power industry. 01:12:00.640 --> 01:12:02.110 align:middle line:84% You're going to support your industry. 01:12:02.110 --> 01:12:03.750 align:middle line:84% You're going to have reactors in the United States, 01:12:03.750 --> 01:12:05.167 align:middle line:84% and that industry is going to want 01:12:05.167 --> 01:12:07.570 align:middle line:84% to export its reactors to other countries. 01:12:07.570 --> 01:12:11.410 align:middle line:84% And on occasion, there will be proliferation. 01:12:11.410 --> 01:12:17.690 align:middle line:84% And so it's a surcharge that should apply to the US industry 01:12:17.690 --> 01:12:21.410 align:middle line:84% because it does promote and push for exports, as it's doing today 01:12:21.410 --> 01:12:23.850 align:middle line:84% with Saudi Arabia, which is almost certainly going 01:12:23.850 --> 01:12:27.090 align:middle line:84% to convert its civilian program into a nuclear weapons 01:12:27.090 --> 01:12:28.890 align:middle line:90% program at some point. 01:12:28.890 --> 01:12:30.490 align:middle line:84% AUDIENCE: Well, why wouldn't you just 01:12:30.490 --> 01:12:33.570 align:middle line:84% like, add that cost to the export cost 01:12:33.570 --> 01:12:36.730 align:middle line:84% as opposed to limiting internal development? 01:12:36.730 --> 01:12:38.990 align:middle line:84% SCOTT KEMP: So there's no export cost to the US. 01:12:38.990 --> 01:12:41.217 align:middle line:90% There's export profits. 01:12:41.217 --> 01:12:43.050 align:middle line:84% AUDIENCE: But there's no reason you couldn't 01:12:43.050 --> 01:12:44.342 align:middle line:90% put an export cost [INAUDIBLE]. 01:12:44.342 --> 01:12:46.870 align:middle line:84% SCOTT KEMP: Then we wouldn't export. 01:12:46.870 --> 01:12:47.823 align:middle line:90% But we do. 01:12:47.823 --> 01:12:49.490 align:middle line:84% Do you see the problem is like someone-- 01:12:49.490 --> 01:12:51.890 align:middle line:84% you're paying this as an externality. 01:12:51.890 --> 01:12:54.830 align:middle line:84% So you're not paying it as an internalized cost. 01:12:54.830 --> 01:12:58.130 align:middle line:84% So you have to attribute the cost somewhere. 01:12:58.130 --> 01:13:00.910 align:middle line:84% And it's all the benefits you get 01:13:00.910 --> 01:13:04.510 align:middle line:84% minus the cost of having that industry. 01:13:04.510 --> 01:13:07.430 align:middle line:84% If you choose to say, well, it's just going to be on the export, 01:13:07.430 --> 01:13:11.010 align:middle line:84% then that would be fine if it actually stopped the export. 01:13:11.010 --> 01:13:13.070 align:middle line:90% But it doesn't. 01:13:13.070 --> 01:13:15.470 align:middle line:84% AUDIENCE: It feels quite dubious either way. 01:13:15.470 --> 01:13:16.610 align:middle line:90% I don't know. 01:13:16.610 --> 01:13:20.190 align:middle line:84% SCOTT KEMP: In any way, it's an enormous number. 01:13:20.190 --> 01:13:21.750 align:middle line:90% I think that's the point. 01:13:21.750 --> 01:13:23.570 align:middle line:90% There was more questions. 01:13:23.570 --> 01:13:26.790 align:middle line:84% AUDIENCE: That $360 billion from sanctions, 01:13:26.790 --> 01:13:28.330 align:middle line:90% that was based on Iran? 01:13:28.330 --> 01:13:33.230 align:middle line:84% Wouldn't that depend on the specific economic trade-- 01:13:33.230 --> 01:13:34.630 align:middle line:90% SCOTT KEMP: Absolutely. 01:13:34.630 --> 01:13:35.710 align:middle line:90% Yeah, absolutely. 01:13:35.710 --> 01:13:37.357 align:middle line:84% AUDIENCE: --with a country in which we 01:13:37.357 --> 01:13:39.690 align:middle line:84% trade more with than that value would be higher or less, 01:13:39.690 --> 01:13:40.550 align:middle line:90% [INAUDIBLE]. 01:13:40.550 --> 01:13:42.350 align:middle line:84% With North Korea, it might be tiny 01:13:42.350 --> 01:13:44.133 align:middle line:84% because we don't do much trade with them. 01:13:44.133 --> 01:13:45.050 align:middle line:90% SCOTT KEMP: Precisely. 01:13:45.050 --> 01:13:46.870 align:middle line:84% We don't do much trade with Iran either. 01:13:46.870 --> 01:13:48.910 align:middle line:90% This is the lost trade. 01:13:48.910 --> 01:13:51.270 align:middle line:90% It's estimate of the lost trade. 01:13:51.270 --> 01:13:54.250 align:middle line:84% We would probably have done a lot of trade with North Korea, 01:13:54.250 --> 01:13:56.750 align:middle line:84% though there are other things other than its nuclear program 01:13:56.750 --> 01:13:58.810 align:middle line:90% that prevents that. 01:13:58.810 --> 01:14:01.970 align:middle line:84% But they have some of the strongest deposits 01:14:01.970 --> 01:14:04.450 align:middle line:90% of rare earths, for example. 01:14:04.450 --> 01:14:07.810 align:middle line:84% We would love to have access to North Korea. 01:14:07.810 --> 01:14:11.610 align:middle line:84% Yeah, it'd be huge, huge, hugely beneficial. 01:14:11.610 --> 01:14:14.010 align:middle line:84% Yeah, so it does absolutely factor in. 01:14:14.010 --> 01:14:17.170 align:middle line:84% You're going to try the sanctions story on South Korea? 01:14:17.170 --> 01:14:18.570 align:middle line:90% Probably not. 01:14:18.570 --> 01:14:20.177 align:middle line:90% Trade is just way too intense. 01:14:20.177 --> 01:14:21.510 align:middle line:90% This would be way too expensive. 01:14:21.510 --> 01:14:27.770 align:middle line:84% So you either capitulate or you do something else. 01:14:27.770 --> 01:14:32.850 align:middle line:84% So it is very case-specific, but I wanted to use real numbers. 01:14:32.850 --> 01:14:35.490 align:middle line:90% Yeah. 01:14:35.490 --> 01:14:37.810 align:middle line:84% All this is trying to show-- these are not real costs. 01:14:37.810 --> 01:14:40.770 align:middle line:84% These are just trying to illustrate scale 01:14:40.770 --> 01:14:42.650 align:middle line:90% that this is not small. 01:14:42.650 --> 01:14:44.930 align:middle line:84% That's all I'm trying to show you. 01:14:44.930 --> 01:14:45.430 align:middle line:90% All right. 01:14:45.430 --> 01:14:47.970 align:middle line:84% So to get to this point, can we be clever and give it only 01:14:47.970 --> 01:14:49.770 align:middle line:90% to the good guys? 01:14:49.770 --> 01:14:51.690 align:middle line:90% This always gets raised. 01:14:51.690 --> 01:14:55.930 align:middle line:84% And it's nice if this were true, but history 01:14:55.930 --> 01:14:57.710 align:middle line:90% has some warnings for us. 01:14:57.710 --> 01:15:00.490 align:middle line:90% So who has seen this? 01:15:00.490 --> 01:15:03.050 align:middle line:90% 01:15:03.050 --> 01:15:04.610 align:middle line:90% Yeah, this is 1970. 01:15:04.610 --> 01:15:09.510 align:middle line:84% This is Iran was our ally, as evidenced by this propaganda 01:15:09.510 --> 01:15:11.590 align:middle line:84% advertisement for the Pilgrim Nuclear Plant 01:15:11.590 --> 01:15:12.890 align:middle line:90% here in Massachusetts. 01:15:12.890 --> 01:15:14.450 align:middle line:84% They were trying to build up support, 01:15:14.450 --> 01:15:17.390 align:middle line:84% and they're saying, look, the Shah wants nuclear power. 01:15:17.390 --> 01:15:20.790 align:middle line:84% Shouldn't you want nuclear power, too? 01:15:20.790 --> 01:15:23.310 align:middle line:84% And indeed, the Shah was building a nuclear plant 01:15:23.310 --> 01:15:26.270 align:middle line:84% using US support and simultaneously 01:15:26.270 --> 01:15:29.910 align:middle line:84% harbored not-so-secret nuclear weapons ambitions almost 01:15:29.910 --> 01:15:34.070 align:middle line:84% identical to what is happening today with Saudi Arabia. 01:15:34.070 --> 01:15:36.550 align:middle line:84% And we were OK with that because the Shah 01:15:36.550 --> 01:15:41.310 align:middle line:84% gave US oil, just like Saudi Arabia gives us oil today. 01:15:41.310 --> 01:15:43.290 align:middle line:84% And there was even a private company, 01:15:43.290 --> 01:15:45.190 align:middle line:84% and I know the head of the company, 01:15:45.190 --> 01:15:51.790 align:middle line:84% who was exporting with official US government license, 01:15:51.790 --> 01:15:55.490 align:middle line:84% uranium enrichment technology to Iran based on laser isotope 01:15:55.490 --> 01:15:56.570 align:middle line:90% separation. 01:15:56.570 --> 01:15:58.290 align:middle line:84% And they were in the process of building 01:15:58.290 --> 01:16:01.050 align:middle line:84% that facility for the Iranians when 01:16:01.050 --> 01:16:02.890 align:middle line:90% the Iranian Revolution came. 01:16:02.890 --> 01:16:07.590 align:middle line:84% And they captured the hostages at the US embassy, 01:16:07.590 --> 01:16:11.090 align:middle line:84% which led to the change in relationship with Iran. 01:16:11.090 --> 01:16:18.370 align:middle line:84% After Iran's revolution, we found a new friend 01:16:18.370 --> 01:16:19.930 align:middle line:90% in the region. 01:16:19.930 --> 01:16:22.930 align:middle line:90% Do you know who it was? 01:16:22.930 --> 01:16:24.010 align:middle line:90% It was Iraq. 01:16:24.010 --> 01:16:27.710 align:middle line:84% So here is former Secretary of Defense Don Rumsfeld, 01:16:27.710 --> 01:16:30.210 align:middle line:84% at that time, in his capacity as special envoy to the Middle 01:16:30.210 --> 01:16:35.050 align:middle line:84% East, shaking hands with Saddam Hussein in 1983. 01:16:35.050 --> 01:16:36.970 align:middle line:84% Saddam Hussein at this point had used 01:16:36.970 --> 01:16:42.290 align:middle line:84% mustard gas, illegal under the Geneva Conventions, 01:16:42.290 --> 01:16:45.770 align:middle line:84% and went on with US technical support 01:16:45.770 --> 01:16:51.290 align:middle line:84% to use nerve agent in 1984 against the Iranians, 01:16:51.290 --> 01:16:54.750 align:middle line:84% both of them banned under the Geneva Protocol. 01:16:54.750 --> 01:16:58.350 align:middle line:84% Eventually, we even exported the chemical precursors 01:16:58.350 --> 01:17:02.430 align:middle line:84% to help Saddam Hussein make these chemical weapons to attack 01:17:02.430 --> 01:17:09.470 align:middle line:84% the Iranians and because the Iraqis were our friends. 01:17:09.470 --> 01:17:12.370 align:middle line:84% And then two years later, we had enough of him, 01:17:12.370 --> 01:17:16.830 align:middle line:84% and we were at war with Iraq over their invasion of Kuwait, 01:17:16.830 --> 01:17:18.052 align:middle line:90% who were also our friends. 01:17:18.052 --> 01:17:19.510 align:middle line:84% And then we learned that the Iraqis 01:17:19.510 --> 01:17:23.670 align:middle line:84% had a big hidden nuclear weapons program, including centrifuges. 01:17:23.670 --> 01:17:26.063 align:middle line:84% And would you believe these guys? 01:17:26.063 --> 01:17:28.230 align:middle line:84% We're so friendly with them, and they go off and try 01:17:28.230 --> 01:17:29.570 align:middle line:90% to make nuclear weapons. 01:17:29.570 --> 01:17:31.470 align:middle line:90% How dare they? 01:17:31.470 --> 01:17:33.990 align:middle line:84% So, look, I don't want to belabor the point for more 01:17:33.990 --> 01:17:37.590 align:middle line:84% than it's worth, but history is full of examples 01:17:37.590 --> 01:17:40.350 align:middle line:84% where it is extremely difficult to predict 01:17:40.350 --> 01:17:45.150 align:middle line:84% the course of foreign affairs, where you're friends one day, 01:17:45.150 --> 01:17:49.070 align:middle line:84% and 10 years later, everything has changed. 01:17:49.070 --> 01:17:55.140 align:middle line:84% So Saudi Arabia is a country that bred 15 of the 19 hijackers 01:17:55.140 --> 01:18:00.020 align:middle line:90% that flew planes during 9/11. 01:18:00.020 --> 01:18:06.820 align:middle line:84% They killed Jamal Khashoggi, the Washington Post journalist, 01:18:06.820 --> 01:18:08.180 align:middle line:90% very violently. 01:18:08.180 --> 01:18:10.980 align:middle line:84% And we just close our eyes to it. 01:18:10.980 --> 01:18:12.480 align:middle line:84% And all the warning signs are there. 01:18:12.480 --> 01:18:14.740 align:middle line:84% And we say, because we want the $10 billion 01:18:14.740 --> 01:18:17.260 align:middle line:84% contract for that nuclear power plant. 01:18:17.260 --> 01:18:20.585 align:middle line:84% And then we're surprised when 10 years later, they're 01:18:20.585 --> 01:18:21.960 align:middle line:84% using it to make nuclear weapons. 01:18:21.960 --> 01:18:24.940 align:middle line:84% We really shouldn't be that surprised. 01:18:24.940 --> 01:18:30.420 align:middle line:84% Here's a more dispassionate way to make the same argument. 01:18:30.420 --> 01:18:33.620 align:middle line:84% We are accustomed to the idea of living 01:18:33.620 --> 01:18:38.500 align:middle line:84% in a relatively stable country, but the whole world 01:18:38.500 --> 01:18:39.720 align:middle line:90% isn't like that. 01:18:39.720 --> 01:18:43.660 align:middle line:84% So here is some of the constitutional rewrites 01:18:43.660 --> 01:18:45.780 align:middle line:90% since the year 2000. 01:18:45.780 --> 01:18:49.880 align:middle line:84% And it just goes to show that countries 01:18:49.880 --> 01:18:53.480 align:middle line:84% are having major political upheaval all the time. 01:18:53.480 --> 01:18:55.840 align:middle line:84% The average age of a constitution in the world 01:18:55.840 --> 01:18:59.640 align:middle line:90% is about 43 years. 01:18:59.640 --> 01:19:01.680 align:middle line:84% The average lifetime of a nuclear power plant 01:19:01.680 --> 01:19:04.072 align:middle line:90% is twice that. 01:19:04.072 --> 01:19:06.280 align:middle line:84% You do not know who you are giving your nuclear power 01:19:06.280 --> 01:19:09.120 align:middle line:84% plant to when you give your nuclear power plant to someone. 01:19:09.120 --> 01:19:12.360 align:middle line:84% You think you know, but you don't. 01:19:12.360 --> 01:19:14.720 align:middle line:90% Things change. 01:19:14.720 --> 01:19:17.600 align:middle line:84% So I say, we need to be very careful about this idea 01:19:17.600 --> 01:19:21.320 align:middle line:84% that we will just give it to the good guys. 01:19:21.320 --> 01:19:24.520 align:middle line:90% And that's all I have to say. 01:19:24.520 --> 01:19:28.640 align:middle line:84% So we will put everything together in the next session 01:19:28.640 --> 01:19:31.590 align:middle line:90% and see where we stand. 01:19:31.590 --> 01:19:49.000 align:middle line:90%