Why Private Billions Are Flowing Into Fusion

Aug 11, 2023

Why Private Billions Are Flowing Into Fusion

As the world searches for a way to avoid climate catastrophe, dozens of fusion startups are putting billions of dollars into the risky quest for unlimited clean power.

#PowerMoves #CleanEnergy #BloombergQuicktake
--------
Like this video? Subscribe: http://www.youtube.com/Bloomberg?sub_…
Become a Quicktake Member for exclusive perks: http://www.youtube.com/bloomberg/join
Subscribe to Quicktake Explained: https://bit.ly/3iERrup

QuickTake Originals is Bloomberg’s official premium video channel. We bring you insights and analysis from business, science, and technology experts who are shaping our future. We’re home to Hello World, Giant Leap, Storylines, and the series powering CityLab, Bloomberg Businessweek, Bloomberg Green, and much more.

Subscribe for business news, but not as you’ve known it: exclusive interviews, fascinating profiles, data-driven analysis, and the latest in tech innovation from around the world.

Visit our partner channel QuickTake News for breaking global news and insight in an instant.

Content

1.876 -> Fusion power has a certain reputation,

5.38 -> something in the realm of a holy grail,

8.174 -> but not in a good way.

10.051 -> There is a famous joke about nuclear fusion

12.554 -> about how it never becomes a reality.

14.639 -> You can Google it if you like.

17.017 -> But really the moment we are in with nuclear fusion today

20.353 -> is probably more exciting than it's ever been.

23.231 -> There's more activity in the fusion world

25.275 -> than ever, and not just in government research labs.

28.695 -> There's also an emerging private fusion industry

31.448 -> that's attracted billions of dollars in capital

33.7 -> in recent years.

35.201 -> Governments and private investors alike

37.12 -> realize that we've got to find a solution

39.456 -> that's going to allow us to get to net zero targets.

42.834 -> This is one of the hardest

43.793 -> but most rewarding problems that humanity could work on.

47.505 -> Ultimately, we all want the same thing.

48.965 -> We want someone to put electricity on the grid

51.176 -> from a fusion power station as quickly as possible.

53.845 -> Frankly, the scale of the challenge,

55.972 -> 3,000 gigawatts of fossil fuel to replace.

60.351 -> There's not many things that can do that.

62.02 -> In fact, there may only be fusion that can really do that.

65.231 -> And while many in the scientific community

67.484 -> predict fusion power will take decades,

70.403 -> some in the private fusion space

72.53 -> believe we'll get there in just a few years,

75.241 -> as soon as the 2030s.

77.702 -> There is a lot of money

78.912 -> going into these companies.

81.039 -> And it's very interesting and exciting.

83.166 -> There are some I really love

86.086 -> and some I would rather laugh about.

88.797 -> The world is desperately searching

90.757 -> for a replacement for fossil fuels.

93.134 -> Now scientists and startups

95.72 -> are betting that a commercial fusion reactor

98.306 -> is finally in sight.

120.245 -> We are looking at visible light

122.205 -> coming from the plasma.

124.332 -> The plasma is this sort of 100,000 degrees temperature,

128.169 -> but what you actually see is the coldest part.

131.339 -> The hottest part you really don't see

133.341 -> because it's too hot to emit light

136.719 -> in the visible, isn't it?

139.848 -> Yes.

142.267 -> When I started, I was doing numerical modeling

145.061 -> and then I realized that running the machine

147.522 -> was a lot more interesting and a lot more fun for me.

153.278 -> My work is stressful.

154.696 -> My work can be very demanding in terms of time.

159.45 -> But this is still

161.286 -> one of the most exciting places in the world.

165.206 -> Here in Oxford, you'll find arguably

167.208 -> the most successful fusion experiment on Earth,

170.42 -> JET, the Joint European Torus,

173.381 -> torus being the technical term for donut,

176.092 -> which is how the reactor's shaped.

178.678 -> JET's been operating since the early '80s

181.514 -> and only by 1997 were we really ready to try proper fusion.

185.602 -> And we produced 16 megawatts of fusion power at the time

188.813 -> which is like a few wind turbines.

190.481 -> It's pretty significant.

191.566 -> And it shows that fusion is possible.

196.696 -> The 1997 experiment set records,

199.782 -> but the reactor was only able to run for less than a second.

203.828 -> The team spent the next two decades

206.206 -> cooking up a new approach.

208.166 -> And in 2021, they gave it another shot.

211.544 -> We always knew that we could do better.

214.631 -> The last two days before Christmas

216.341 -> were dedicated to these experiments,

218.051 -> just this window of parameters where we could get more.

221.596 -> And we did.

223.306 -> Three, two, one.

229.604 -> JET more than doubled its previous record,

232.273 -> producing more energy than any fusion experiment in history.

236.527 -> We couldn't hug.

238.613 -> We couldn't high-five, nothing,

240.281 -> because we have to be at two meters from each other.

243.826 -> But, you know, it was obvious that this was a record.

247.622 -> It was successful.

248.498 -> You could see that it was successful.

252.001 -> It's a real step

252.835 -> towards the ultimate promise of fusion,

255.672 -> a cheap, emissions-free power source

258.216 -> with virtually unlimited fuel.

261.344 -> But maybe don't break out the champagne just yet.

264.973 -> So in the 1997 experiments

267.016 -> we produced a lot of power, but very transiently.

269.394 -> So it ramped up and then we lost control.

271.604 -> Now we ramp up and sustain for five seconds.

275.858 -> So what exactly makes fusion

277.86 -> such a tough problem

279.279 -> that sustaining it for five seconds

281.698 -> constitutes a world record?

284.075 -> Nuclear fusion.

285.743 -> Once it's perfected,

287.328 -> fusion power will give us an unlimited supply of energy.

291.791 -> So nuclear fusion is really what is happening

294.836 -> inside the sun,

296.254 -> where lots and lots of hydrogen atoms

300.3 -> are moving at immense speeds.

304.262 -> And every so often some of them fuse together

307.348 -> to form helium.

309.892 -> Now the process at the atomic level

311.894 -> leads to a very small amount of loss in mass.

316.316 -> And that little amount of mass

318.735 -> actually generates a lot of energy.

321.696 -> And you do that millions and millions of times

323.948 -> and you get the sun.

326.659 -> Well look, we know fusion works.

327.744 -> It's happening right now in our sun.

329.579 -> But the reason it's happening in the sun

330.955 -> is because of the mass of the sun.

332.582 -> It's so massive,

333.958 -> has this huge gravitational force

336.419 -> which is pushing those isotopes of hydrogen

338.796 -> close enough to fuse.

340.506 -> We obviously cannot recreate the mass of the sun

343.843 -> here on Earth.

344.969 -> So instead, we have to give that fuel even more energy.

349.807 -> So we take a gas, we put a huge amount of energy into it,

354.062 -> and that turns it into the fourth state of matter, plasma.

356.647 -> If you consider water, for example, it's ice,

359.275 -> then you warm it up

360.234 -> then you get a fluid,

361.527 -> and then you warm it further up, you get steam.

364.03 -> And if you then increase the temperature even further

365.865 -> then you get plasma.

367.325 -> Common plasmas include lightning,

369.869 -> neon lights,

371.162 -> and these things.

373.456 -> We need temperatures

374.791 -> 10 times larger than in the solar interior.

377.585 -> So this is about 100, 200 million degrees.

380.338 -> And only at those sort of temperatures

381.964 -> do you get fusion to happen here on Earth.

385.093 -> Heating something

385.927 -> to 10 times the temperature of the sun

388.137 -> is, to use the technical term, very hard.

391.766 -> Scientists have been working at it since the 1930s.

395.269 -> The major breakthrough has been

396.813 -> the first tokamak experiments in the 1960s

399.232 -> in the former Soviet Union.

401.15 -> They achieved several million degrees,

403.736 -> and this was a real breakthrough.

405.947 -> Tokamaks is still one of the most popular ways

408.241 -> to create fusion.

409.659 -> That's what JET is,

410.952 -> along with many other government-run reactors

413.287 -> around the world.

415.081 -> Using powerful magnets to contain the plasma,

417.625 -> they've achieved temperatures of 100 million degrees

420.461 -> and well beyond.

422.505 -> But there's still one big milestone we need to reach

425.466 -> before fusion power becomes a reality,

428.928 -> net gain.

431.347 -> We need more power out than put in to heat the fuel

435.101 -> in the first place.

436.686 -> If we can't generate more power from fusion

438.938 -> than we put in

440.148 -> then the whole thing's a bust.

442.233 -> Unfortunately, no one's ever done it,

444.694 -> not even the brilliant minds at JET.

447.822 -> That's all kind of part of the plan though

449.907 -> because JET isn't actually designed to solve fusion

452.702 -> all by itself.

454.078 -> It's just set up for a much bigger project called ITER.

458.708 -> Uh, no, no.

459.542 -> ITER.

461.294 -> We are building this experimental machine

465.59 -> in the south of France, ITER, which is going to be big.

468.468 -> And it's the first one that will produce really more energy

471.679 -> than what it consumes.

473.973 -> ITER is a massive international collaboration

476.559 -> between 35 different countries.

478.644 -> And everyone involved seems pretty confident

480.897 -> it's going to get to net gain for the first time.

484.775 -> So fusion's just around the corner, right?

488.196 -> Well...

489.363 -> With the ITER project, the first plasma

492.033 -> is supposed to be created by 2025,

495.828 -> but full fusion reaction isn't expected until 2035.

501.125 -> So if ITER were the only bet that we were making on fusion

505.838 -> then it would be a safe bet to say

507.423 -> that nuclear fusion is going to take decades,

510.343 -> and that's too long for the climate fight.

514.472 -> Even the most optimistic scientists

516.474 -> think that fusion power may not be developed for 50 years,

519.977 -> if then,

521.437 -> if ever.

523.189 -> Some think we can get there much faster.

531.989 -> There's more than 30 private fusion companies

534.659 -> around the world.

536.077 -> And it feels like every month or two

538.538 -> there's another private company that springs up somewhere

541.666 -> with another great idea about how to do this.

544.71 -> The amount of funding going into fusion

546.337 -> has also been scaling.

547.797 -> There's more private funding going into fusion per year

551.008 -> than there is federal government funding

552.51 -> in the United States.

553.886 -> The private fusion space is still small

556.472 -> in terms of budget and people

558.558 -> when compared to the mainstream.

560.351 -> But if you look at the rate of progress,

563.02 -> I would say it's much, much faster.

565.356 -> And I do think that it will be the private side

569.36 -> that produces the first vital technology.

577.493 -> Canadian fusion company General Fusion

579.954 -> is steering away from the traditional tokamak design.

583.708 -> If you have followed fusion at all in recent years

585.96 -> you might have come across this steampunk octopus,

589.297 -> an earlier prototype of their reactor.

592.174 -> It's very analogous to

595.636 -> a fusion version of a diesel engine.

597.471 -> So you basically,

598.806 -> you have this very large cavity

600.766 -> that's opened up inside of liquid metal.

604.77 -> And into that liquid metal

606.314 -> we inject a high temperature plasma of hydrogen.

611.527 -> We can now perform this compression to heat up that plasma

615.698 -> very much the same way you think about a piston

619.201 -> compressing and heating the fuel in a diesel engine.

622.455 -> This is a steam-driven compression process

625.124 -> using an array of drivers.

627.543 -> It compresses and heats this magnetized plasma

631.255 -> to fusion conditions.

634.258 -> General Fusion's reactor

635.468 -> would create brief bursts of fusion energy,

638.095 -> an approach they hope will achieve net gain more easily

641.057 -> than a tokamak.

642.808 -> Founded in 2002, and supported by investors like Jeff Bezos,

647.188 -> they're one of the furthest along

648.689 -> in the field of fusion startups

651.025 -> with plans to build a demonstrative plant by 2026 in Oxford

655.112 -> on the same scientific campus as JET.

658.157 -> This thing won't actually put megawatts on the grid,

660.66 -> but it will prove that our approach to fusion

663.996 -> in a power plant relevant environment

666.207 -> can actually make fusion happen.

668.876 -> Aiming to break ground in a matter of months,

671.128 -> the company is running full tilt

673.172 -> to work out all the kinks before showtime.

677.009 -> This is about 2/3 of the full scale

678.844 -> that we'll need in our fusion demonstration plant.

681.472 -> So it's mostly at this point

683.766 -> a question of understanding the properties of the plasma

687.186 -> and the plasma physics,

688.187 -> to be confident that that will scale

690.398 -> as we build the larger version.

693.818 -> The alarms you hear when we start charging

695.82 -> are alerting you to the fact

696.946 -> that we're starting to charge this machine.

699.99 -> You hear a ping from when the plasma

702.66 -> is interacting with the wall of the machine.

711.627 -> You see, at the time you heard that thump,

715.005 -> that's when the machine created the plasma.

719.552 -> It's really exciting to see

721.011 -> what's happening in the private fusion industry.

725.057 -> Of course, I came to General Fusion

727.017 -> because I like General Fusion's technology.

729.729 -> That's one man's opinion.

731.814 -> I think we have,

733.441 -> my view is the first really good shot on goal.

736.569 -> But what really feels good

738.279 -> is there's going to be lots of shots on goal, right?

740.573 -> And I'm very confident

742.408 -> we're going to score the win that we need.

746.954 -> Not far from General Fusion's headquarters,

748.956 -> you can find another fusion company.

752.042 -> Helion Energy, one of the most buzzed about

754.17 -> fusion companies in the world

755.796 -> is hunkered down in some former Boeing hangers

757.923 -> just outside Seattle.

760.259 -> What we have tried to do at Helion

761.927 -> is approach fusion from a different direction

764.138 -> than a lot of other people.

765.723 -> We looked at the state of the art

767.433 -> of what was being built in fusion

769.351 -> and thought really there has to be a better way

771.729 -> to get to commercial fusion faster.

774.815 -> That might sound like hype,

776.4 -> but there really is something radically different

778.903 -> about Helion's reactor.

781.238 -> Almost every idea behind a nuclear fusion startup

784.742 -> or a company

785.576 -> is to rely on the heat that is generated by fusion.

790.289 -> And then converting that heat into steam,

793.667 -> which is then used to turn turbines to generate electricity.

797.63 -> Now, Helion says that it doesn't have to go through

800.966 -> that heating and turning a turbine phase.

805.179 -> So we do something called direct energy conversion

807.556 -> where we take the magnetic energy of the fusion system,

810.392 -> the charged particle energy of the fuel,

812.561 -> and directly extract that to electricity.

815.231 -> We inject our fuel.

816.482 -> We magnetically compress that fusion fuel.

819.568 -> Fusion begins.

820.694 -> It pushes back on that magnetic field.

823.155 -> So, a good analogy is regenerative braking in your car.

826.325 -> We then directly regeneratively

828.369 -> take the electricity back out of that fusion expansion

831.08 -> and turn that into electricity.

833.791 -> By cutting out the steam step for fusion

836.377 -> we can radically increase the overall engineering efficiency

839.672 -> of the system.

840.84 -> We aim for a system now that can be much smaller,

843.717 -> required a lot less of the complexity and the challenges,

846.387 -> and really, from my point of view, a lot less of the time.

850.599 -> That pitch was good enough

851.851 -> to get Silicon Valley giants Sam Altman and Peter Thiel

855.062 -> on board as investors.

857.064 -> The company's $500 million Series E round

860.401 -> makes them one of the best-funded fusion companies

862.945 -> in the world.

864.822 -> They're also making one of the most ambitious predictions

867.867 -> in the industry.

869.201 -> The goal is to get this built by 2024,

872.58 -> running, generating net electricity from fusion

874.832 -> for the first time.

876.542 -> Timeline is the driver.

877.71 -> It's always the driver.

879.128 -> And so if it's a question of,

880.379 -> well, it could be a little bit better,

881.672 -> but take an extra year,

883.048 -> we say, no, we're going to make it a little bit worse

885.009 -> but get it done a year earlier.

887.052 -> It's the Silicon Valley mentality

888.554 -> of how can you build as absolutely as fast as possible?

897.146 -> Yeah, help me with the gas pressure.

898.439 -> Everything's isolated.

900.774 -> You got the gas pressure.

903.027 -> Fire. Firing.

906.989 -> Nice.

907.823 -> Good shot.

909.366 -> That was easy.

911.368 -> Back in Oxford,

912.453 -> First Light Fusion is taking what might be

914.955 -> the most original approach of any fusion company,

918.208 -> borrowing from a branch of fusion

919.793 -> called inertial confinement.

923.672 -> The idea behind inertial confinement is you hold plasma

928.177 -> for a very short time

930.638 -> in a very small space.

932.556 -> So, one example of inertial confinement

935.225 -> is where lasers, very high-powered lasers,

938.145 -> are used to heat a very small amount of hydrogen

941.357 -> for nanoseconds, that is billionths of a second.

945.402 -> This has been done before,

946.946 -> most famously at the National Ignition Facility

949.657 -> in California.

951.158 -> But First Light has come up with a new approach.

954.37 -> We call it projectile fusion.

955.829 -> We have a high velocity projectile.

957.957 -> It flies in and it hits into our, what we call a target.

961.502 -> And the target has to focus the energy of the projectile

964.713 -> into the fusion fuel.

966.131 -> This is one of our targets.

968.926 -> This is the key technology to our approach to fusion.

971.553 -> So this is completely turned into a plasma

975.099 -> by the force of the impact and the energy released.

980.646 -> In a power plant, one of these targets

983.232 -> would release enough energy to power the average UK home

985.859 -> for over two years.

988.112 -> That sound.

990.739 -> That's the projectile being fired out of this gas gun

993.659 -> at around 15,000 miles an hour.

996.745 -> To generate power

997.705 -> you have to do that at a repetition rate.

999.665 -> You have to do a certain number

1001.917 -> and it's the energy you release every time

1004.461 -> times by the frequency

1005.421 -> and that's the power.

1006.588 -> So in our power plant design

1008.09 -> we'd be doing this about once every 30 seconds.

1011.468 -> So we recently showed fusion in our lab

1014.763 -> with a projectile driven approach for the first time.

1017.307 -> Just simply experimentally it's a big proof of concept

1019.518 -> that it can really work.

1020.894 -> If we look at the actual amount of fusion we produced

1024.314 -> then that number by itself, it was 50 neutrons.

1026.9 -> And we're not hiding it.

1028.193 -> It's not very impressive.

1029.57 -> But the point is

1030.738 -> that's exactly what the simulations predicted.

1032.99 -> And that's what gives us the very rapid, we hope,

1036.702 -> pathway forwards to improve that number.

1040.372 -> I hope we're not talking about 50 neutrons

1042.916 -> at the end of the year.

1047.963 -> So there's these disruptive technologies

1049.548 -> which are coming into our space, which is great.

1051.717 -> We want new ideas.

1052.76 -> We want people to come and tackle

1054.47 -> the big challenges that we face.

1056.013 -> There's lots of startup companies.

1057.806 -> Some of them will fail.

1059.183 -> Some of them will go to the wall.

1060.476 -> Some of them will wildly succeed.

1062.394 -> That's just healthy.

1064.354 -> Some on the scientific research side of fusion

1066.899 -> are more critical of the startup phenomenon.

1069.943 -> Few private fusion companies have shown any results

1072.529 -> beyond fusing a few atoms.

1074.823 -> And many scientists worry that these startups

1077.493 -> are promising too much.

1079.912 -> The cons in my mind

1081.705 -> are the promises by companies

1084.416 -> who either have a concept

1085.959 -> that we have already, for good reason, put apart

1088.253 -> and knew why they won't work,

1090.798 -> or completely new concepts.

1093.175 -> Some change them every other year.

1096.386 -> And then they still promise fusion energy

1098.722 -> by 2030 or something.

1101.266 -> And I don't like that too much because I'm afraid

1103.435 -> that if there are so many promises not kept

1107.272 -> that this would not be positive for fusion energy

1109.942 -> in general.

1111.568 -> If you look at the timescales that governments work at

1116.99 -> in research,

1118.617 -> that's completely different than the timescales

1121.036 -> that entrepreneurial private industry works at.

1123.956 -> It's not that there's a problem there with government,

1126.375 -> but, you know, they're a different set of metrics there.

1129.211 -> They're largely research-driven organizations.

1132.84 -> You need look no further than commercial access to space.

1136.677 -> SpaceX.

1137.803 -> They've created what people thought would never be possible,

1141.306 -> which is reusable rockets.

1143.809 -> NASA was never able to achieve that

1146.145 -> because they weren't motivated by the same things.

1149.982 -> And so when it comes to the last mile of commercialization,

1153.861 -> that's where you really need to hand the baton

1156.113 -> from government to private industry.

1159.658 -> From what I know, some of the private sector

1161.994 -> is very serious

1163.579 -> and is bringing

1165.164 -> significant capital and significant engineering advances.

1170.002 -> It needs to promise things

1173.213 -> on the time

1174.047 -> on a realistic timescale.

1175.34 -> So, not next year.

1177.593 -> And the timescale of about 20 years is realistic.

1182.723 -> It remains to be seen

1184.016 -> whether the Silicon Valley VC type approach

1186.685 -> will actually get us to commercial fusion faster.

1191.607 -> One thing is for sure though,

1193.525 -> progress, however slow and incremental, is being made.

1197.738 -> And if we're going to get to a fusion powered future,

1200.866 -> we're probably going to need both the painstaking research

1204.328 -> and the risky new ideas.

1207.08 -> I think the rate of progress is really driven

1208.707 -> by the imperative.

1209.791 -> And to be honest, I mean, let's just be honest as a society,

1212.961 -> over the last 20 or 30 years

1214.254 -> we've been pretty comfortable with burning fossil fuels.

1216.34 -> The fact that we have to stop it

1218.508 -> now drives the imperative.

1220.26 -> This is the most important question for mankind

1222.471 -> because, I mean, you need electricity or energy

1225.057 -> for everything.

1226.558 -> We have a lot of developing countries,

1228.477 -> and if these countries

1229.645 -> want to come to our standards of living,

1231.355 -> there's no way other than provide them with cheap energy,

1234.566 -> and better, CO2 neutral.

1236.318 -> Otherwise, we have a climate crisis.

1239.571 -> Each fusion technology

1241.198 -> has a slightly different flavor

1243.158 -> in terms of its ultimate value proposition.

1246.119 -> And so I think that you're going to see

1248.997 -> a portfolio of technologies being commercialized

1252.292 -> over the coming decades

1253.835 -> that address different parts of the market.

1255.963 -> And the market is huge and diverse.

1258.548 -> The total addressable market for fusion

1261.927 -> is on the order of $1 trillion a year.

1264.888 -> It might be decades

1265.931 -> before fusion becomes a reality,

1268.6 -> but it's the kind of cheap, unlimited power

1273.021 -> that you would want to give a civilization

1276.566 -> to be able to do all the things it wants to.

1280.946 -> Imagine if you could make a lot of rocket fuel

1283.532 -> to create an entire fleet of rockets

1285.534 -> that would go and mine asteroids

1287.369 -> because, you know, at some point

1288.745 -> we're going to run out of metals.

1291.206 -> Or you could remove

1292.374 -> existing carbon dioxide in the atmosphere

1295.085 -> back down and bury it deep underground.

1297.838 -> That process requires a ton of energy,

1301.091 -> but if you were able to create nuclear fusion

1303.802 -> you could imagine restoring the Earth's atmosphere

1306.68 -> to pre-industrial levels.

1309.808 -> And these are the kinds of application

1311.351 -> that really cheap and clean energy

1314.146 -> can enable humanity to do.

1320.569 -> I want my kids to have a future

1323.947 -> that has the possibilities that I've had.

1328.91 -> So I hope we get fusion,

1331.705 -> and I think private, public,

1334.082 -> it all needs to come together because this is worth it.

Source: https://www.youtube.com/watch?v=Dp6W7g9no0w