Science AMA Series: We're scientists at the MIT Plasma Science and Fusion Center. We broke the world record for pressure in a magnetic fusion device on our tokamak’s last day of operation. AUA about our scientific work, our future as a lab, and fusion in general!

Abstract

Hi Reddit, we’re scientists working in the Plasma Science and Fusion Center at the Massachusetts Institute of Technology. Last time we were here, redditors helped to extend the life of our tokamak, Alcator C-Mod, to FY2016. This time we are here because we set a world record plasma pressure in Alcator C-Mod on our last day of operation! While C-Mod’s shutdown is definitely a loss to our community of scientists, we’re now turning our attention to the next generation of fusion reactors. We’re particularly excited because we’ve begun to work with high temperature superconductors to design much smaller fusion reactors than ever before, opening up the “high-field approach” to fusion.

You can read about our record here, or watch an awesome 360-degree tour of our machine on YouTube.

Today we have several research scientists, postdocs, graduate students, and faculty in the C-Mod control room to answer your questions, so ask us anything about our work, future plans, or about fusion in general!

We're now live in the control room of Alcator C-Mod until 4pm EST. Ask us anything!

Currently here:

Prof Ian Hutchinson - MIT Professor of Nuclear Science and Engineering (ih)

Bob Mumgaard - Research Fellow (bm)

Seung Gyou Baek - Research Scientist (sb)

Adam Kuang - Ph.D. Candidate (ak)

Norman Cao - Ph.D. Candidate (nc)

Libby Tolman - Ph.D. Candidate (lt)

Alex Creely - Ph.D. Candidate (ac)

Sean Ballinger - Ph.D. Candidate (sean)

Alex Tinguely - Ph.D. Candidate (at)

EDIT 4:08PM: Alright, looks like it's time for us to pack up. Thank you for all of your questions! We'll be still be checking in on this thread regularly so feel free to post any other questions or comments you have :)

First of all, congratulations to your results and to the media coverage you got!

  • (1) What were the plasma temperature and density during that record shot ?

  • (2) What was the background magnetic field strength and the related plasma beta value ?

  • (3) What made that shot a record shot, i.e. was it particularly high heating power or better <something> than usual ?

  • (4) For how long could you sustain that record pressure ?

  • (5) Did you do it on purpose at the very end of the experiment's life because you were afraid of unusual high power fluxes (due to the high pressure) to the plasma-facing components ?

  • (6) Did the plasma-facing components (or anything else) got any damage during that record discharge ?

  • (7) What will happen now with the Alcator C-mod device and the people working on it ?

  • (8) If I understand it correctly, your approach is somewhat different (although very similar) to ITER in that sense that you have a rather small device achieving nevertheless high density and temperature due to the higher magnetic field (correct me if I am wrong). There was once a device called “IGNITOR” which was planned to achieve some fusion power output, do you know the status of that device ?

  • (9) Does the comparable high magnetic field strength in your approach prevent you from using superconductors at a reactor size level (or IGNITOR-size) ?

  • (10) Will the scientific details about that shot be presented somewhere (or are they already somewhere?) ?

  • (11) Does this record shot have a scientific significance or was it more something like “let's see what we can squeeze out of that device” ?

  • (12) Thank you very much for reading all those questions and maybe even trying to answer them :)

I wish you all the best for your future.

DerPlasma

Great, technical questions. I'll try my best to put this into laymen terms.

Check out the FAQ about the record shot for more info: https://www.psfc.mit.edu/news/2016/pressure-record-faq

(1) The plasma temperature and density during the record shot was about 30Million C in the center, the density was 5*1020 ions per cubic meter. To put this into perspective, the 2 atmosphere shot was like gas at 100,000 times hotter and 100,000 times less dense. The density was very high for a tokamak, while the temperature was not as high as been acheived on other tokamaks.

(2) The background magnet field was 5.7T or about 140,000x the Earth's magnetic field. The Beta was about 1% which is low for C-Mod.

Check out some other traces here: https://www.psfc.mit.edu/news/2016/pressure-record-faq#q_3_5

(3) The record shot was not particularly unusual, just more highly tuned. We developed a better way to fuel the plasma to higher density for the record. But actually, we got very near record performance on shots that were very different in three different ways.

(4) The highest pressure shot had nearly that high of performance for the full anticipated length of 0.5s of heating power, it was not transient. Things had to be stopped because C-Mod's magnets were heating up from being on and the internal components were heating up from the power to and from the plasma. This is normal.

(5) There were many reasons to push this to the end of the experiments life time. One was we were learning as we go. We did not particularly stress the machine, only perhaps marginally more than usual. The scheduling for collaborators coming in also played an important part since many team members were traveling to C-Mod for the last day and we wanted everybody to participate. More can be found here https://www.psfc.mit.edu/news/2016/pressure-record-faq#q_3_7

(6) We do no think we damaged anything. We did not see any deleterious effects in the plasma nor on cameras or signals and plasmas ran well afterwards. We'll go inside the machine to inspect next week. See how we do this: https://www.youtube.com/watch?v=q3pQgozt3cM

(7) The device is now undergoing various calibrations so we can interprete the reams of data. Once we're done with that we'll put it in "cold storage" which is one step up from mothballing it. It is unclear what it's future is. We'd like to see it used for something scientifically interesting (Like converting it for this experiment: http://www.psfc.mit.edu/research/topics/adx-advanced-divertor-experiment) but it will be up to the DOE. The science team will continue to analyze data from C-Mod and will also continue collaborating as a team on the other facilities around the world. We are at the same time developing new technologies that will enable another generation of even more powerful tokamak reactor.

(8) Our device is similar to all tokamaks which includes ITER and IGNITOR. These devices are distingished by two important parameters, their size and magnetic field. If you make the field higher the size can be smaller and it is very non-linear. But there are limits from technology for the magnets. IGNITOR is still a Russian-Italy project but has not seen much progress recently due to funding, likely due to econimic down-turns.

(9) Previously the high field approach with tokamaks required using copper magnets (like C-Mod and IGNITOR) because superconductors don't work at the required magnetic field. However, we're really excited about building tokamaks at around or above 10T using new high-temperature superconductors: http://news.mit.edu/2015/small-modular-efficient-fusion-plant-0810. This new magnet technology enables IGNITOR sized experiments that could simply not be done with superconductors before. But for a toroidal reactor (like a tokamak or stellerator) the minimium size is set by the neutrons penetrating the wall and getting to the magnet.

(10) The scientific details were presented at the IAEA fusion conference this week at Kyoto http://www-pub.iaea.org/iaeameetings/48315/26th-IAEA-Fusion-Energy-Conference This is the premier world-wide conference. The results will be published in a journal article that will come out soon (it is a slow process) The results will also be presented in two weeks at the American Physics Society's Plasma Physics meeting in San Jose which is the annual large conference on plasma physics.

(11) The record shot does have scientific significance because it was significantly higher and longer than had been done before and we used a new way to fuel that was somewhat unexpected to work. Additionally, the near-record shots were in some ways even more scientifically interesting since they saw new phenomenon in the edge of the plasma and expanded the range we could get favorable operation in many ways. All the of the shots expand the international database for how tokamaks work in new directions. ...Personally though, nothing beats passing through an integer in an important parameter. P=1.95atm just doesn't seem as good as p=2.05atm....

(12) You're welcome. Glad you asked some good questions.

  • BM

Considering that this is the technology that would be most instrumental in Saving Humanity from our own Extinction, how much money is needed to fast-track this technology? There must be a place that we can crowdfund this into a Philadelphia Project level experiment. Basically do you have a Kickstarter page?

icansolveanyproblem

This is a bit of a tricky question but I will try my best to answer it. So to fast track fusion energy and build a reactor that would actually put electricity on the grid. You would need to go through the standard design cycle. You brainstorm -> design -> build -> test -> fail -> learn.

However the problem at the moment is that each of these iterations is taking too long and costing too much money. We need to find a way to streamline this learning process. To this end, the scientists here at the PSFC have decided that we've got to make smaller reactors that can be built faster and cheaper. We've got a REALLY good idea of how well they will perform and if everything goes smoothly we're not actually that far away. Even at these smaller cheaper machines though, each of them will be in your 100 million - 1 billion dollar mark.

We haven't started a kickstarter page but it is a good idea :)

~ak


What can I do to help humanity reach the fusion power dream faster?

I just got my bachelors degree in computer science, majoring in software development. I'm from Finland if that matters.

futurefusionmaster

Hi! The best thing you can do to help humanity reach the fusion power dream is getting excited about fusion, and using that excitement to get involved in something that interests you. This can take any number of forms: pursuing higher education to do research in the field of fusion, writing to your local government to push for more funding of fusion and science-related education, learning about fusion through books and the web and spreading that knowledge, or simply sharing exciting news articles like these with people you know. -nc


For fusion reactors, one significant limitation seems to be radiation damage to the first wall. What is the current status of research into first wall material and what are some promising candidates?

Is materials degradation likely to remain a significant technical challenge to the development of fusion reactors, as it has been for fission reactors?

adirondack928

Radiation damage from neutron bombardment needs to be considered for all the reactor components. We already know a lot from fission reactors about radiation damage, but fusion neutrons are more energetic on average so we need to supplement our knowledge. Small sample tests can be done on existing facilities, but in the longer term the big components need to be tested and proved in a dedicated fusion materials test facility. ITER will also give limited information on this topic.

The plasma bombardment of the plasma facing materials is another whole issue. ~ih


Why is aneutronic fusion (specifically Lawrence Plasma Physics approach) so overwhelmingly underfunded? Their reports make it sound that with just a bit of funding they could almost completely change the fusion field forever? Are they merely hyping?

carutsu

Aneutronic fusion means fusion with no neutrons produced by the reactions. There is only one remotely plausible reaction with this property Proton Boron11. This reaction is at least a 100 times harder to make work than the approach most people are taking which is Deuterium Tritium (DT). That's why we are concentrating on DT as the most direct route to fusion energy ~ih


Is there any expected time when the human society will be able to make use of fusion energy abundantly?

aturtlefromhongkong

This is always a really important but difficult question to answer. Short answer: 2030s??? Plus minus a decade?

Long answer: Scientifically we expect to be able to harness fusion energy. There are many approaches to try to do this, some will likely flame out, but everything we know about the nuclear reactions and the plasma physics says it will work (see: The Sun).

But... Science is not the only part of this. Technology is important as well since it enables harnessing scientific understanding. This is getting better and better as new materials (See REBCO superconductors http://www.psfc.mit.edu/research/topics/high-field-magnets) are invented and manufacturing improves.

And Society and Economics are important since they set the acceptance and the capital to make progress. People have to want the products at a price that can turn a profit.

Historically technologies can seem like they are a long way off and then suddenly all of these parts come together and things accelerate quickly. For example, fission went from first light bulbs to big privately funded power plants in 8 years! https://en.wikipedia.org/wiki/Nuclear_power_in_the_United_States#/media/File:US_Nuclear_Electricity_1949-2011.png

We think that fusion may be approaching this breakout point. The science is looking more and more sound. The technologies are developing. There is more and more need for clean abundant power to combat climate change.There is increased interest in the private sector in funding. The slowest path is likely the governments building ITER and they plan on power in 2050. But many fusion scientists think it could be even sooner, like in the 2030s. Some say even sooner like the 2020s but personally, I don't think this is credible from what I have seen.

-BM


What is the likely best method for extracting electric power from fusion devices? Are there any attempts to overcome the Coulumb force other than applying energy?

MuzzleCream

The heat carried by neutrons from fusion reactions will be absorbed in the lithium blanket (or the first material wall) with the heat exchanger. From there, it is similar to conventional power plants with a steam generator and a turbine that produces electricity. You are right that one needs to achieve very high temperatures to overcome the Coulomb force. Two major approaches to confine plasmas magnetically or inertially.


The following question is not a misspelt piece of cross-universe Fallout/Marvel fanfic:

Will it be your laboratory that moves on to work on the REBCO based ARC reactor? If yes, then is the end of the Alcator C-Mod really a loss to science, or is it putting ole Blue out to pasture so you can put what you learned to use?

Oddball_bfi

It's great that you've heard about REBCO and the ARC reactor! We think that REBCO is an incredible new technology and will hopefully have a huge impact on the future of fusion research. We definitely plan on (and already are) doing quite a bit of research and development of the new REBCO technology. Hopefully we'll be able to apply it in new high-magnetic field fusion experiments soon, allowing us to build off of everything we've learned from C-Mod. The ARC reactor would be a much longer term project, so isn't something that we're planning on pursuing immediately. Along those lines you're exactly right, though. We're sad to see C-Mod go, but it does open up tons of new possibilities for future research! ~ac


I am a current undergraduate hoping to pursue graduate studies in plasma physics. I was wondering what your thoughts are on the funding situation for plasma physics, both at your institution and at universities in the United States in general. Specifically at your institution you mention switching from the C-mod to other projects, will this result in a general downscaling of resources?

alecslo

I'm excited you're considering graduate school in plasma physics! For me, it's been very fun and rewarding. As far as funding goes, the first thing to remember is that plasma physics is a very broad field. Plasma physicists work not just on fusion projects, but also on space plasma (check out the work of MIT's Nuno Loureiro here: http://web.mit.edu/nse/people/faculty/loureiro.html ), novel technologies (see one related to nuclear disarmament here: http://www.pppl.gov/news/2016/09/pppl-and-princeton-demonstrate-novel-technique-may-have-applicability-future-nuclear ), plasma thrusters (http://www.nasa.gov/centers/glenn/about/fs23grc.html), and many other topics. These areas of research are funded by a wide variety of sponsors, so it's hard to make blanket statements about the funding of the entire field.

For our particular situation, it's true that the loss of C-Mod funding deprives us of the opportunity to do tokamak experiments here at MIT for now. But our lab is funded very strongly to continue research collaboratively at other fusion experiments throughout the world, and to continue different forms of research here. We're also thinking carefully as a lab about what sort of fusion experiment we might like to build next, and how we might fund it. These thoughts are early stage and we can't give you details yet, but we're optimistic about our future! -lt


What would be the result of a sudden catastrophe loss of your magnetic containment field? What safeties and redundancies exist to keep the fusion running safely and smoothly?

EnigmaEcstacy

These events, when the plasma suddenly terminates, are called disruptions and have the potential to damage the vacuum vessel. The rapid loss of thermal and magnetic energy can create high heat fluxes, large currents/forces in the walls, and even relativistic particles inside the machine. Disruptions happened all the time in C-Mod, on about 1/3 of all shots. (Note that our record-breaking plasma pressure shot did not disrupt.) But the tokamak can be designed to handle disruption forces and heat loads, and mitigation schemes are also being studied. For instance, we can puff gas into a disrupting plasma to more safely radiate away the thermal energy. Scientists are also using physics-based and machine-learning algorithms to better predict and avoid disruptions. Finally, MIT is interested in the high magnetic field approach to fusion, which actually can make disruptions less likely as we are operating in regimes farther from known limits. - at


Is the funding of those operations sufficient ? I sometimes read about scientific achievements not reachable because priorities lie elsewhere.

Is that the case for those advances too ? I hope not, this seems to be one of the most important improvements that could happen to human energy sources.

Zodyac

Funding has been far from ideal for a long time. In FY2017, the US fusion program is going to represent around 7% of the Department of Energy's budget, itself 0.5% of the $1 trillion discretionary spending budget of the US government. It's a third of what we spend on particle accelerator research, so there's lots of room for improvement!

Right now, the US fusion program is being scaled back to help fund the ITER experiment, a huge, internationally-run tokamak projected to produce a burning plasma that creates more energy than is required to sustain it. ITER is a very important step for fusion, but we think we could achieve similar results faster using small tokamaks with very high magnetic fields that take advantage of newly developed high-temperature superconducting magnets.

Fusion would indeed be an ideal source of energy in the long run, and we would definitely see faster progress with more funding. -sean


My understanding is the larger the reactor the closer the machine gets to producing more energy than it consumes. What are the major pitfalls in scaling the device to one that produces more energy than it consumes? Is it the magnets? If so, what physical property of magnetism causes the asymptote in field strength?

mt1336

Whether more energy is produced that lost from the plasma depends how fast it is lost; in other words how good the confinement is. Making the plasma bigger slows the energy loss rate. We know this from theory and from experience with experiments. However, since we don't yet have a complete fundamental theory of the important loss processes there's some uncertainty in the extrapolation. That's why the proof requires us to demonstrate good confinment. ITER is planned to do that experimentally. ~ih The magnets are important, but not the major uncertainty.


Hi guys, not a native speaker here so bare with me.

  • I was under the impression that the Tokamak has to work in pulses rather than continuously. How does this affect the use of Tokamaks in energy generation? Isn't the Stellerator better suited for that application?

  • Are there any recent improvements in regards to keeping the plasma stable over a longer period of time? If I remember correctly that's what keeps us from actually reaching the triple point. How far off are we still?

Glad you take time to answer some questions!

Finrod04

Most Tokamaks today have plasma current driven by induction, which limits their pulse duration. It is planned that a reactor will use non-inductive current drive and operate steady-state. There have been a few experiments that have demonstrated this is possible, and have operated plasmas for hours at a time. ITER will run inductively for up to 1000 seconds at a time, but may also run steady state, although at reduced performance. The stellarator is a confinement scheme that does not need plasma current and so does not face this challenge. That's a big potential advantage, but stellarators don't yet have as good confinement as tokamaks. It's a good thing to pursue different approaches, with different advantages and disadvantages.

The key challenge preventing us reaching ignition or a burning plasma state is to have sufficiently good confinement. Heat leaking out slower than it is generated by fusion reactions inside the plasma. The losses are affected by small scale turbulence arising from unstable perturbations. But the plasmas are expected to be stable on the larger scale. Indeed they need to be!


Im curious, since the record was set on the very last day of operation how close did you come to the safety margins? Was there any damage to the reactor as a result?

mTesseracted

No there was no damage to our tokamak. It is still in working order, but we are not funded to operate it. It takes a team of engineers and technicians for that. We did operate that day at our the highest fields we have use. However, our device was designed for even slightly higher fields, so we did not have to approach safety limits.


What are fusion reactors and how do they work?

J29736

Hi! So great that you're curious :) You may be familiar with fission, which is the nuclear process which splits a nucleus into smaller parts, releasing energy. Fusion is in some ways the opposite. It happens when the nuclei of two light elements (for many fusion reactors, deuterium and tritium in particular) come together, releasing energy. One way to make fusion happen is to take a plasma (basically an ionized gas) and heat it up really hot. It's difficult to do this because you have to heat up the plasma so hot that it would damage a material container. So fusion reactors use really strong magnetic fields to hold the plasma in place, and then heat up the plasma by driving a plasma current and using radio frequency heating (and some other methods as well). This allows fusion to start happening. In the case of a DT reactor designed to produce electricity, this fusion would create very energetic neutrons that would then heat up a blanket, allowing the production of electricity. Let us know if you have more questions! -lt


How much progress have there been made in the past 10 years or so in energy production by fusion? Have there been any major breaktroughs and will there soon be one?

Tiiqr

In the past 10 years, there has been an advance in understanding plasma physics in confined plasmas with the developments of advanced experimental diagnostics and simulation capabilities to understand the plasma behaviours. More recently, a lot of focus is given to the high temperature superconducting magnets that was not available before, which will allow to build a tokamak at a smaller volume but at a faster time scale.

In the near future (within the next 20 years), ITER being built on France will demonstrate the output fusion power (~500 MW) can exceed the power you put into the plasma (~50 MW). ~sb


Why is progress so slow and are giant strides expected with the next generation of fusion experiments?

Shitbadger16

Developing new large scale technology takes a long time. That's true across the board, but fusion has a particular problem in that we don't know how to make it generate small amounts of NET energy. Therefore we can't prove that fusion will work in a small experiment and then gradually scale up to larger scale. Instead we have to go to full scale for the proof of feasibility. If the next steps like ITER or like a more compact burning plasma are successful, then we will have demonstrated scientifically fusion energy at the needed scale. There will still be engineering challenges to reliability and so on, but I think an experimental demonstration of burning plasma would be a huge stride.


Please forgive the errors that are likely in my questions, I am not a scientist, merely an enthusiast.

The two atmosphere pressure of the plasma was magnetically confined? That was within the reaction chamber. What was the pressure in the chamber outside of the magnetic bottle? How tightly was the plasma contained? What was the temperature of the plasma?

I find this absolutely remarkable! Thank you so much for doing this AMA!

joegee66

The plasma 2 atmospheres was contained by magnetic fields that had a much higher net (magnetic) pressure roughly 200 atmospheres. Those magnetic fields are generated within our electromagnets and withstood by a massive stainless steel mechanical structure. The plasma temperature was 30 million degrees Celsius. ~ih


Good work!

Did you do this on your last day because if something broke due to the high power, it didn't matter?

On a more serious note, 5.7T isn't that big of a magnetic field, especially for superconducting magnets. Why not get a bigger magnet?

Third, what is the highest temperature superconductor you guys have worked on? Are we anywhere near practical liquid nitrogen temperatures superconductors yet?

arden13

Obviously the knowledge that there was less to lose meant we were less nervous about breaking things, but there was no damage to our tokamak. It is still in working order. We are just not funded to operate it. It takes a team of engineers and technicians for that. We did operate that day at the highest fields we have used, namely at 7.8T which is bigger than the 5.7T at which the record was set. Actually, our device was designed for even slightly higher fields, so we did not have to approach safety limits. The reason the record was at less than full field was because our RF heating was not designed to work so well at 7.8T. If we'd had the funding (previously) to implement the best high-field heating, I'm confident the record would have gone a lot higher.


Good work!

Did you do this on your last day because if something broke due to the high power, it didn't matter?

On a more serious note, 5.7T isn't that big of a magnetic field, especially for superconducting magnets. Why not get a bigger magnet?

Third, what is the highest temperature superconductor you guys have worked on? Are we anywhere near practical liquid nitrogen temperatures superconductors yet?

arden13

To the second point about the strength of the magnetic field.

You are right, that 5.7T is not that high for superconductors. There are small magnets around the world that can go higher than C-Mod's 8T field for things like NMR. However, it is quite high for a large magnet. which has to be very strong and carry lots of current in a small area. Most tokamaks are at 2-3T superconducting or not, C-Mod being the exception.

There is also a subtlety with the field. The superconductors care about the field at coil which is in the middle (the hole of the donut) The field here is usually about twice as high as the field at the plasma (Maxwell's equations say so). Traditional superconductors become very poor superconductors above about 12T. (see http://fs.magnet.fsu.edu/~lee/plot/Archive/Jeprog-070813-1358x974-256pal.png and look at the NB3SN line, it is dropping fast). This limits superconducting tokamaks to about 5-6T at the plasma using old superconducting technology, this is what ITER is at. The new superconducting technology, HTS or REBCO doesn't drop with field allowing much higher fields to be built. We'd like to build tokamaks at 10T at the plasma or 20T at the coil using this superconductor.

For reference for magnetic fields. Most MRIs are at 1.5T, some are at 3T. You have to have a board of doctors certify the procedure to put people in fields above 7T.

-BM


In which year do you believe fusion energy will be commercial avaible? How big, in comparison to ITER or ARC tokamak, the first power plant will be? And how much it will cost?

Alfonso_Italy

My personal opinion as a first-year graduate student is that with very serious and unrealistic funding, we have the capability to build a tokamak power plant that supplies 100-500 GW of power to the grid in 20 years. This would include the time to very quickly build experimental tokamaks, which we need first in order to study a burning plasma and relatively unstudied reactor components like blankets and divertors. At the current level of funding, projects take a long time and have conservative designs. You can't build a tiny device that scales up well to power-generating scales. You need to start with a relatively big device to get useful data. That takes a long time to build, then you discover some unexpected plasma physics, like H-mode or the need for a divertor, and you can build a new tokamak with an improved design. We would definitely see quick progress, and maybe even less overall spending, if the field had more funding today. We could build many machines, and weirder machines, each testing out the specific components that need to be developed for power generation.

Being competitive with other sources of energy will take time, but we will constantly benefit from advances in plasma science, materials, magnet technology, and manufacturing.

At MIT, we're very interested in creating commercial plants along the ARC model, which is much smaller than ITER and uses a much higher magnetic field to confine its plasma. We believe this could be a faster path to commercial viability, and we want to investigate it! -sean


Are there any waste products associated with fusion: in the way of off-gassing or spent fuel? What is needed as a catalyst to create fusion? Is there ever any worry about running out of the catalyst?

Edit: would any of these ingredients have a negative environmental cost associated with them... thereby making fusion a wash.

bizmarxie

Yes there is! Helium! But it is a safe gas, we breathed it in as kids to make our voice go squeaky. We're currently experiencing a shortage so this hopefully will be a plus!

Unlike traditional nuclear fission processes there isn't going to be long lived radioactive waste produced that will last 10,000-100,000 years. However the machine itself will gradually become radioactive over the course of operations. The nice part though is that we're talking a cooldown period of 50-100 years to become safe again. This is something that is manageable within a 'human' time frame and so we do not foresee it being a problem.

No catalyst are used in the process, only the fuel with is isotopes of hydrogen called deuterium and tritium. Deuterium is naturally available and very abundant, there is sufficient deuterium to last millions of years. Tritium is produced through reactions with lithium of which we have thousands of years worth in supply. By then the hope is that we will have designed or reactors to only run on deuterium.

~ak


Can you provide any updates on the ARC proposal? The videos on YouTube make it sound very promising.

Do your recent experiments at C-Mod help to address any of the design questions in ARC/SPARC?

boraxx

So the original ARC design did not include a divertor, which is basically the exhaust system of the tokamak. This past spring, our fusion engineering and design class, led by PSFC Director Prof. Dennis Whyte, designed the appropriate magnetic configuration and heat exhaust system for a long-leg divertor, which could adequately remove the power of a 1 GW fusion power reactor. (A paper is in the works!) C-Mod certainly has the highest divertor heat fluxes in the world, so all research there is important to ARC! We are also in hopes of a new experiment called ADX which would study advanced divertor geometries, like those that we would need for a reactor. You can read more at http://www.psfc.mit.edu/research/topics/adx-advanced-divertor-experiment - at


Thank you for doing AMA.

If I remember my physics course, gas heats up when you apply presure on it. Is it same with plasma?

Aken_Bosch

Thanks for your question!

Yes - if you compress a plasma, it will heat up! This is the idea behind inertial confinement fusion, to compress a target using lasers to heat it up to the extreme temperatures necessary for fusion.


One thing I've always wondered is how are you able to measure how much energy is given off by the fusion reaction in order to calculate efficiency? Do you calculate how hot the plasma should be when X amount of energy is supplied by the lasers factoring in absorption rate, measure the actual temperature and subtract the two? If so how do you measure the temperature of something that hot? Does the containment magnetic field factor in at all in the calculation? Thanks for the AMA!

scuzzy987

Fusion experiments have a wide variety of diagnostics that measure different properties of the plasma. The diagnostics that are most directly tied to fusion rate are neutron detectors (if you want to read an article about one type of neutron detector, see here: http://www.ccfe.ac.uk/assets/Documents/AIPConfProcVOL1612p93.pdf ). Since fusion produces neutrons, more neutrons will be present when more fusion is taking place. By comparing the number and energy of the neutrons produced to the amount of energy it takes to run the machine, we can get a measure of efficiency. We also have diagnostics which measure temperature though, and they're really cool! One of these is Thomson scattering (https://en.wikipedia.org/wiki/Thomson_scattering), which shoots a strong laser through the plasma and records how much light gets scattered. This parameter depends on the plasma density and temperature. -lt


Thank You guys so much for doing this. I apologize in advance for poorly formed questions, I am neither a nuclear scientist nor a native speaker.

(1) Are you following the news/developments from "cold fusion" (LENR) approach to nuclear fusion?
(2) Do you completely dismiss the "cold fusion" (LENR) approach to nuclear fusion as non scientific?
(3) If you had to bet/guess, with which probability would you say that Andrea Rossi's approach is a complete hoax with absolutely no results?
(4) Do you think that some funds should be invested in further investigation of Fleischmann-Pons experiment?
(5) Is there any way to directly invest in your next project?
(6) Is there any way to directly donate to your next project?

Thank You!

LovroX

1) Yes I'm familiar with Cold Fusion and was interviewed about it on CBS evening news on the day of its announcement in 1989. 2) Careful scientists don't "dismiss" things out of hand. That's why we took a serious interest in evaluating and understanding the claims being made. It took us 2 weeks to be sure that Pons and Fleischmann had made some serious errors in their experiments and interpretation. It took several months and lots of independent experiments to demonstrate that cold fusion can't be reproduced. Since then, the enthusiasts have continued to explore cold fusion, but have not succeeded in demonstrating there is really fusion going on, or that net energy can be released. 3) Rossi has a track record of dubious business dealings that ought to lead anyone to treat his claims with a degree of scepticism. When you know the physics, you are even more sceptical. 4) Not public funds, but if individuals wish to pursue it, why not? 5-6) Yes, please contact the Plasma Science and Fusion Center http://www.psfc.mit.edu/


I'm a mechanical engineering student and it's been my dream to work on fusion energy.

What advice (like what topics or institutions that I can/should look to apply to) can you give to get started in this field?

boiboi95

I did my undergrad in mechanical engineering as well! I started studying mechanical engineering because I wanted to get involved in clean energy. When I learned about fusion I realized that it was both super interesting, and hopefully an almost ideal source of clean energy.

Pretty much everything in mechanical engineering is related to fusion in some way. Fluids, thermodynamics, material properties, etc. Having a strong physics background is also super helpful. If you're looking for a fusion-specific source, here's one of the textbooks that we use at MIT:

https://www.amazon.com/Plasma-Physics-Fusion-Jeffrey-Freidberg/dp/0521733170

There are a bunch of institutions that do fusion research in the US and around the world. MIT, Princeton, University of Wisconsin-Madison, UCLA, and UT-Austin just to name a few. If you're really interested keep an eye out for graduate programs and try to decide which fits your research interests best. There are also a number of undergraduate internship programs at a variety of places, such as the SULI program from the Department of Energy (which I did as an undergrad). Some institutions will also offer separate internships to undergrads, so you may want to contact them individually.

Hope this helps!

~ac


Two questions for you all:

  1. I assume each shot generates large data sets. How are you taking advantage of the modern advances in data science and Bayesian inference? How important is data science to your work?
  2. Are you planning an autopsy on Alcator C Mod to see how the high-pressure shots affected it?
hollowpoints4

So this would a be a yes and no sort of answer. A lot of work has begun on analyzing the data using modern data science technique for more robust curve fitting tools, error propagation and other similar concepts. However, a problem emerges that in order to have a good idea of what is sufficient statistics or to apply information theory one has to generally work on either having a good model of the problem or heavily simplifying the system. Both approaches have proved very challenging on our experiments as we are essentially trying to stabilize chaotic dynamical systems. So the long and short of it is yes but there is a lot of progress that needs to be made to improve things further.

Not quite an autopsy but close! We've just brought the machine up to atmospheric pressure to open it and go inside. We've just taken a while to do an external inspection and warm the machine up so when we go in we can do an inspection of the interior.

Click on this link to see a fly through of the inside of the tokamak.

~ak


Do you think your breakthrough will help get you funded for 2017? What happens next if it does? If it doesn't?

almosthere0327

The federal funding cycle is complex, and it's hard to say what effect the breakthrough would have on prospects of C-Mod getting refunded. C-Mod is currently in a "safe shutdown" mode and could be reactivated if there was enough pressure from the scientific community to do so. If C-Mod were to be refunded, we would probably resume doing the experimental work we were doing before the shutdown. Otherwise, the experimentalists here would probably work on the collaborations we have in place with other experiments and machines across the globe. There is also a hope that major components of C-Mod could be reused on machines elsewhere.

However, with or without C-Mod, we want to continue pushing the development of fusion forward. Our broader goals revolve around developing the high-field approach to fusion. C-Mod is unique in its ability to produce reactor-relevant plasmas in such a small device, which is a result of its high magnetic field. We hope to leverage our experience working with C-Mod and new advancements in high-temperature superconducting magnets to develop smaller and cheaper reactors that could be designed and built quickly. -nc


I was fairly excited by the Lockheed Skunk Works compact fusion reactor that was in the news a couple of years ago.

Does the wider fusion research community have critiques of their approach? Insurmountable hurdles etc?

Are there any design concepts from high beta designs (going from Wikipedia, I recall none of this from Phys 102) that might be applicable to tokamak reactors?

Sanpaku

I think one of the main concerns from the fusion community is the lack of transparency from Lockheed over the design of their reactor and overall progress they've made. It's an interesting design, and researchers are generally open-minded to new ideas, but it's hard to evaluate critically or learn from their design without data.

Beta is a metric measuring how much pressure you can contain for a given magnetic field, and hence how energy-dense your plasma can be. If we can make it much cheaper to use superconducting magnets in tokamaks, then we wouldn't need higher beta than we do now to achieve viable fusion reactors.


Hi! I'm a senior in high school and I'm doing a project on nuclear fusion!

Why is plasma pressure important? Is it because it is more difficult to keep particles moving that fast at higher pressures? Do JET pulses at higher pressure yield more energy (supposedly because there is more fuel in the chamber)?

Also, what do you know about alternative ways of harnessing energy from fusion? I know that ITER uses the released neutrons to heat up the walls. I also have heard about something called a magnetohydrodynamic generator; what is this and how would it work? I understand that it would somehow use the helium produced to generate energy.

One last thing- how old are all of you and how did you get your current job? I want to become a scientific researcher and I'd like to know how one does that.

Thanks in advance for any answers!

The_Purplewalrus

Glad to hear you're interested in fusion!

So the main reason that plasma pressure is important is that plasma pressure is the main driver behind fusion reactions. Basically, in order to get fusion reactions to happen you have to smash deuterium and tritium nuclei together really hard. The nuclei are both positively charged, so you need a lot of energy to overcome the repulsion of like charges and get the reaction to occur. The way that we give the nuclei enough energy to get close and fuse is by generating a very large plasma pressure, which can smash the nuclei together with quite a bit of force. In the sun, this pressure is contained with gravity. Unfortunately, we're just a bit smaller than the sun (only like a factor of 1027, one octillion, times smaller), so instead of gravity we use strong magnetic fields to hold the plasma together. The stronger the magnetic field, the more pressure that you can contain.

JET pulses yield more energy for a few reasons. Primarily, JET is a much larger machine than C-Mod. We only have about 1 cubic meter of plasma, while JET has 100 cubic meters. This means that even with a much higher pressure, C-Mod doesn't generate nearly as many fusion reactions as JET.

You're exactly right about ITER. Most current designs for fusion reactors involve capturing the fast neutrons generated by the fusion reactions. This is typically done in either a solid wall or a liquid blanket that surrounds the plasma. The walls or blanket would then transfer energy to water, which would boil, spin a turbine, and generate electricity in the same way as a traditional power plant. There are, however, other possibilities for extracting the energy. One of which is the magnetohydrodynamic (MHD) generator. This would direct fast helium, which is generated by the fusion reactions, through a set of magnets in order to generate electricity. To some extent this works the same way as a normal generator, except in stead of mechanically moving a permanent magnet inside of a another electromagnet, you move charged particles through an electromagnet. This technology is definitely promising, but is much less developed than other methods of energy extraction. It could prove to be really cool, but we'll have to do a lot more research before we're sure how well it'll work.

As far as your last question, I'm a 3rd year PhD student here at MIT. I studied mechanical engineering and physics at my undergraduate university and am now in the Nuclear Science and Engineering department at MIT. The typical path to become a fusion researcher is to go through undergraduate in engineering or physics, and then to do a PhD in a similar field. So glad to hear you're interested in the field!

Hopefully these mostly answered your questions!

~ac


Among the alternative fusion approaches, which one do you consider the most promising?

Alfonso_Italy

By far the most promising non-tokamak approach to magnetic confinement is the stellarator because it probably avoids disruptions and the need for current drive. Stellarators have decent energy confinement though not (yet) as good as tokamaks. No other magnetic confinement concept has energy confinement within a factor of 10 as good as an equivalent scale tokamak. They don't confine. That's why the main line is tokamaks and stellarators. The Spherical Torus is a kind of tokamak, but with very low aspect ratio. In Europe ST stands for Spherical Tokamak. The ST has decent confinement too. Short of an EXTREMELY dramatic unexpected breakthrough, none of the approaches based upon compact torus configurations are going to prove even scientifically feasible, because their confinement and stability is inadequate. One can dream, but responsible research takes into account what we already know in the science of plasma fusion.

Using very high magnetic fields in tokamak or stellarator designs is I think the most promising way toward fusion energy.


Hi! As an educator of high school students, could you suggest ways that I could help my students understand better what Fusion energy is, why it is important that we keep researching in this area, and what the future of energy will look like (ideally)? I'm trying to construct a module on the need for continued scientific endeavor and I think that what you're doing is an excellent example of this.

ilikebigbuttsss

It's great that you want to teach your students about nuclear fusion, I didn't know much about it in high school! If you're near one of the large tokamaks in the US, you might be able to take your students on a tour of DIII-D, NSTX, or Alcator C-Mod. PPPL has great learning resources online, including a page that lets you remotely control a plasma. If you like building things, you could make something like that, or a fusor, and show how you can bend the plasma with magnets. MAST also has a cool 3D model of their tokamak online. And here are some videos I've saved:

Here are some answers to your questions to get you started. What is fusion energy? It has a lot of faces: tokamaks, stellarators, inertial confinement, private companies like Tri-alpha and General Fusion with outlandish ideas... Why is it attractive? Deuterium can be derived from seawater, tritium can be bred using Lithium, fusion reactions release a lot of energy (compare to fission using this chart going from 2->4 vs 235->140), radioactivity is low, accidents are no big deal, and nothing bad is released into the atmosphere.

At MIT we're interested in small tokamaks with high magnetic fields, which we think are a faster and cheaper path to commercial fusion power. -sean


MIT is my dream college and I'm interested on what you did to get there. I know MIT is very selective but what are some good rule of thumb things that will help.

Also in regards to plasma, what is it? I heard it was a new state of matter, but why is it important? Also do you know anything about super solids?

johnsmithinmyass

So I did my undergraduate at Princeton University and am now at MIT for graduate school. There are a bunch of factors that universities take into account when considering applications, and unfortunately I don't know too much about the details of the process. Interest in science and technology is definitely important for MIT, as well as strong academics and other activities. I think the best info might be at the MIT admissions site:

http://mitadmissions.org/

As far as the importance of plasma, there are a bunch of reasons that it's important. Plasma is often called the '4th state of matter.' When you heat up a solid (ice) enough it become a liquid (water). If you keep heating it up, it becomes a gas (steam). If you heat up a gas even more, then it becomes a plasma. There are actually plasmas all around us. The sun is one giant ball of plasma. Lightning, neon signs, fluorescent lightbulbs, and the northern lights (aurora borealis) are all plasmas as well.

Specifically we're interested in plasmas since we'd like to generate energy from nuclear fusion (like the sun). In order to get hydrogen atoms to fuse into helium and release energy (based on E=mc2), you need to heat the hydrogen up to very high temperatures. At these temperatures (100 million degrees!), pretty much anything is a plasma. Hopefully, if we learn enough about plasmas we can harness the energy of atoms and generate a new clean energy source!

I don't know too much about supersolids (quantum states of matter with superfluid properties) myself, but I know that other labs at MIT are studying them. It sounds like awesome research!

~ac


MIT is my dream college and I'm interested on what you did to get there. I know MIT is very selective but what are some good rule of thumb things that will help.

Also in regards to plasma, what is it? I heard it was a new state of matter, but why is it important? Also do you know anything about super solids?

johnsmithinmyass

I did my undergraduate here at MIT, so I can add some additional info about getting into MIT. The admissions site (http://mitadmissions.org/) is probably your best resource, but as a rule of thumb, MIT looks for smart students who are passionate about something. Getting good grades (especially in math in science) is important, and so is getting involved in extracurricular activities. However, don't do extracurricular activities just to pad a resume, do them out of interest or passion. If you like making things, MIT's maker portfolio is also a good chance to show off things that you've made. -nc


What would've likely been the result in the event of complete and total catastrophic failure across all systems and safeties during your stress test?

Luk3ling

That is the worst case scenario within our risk assessments, where the machine blows up. The building that these machines are in are designed to take the loads. Realistically speaking though a catastrophic failure would cause an 'implosion' rather than an 'explosion' since we run a vacuum for the machines. So the forces on the external structure (i.e. the building superstructure) is not that large.

~ak


Hi guys, I appreciate very much your work and I constantly follow the progress in the field of nuclear fusion. In the last few years there is a lot of entusiasm around the idea of "spherical tokamak" as faster and cheaper route to fusion energy. If this is true, why "ARC" and "SPARC" reactors, which you are proposing, are "traditional" tokamaks and not spherical ones?

Alfonso_Italy

Thanks for following things closely!

The spherical tokamak is a kissing cousin to the "standard aspect ratio" tokamak. It is basically squished. It has some interesting properties from a physics standpoint such as the highly turbulent and unstable plasmas and the current driven by pressures. However, for a reactor it is still significantly less demonstrated and their fast and cheaper aspects are unfounded. For example NSTX, the highest performance spherical tomakak has 1/10th the pressure and shorter confinement times compared to C-Mod despite being 14x larger and 4x more expensive. Their compact shape also proven more difficult to build and operate since the magnet has to fit in such a small central region and NSTX has had repeated problems in this regard. http://scitation.aip.org/content/aip/magazine/physicstoday/news/10.1063/PT.5.1092.

As such, we are still excited about the more traditional tokamaks but hope that the spherical tokamak will make progress as well. - BM


How close are we to achieving fusion power plants. Last time I checked on fusion I was told we haven't been able to complete proper fusion. Are we still at this stage?

MrDobert

At the moment, ITER, a tokamak being constructed in France, is planned to be an only experiment in the next 20 years, which will demonstrate that fusion power production can exceed the input power. IF ITER can successfully demonstrate this, the next step will be a prototype fusion reactor.

MIT is currently proposing an ARC reactor that uses new superconducting magnets, instead of conventional copper coils or low temperature superconductors, which can lead to a smaller, cheaper fusion power reactor. You can check out this PSFC webpage that illustrates the concept of the ARC reactor: https://www.psfc.mit.edu/news/multimedia/2015/arc-the-smaller-faster-cheaper-fusion-reactor

Meanwhile the fusion community is actively researching how to properly exhaust the heat from plasma and how to properly heat the plasma by innovating the tokamak design. For example, another MIT proposed experiment, ADX, is trying to address these two critical issues: http://www.psfc.mit.edu/research/topics/adx-advanced-divertor-experiment

So the coming years might be an exciting moment for fusion power plant development. (sb)


How close are we to achieving fusion power plants. Last time I checked on fusion I was told we haven't been able to complete proper fusion. Are we still at this stage?

MrDobert

Your question really has two parts. Tokamak experiments create a huge number of fusion reactions every day. Our FAQ on our the record setting experiments (http://www.psfc.mit.edu/news/2016/pressure-record-faq#experiment) includes a plot of the fusion reactions we created, in trillions per second! So we definitely have completed proper fusion. However, the challenge now is to create more energy from fusion than it takes to run the experiment in the first place. Doing this will move us close to creating actual fusion power plants. ITER, a tokamak that's currently being built, is projected to meet this goal. -lt


In one of the articles about this it said that a target number for a commercially viable fusion reactor is 50 million degrees. Why is such an incredibly high temperature needed? I always thought the biggest problem with fusion is that it can't run for long, not that it can't get hot enough, wouldn't a lower more sustainable temperature work better?

Aetrion

The rate of fusion reactions is an extremely strong function of energy, which means a strong function of temperature. At room temperature there is less than one fusion reaction in the lifetime of the universe. We need to get that rate up to greater than a trillion reactions per second. Raising the temperature does that. Actually an efficient fusion reactor probably will have a temperature more like 100 Million degrees C. We have demonstrated those temperatures in experiments for the past 20 years or so. We can do that. What we have to do at the same time, is to keep the energy in the confinement device long enough for the reactions to take place. That means we need good confinement AS WELL as high temperature. ~ih


Any luck with outside funding? Really bummed funding has been pulled from this in favor of that mess in the EU :(

relkin43

It should be noted that the funding for Alcator C-Mod came from the tax-payers via the Department of Energy and the payback is scientific knowledge about fusion plasmas for everybody. We are very thankful for this! It is sad that there is limited money for fusion, but priorities are out of our control.

Fusion general has attracted significant funding from venture capital and such over the last few years, this is good for the whole field. However, to date, our group has not been actively pursuing that except for small special high-risk projects. That might change in the future if there is interest.

Hopefully there continues to be excitement and high-quality science and technical results around fusion. In a round-about-but-real way this contributes to more funding across the board. A rising tide raises all boats. - BM


Hi!

I'm a 22 undergrad, so I suppose I'll be able to ask this question to someone at one point or another regardless.

Anyway, the timing of the record-breaking run and Alcator C-Mod's last day of operation lends me to think that the run was done as a "If this breaks the reactor, ¯_(ツ)_/¯" type thing.

Is that the case?

jkium

Have I TA-ed you before -.-

On a more serious note the idea to do this record-breaking run on the last day of operation was more to end Alcator C-Mod's long and successful experimental life on a high note. Something for everybody to celebrate about. In addition it was to further validate the current MIT PSFC mindset that the high field approach to fusion is the most viable one. Setting the new record is something we can use to help support our argument.

~ak


Do we have any guarantee that we can use fusion to provide more energy out than in?

tomholder

Fortunately we've already got a working example of a fusion reactor: the sun! The sun (and all stars) have been producing net energy gain from fusion for billions of years. The hard part is trying to get that same process working here on earth. We've actually gotten pretty close though. Two machines, JET (in the UK) and TFTR (at Princeton), both produced about 60% of the energy that was put in (16 MW of fusion power from 24 MW of heating power in the case of JET). While this isn't useful for a power plant, it's getting close to net energy gain. The challenge now is to cover that remaining gap and then make sure that we can continue to produce steady power for long periods of time. ~ac


Have you been deemed a pirate?

Tantric75

Quick poll of the control room reveals none licensed to be a pirate :( -nc


How will power be produced by the fusion reactors? Will there be circulated coolant to a steam generator similar to how a PWR works? For what its worth, I'm submitting this question from a power plant undergoing refueling, presently.

GMark73

The fusion products transmit their energy to the plasma and to a surrounding blanket where more fuel is bred. The heat is taken up by some sort of coolant, for which there are different options. That heat is then to be used in a generator, probably not terribly different from the steam generators used in current power plants, to generate electricity. There are ideas for other routes to electricity generation, but this is the best understood route.


Hello there. Was is scary taking it to the limit, when your device can get hotter than the sun?

Sexualwhore

In fact, it is generally exciting to see hot plasmas on C-Mod. The machine operates at its engineering limit, and engineers and technicians take a good care of the machine. In the last 20 years, the Alcator C-Mod tokamak has been routinely operated at temperature hotter than at the center of sun (15 Million degrees). With external radio frequency heating, C-Mod achieved plasma temperature over 70 million degrees. ITER will operate at 150 million degrees with an aim to make net power out of fusion reactions! ~sb


Now that it has reached the end of it's lifetime, what is being done with the components of C-Mod?

Hopeles5

Any instruments or equipment that has been borrowed from other universities or national laboratories that we collaborate with are being dismantled and returned.

As for the main machine and the rest of it, the plan is to place it in a 'safe shutdown' mode by feeling the chamber with an inert gas and keeping the machine is standby. The idea being that if the DoE ever decided to provide funding for the operations of the experiment we would be able to start it back up right away. ~ak


Do any of you make plasma sculptures as well? Or experiment with plasma in other ways, outside of work?

Boroflow

Dang, plasma sculptures are cool! We actually had never heard of them, but we certainly want to make some now! I actually built my own plasma speaker a while back. It was really fun. All you had to do was plug in your phone, and little sparks would play your music for you! - at


Now that C-mod will be shutting down, what is the next envisioned tokamak device that will be constructed at MIT, and what is its timeline?

theSquatSasquatch

It's still together at the moment. We have ideas for using some of the components in the longer term for a new experiment called ADX. You can see some of these at http://www.psfc.mit.edu/research/topics/adx-advanced-divertor-experiment


One of the big goals of ITER is produce ~10X return on input power. What recent advances (scientific/engineering) have led to such claims for fusion devices, given that even achieving energy parity has been difficult (to my admittedly limited knowledge)?

brownsugar370

Great question :) The projected performance of ITER isn't actually based on recent science advances, since a lot of its design work occurred in the 1990s. Rather, ITER takes established plasma physics and uses it in a machine that is far larger than any current machines and that has a stronger magnetic field than many (but not all) current machines. Creating this sort of machine has required very careful and advanced engineering, and that's a large part of what has taken ITER so long to be constructed. This engineering process has led to many advances, and so will the science learned from ITER's eventual operation.

Here at MIT, we're also interested in thinking about how more recent scientific and technological advances (high temperature superconductors, I-Mode, advanced divertors, to name a few) could be used to achieve some of ITER's goals in other ways as well. Check out some of our ideas about this here: http://www.sciencedirect.com/science/article/pii/S0920379615302337 -lt


One of the big goals of ITER is produce ~10X return on input power. What recent advances (scientific/engineering) have led to such claims for fusion devices, given that even achieving energy parity has been difficult (to my admittedly limited knowledge)?

brownsugar370

You are correct, ITER's primary goal is to achieve 10x the return power than the input power. Also, no fusion device has achieved energy parity yet.

The scientific consensus is that ITER will get to Q=10. It does this mostly through making the device very large and thus the heat leaks out slower. Much larger than the other devices.

ITER also relies on the accumulated knowledge from all the tokamaks across the world. Some key advances for ITER include how to control the plasma, how to stabilize parts of the plasma that can cause heat to leak and how to handle the heat flowing to the exhaust port. Each of these is important for a reactor (and ITER) but mostly ITER does it by being so big.
-BM


Slightly off-topic: Consider I want to write a short blog-article about your results, who do I need to ask for the permission to use a few of the nice pictures you have on your website?

DerPlasma

You may contact to Paul, the PSFC communications and outreach coordinator. https://www.psfc.mit.edu/people/administration/paul-rivenberg


Is there a theoretical model for the efficiency of an idealized reactor? I.E. is there a theoretical maximum for the power output per unit input for a reactor, and if so is this maximum gain large enough to offset the costs of construction and maintenance of a reactor and make fusion eventually feasible, even just on paper?

azd685

This is a very interesting question.

Unlike, say a gasoline engine, we don't have a first-principles understanding at the level required to calculate a realistic theoretical efficiency for a fusion reactor. What we need is the ability to predict how fast a plasma leaks heat and thus cools.

The closest we have now is something called "neo-classical" theory which predicts that fusion reactors should be incredibly powerful because they should not leak heat and should instead confine their plasmas very well to the point that fusion power out/heating power in approaches infinity, we call this ignition. However, all experiments see that heat does leak out.

The problem is turbulence. (but in like 6 dimensions!) which makes heat leak 10-100x faster: https://www.youtube.com/watch?v=RLI6QW2x4Lg. Many scientists are studying this using supercomputers so we can minimize the heat leakage and over time ways are devised to limit it by spinning plasmas, or changing the internal profiles or launching waves.

To make things worse, typically most plasmas see increased turbulence as they get hotter and denser. So far tokamaks have had the least amount of heat leakage experimentally.

Though we always want better confinement, even with the existing level of heat leakage a large enough reactor (like ITER, which is battleship big) or a smallish tokamak at high enough field can approach ignition.

Various studies have been conducted about the economics of fusion reactors and indicate they could be competitive, but we don't know enough about their costs since we only have built one-off experiments thus far.


In particular, what do you think about the work done at Tri Alpha Energy and, in general, about the effort to obtain fusion using proton and boron 11 as fuel? Consider that Tri Alpha has raised over 500 million dollars which is several time your annual budget.

Alfonso_Italy

Aneutronic proton-boron11 fusion, while attractive, is much harder to achieve than DT fusion, in terms of the temperatures required and the level of plasma confinement that must be achieved. The work being done at Tri Alpha Energy is definitely interesting from a physics perspective and they have made a lot of progress since their initial prototypes, but it is hard to compare their progress to tokamaks today. The performance of tokamaks has improved dramatically since they were invented (they even followed a trend that looked like Moore's law for a time: https://static.iter.org/all/newsline_1_120/img/53/moores.jpg._1024.jpg), but progress has stalled in recent years. Scaling up the Tri Alpha Energy design to a reactor scale design would be a major achievement. -nc


A hearty congrats to you guys!

Here's a question that always comes up with a few buddies of mine when fusion comes up in the news;

How much closer are we to Mr. Fusion from BTTF now with the new Tokamak designs?

MammothCat1

I've got one in my basement, it works great. A 12 pack of beer keeps my house powered for a full year! You've got to work in the field to get your hands on one of those models.

On a side note, a long way off. At least at the moment, it is not clear how your make something that small or fuel it with beer and banana skins.

~ak


In the past I was fascinated by the results of Levitaded Dipole Experiment (LDX). Which is the current status of that experiment and how far is this simple and elegant solution from a burning reactor?

Alfonso_Italy

LDX is certainly a fascinating experiment, helping us learn more about plasmas in our solar system like the solar wind. The device still exists at MIT, but has not been in operation for several years. Thus, it still has a long way to go before it would be viable as a fusion reactor. -at


Please can you give a cost and time estimate for the commercial operation of fusion power as the tokamak has been in development for around 60 years.

funkenstein_1

The tokamak has been around for a while now and many people question the feasibility of the design considering how long we have been working on it. But to-date it is still the most promising design and the closest to achieving reactor conditions. The thing is that we've learnt a lot over the years now about plasma and how to confine them using magnetic fields and that understanding is definitely applicable to any other magnetic field based fusion reactor design. But we've also learnt a lot of things about tokamaks and we think we are close.

So to build a reactor that would actually put electricity on the grid. You would need to go through the standard design cycle. You brainstorm -> design -> build -> test -> fail -> learn.

However the problem at the moment is that each of these iterations is taking too long and costing too much money. We need to find a way to streamline this learning process. To this end, the scientists here at the PSFC have decided that we've got to make smaller reactors that can be built faster and cheaper. We've got a REALLY good idea of how well they will perform and if everything goes smoothly we're not actually that far away. Even at these smaller cheaper machines though, each of them will be in your 100 million - 1 billion dollar mark and will take 3-5 years each to build.

~ak


Additional Assets

License

This article and its reviews are distributed under the terms of the Creative Commons Attribution 4.0 International License, which permits unrestricted use, distribution, and redistribution in any medium, provided that the original author and source are credited.