UKH

Yes I know the news about this is always that it's just round the corner. I just read this article and not for the 1st time I wondered about something. 

Presumably these experiments are carried out inside a chamber of some sort, made of metal, and maybe some plastic? If they reach temperatures of 180 million degrees, why doesn't the whole melt, or even just vaporize? 

The hot stuff in a tokamak, spheromak or stellarator reactor is a plasma, which consists of charged particles (electrons, and atoms missing some electrons).  The plasma is inside a vacuum vessel which is made of metal, but whilst the vacuum vessel keeps air out, it does not contain the plasma.  Instead, the plasma is contained inside the vessel away from it's walls by a magnetic field.  It can be so contained because it's charged.

If for some reason the containment fails (a runaway turbulent breakdown of the plasma starting with an instability known as an ELM perhaps) the plasma can hit the physical wall of the vacuum vessel.  The temperature isn't really a problem; the vacuum chamber on ITER weights something like 5,200,000 kg.  If I've got my late-night maths right, the quantity of plasma contained inside the magnetic field inside the vessel is much less than 1 kg.  

Heat itself isn't a problem, rather how much heat the hot thing can give, and how much heat the cold thing can absorb before melting.  Have you ever got a blob of molten solder on your hand?  It's at 300C or so, but barely even leaves a burn - your big hand can absorb more heat than it can give.  Same with the plasma in the vessel.

However, if I recall someone's pub maths correctly, the entire reactor vessel might jump half a meter up off its mounts which is going to cause some problems.

The one bit of this kind of reactor that gets hits by a lot of plasma is called the "diverter" region; it's designed to collect all the bits of plasma that escape the magnetic field and it's design is one of the really interesting/challenging parts of this kind of fusion machine.  The "MAST Upgrade" experiment down at Cullham is due to test a new kind of diverter called "Super-X".  A little bit of Wintertree magic might even be in one of the diagnostics looking at the diverter region there...

Edit: I really like how each post is getting more succinctt.  That's a high bar set for whoever follows gowain.

In Tokamaks, at least, which are arguably the most developed of the current technologies, magnetic confinement is used to keep the plasma away from the vessel walls. In essence, the reactor vessel is a huge vacuum chamber, and extremely strong magnetic fields are used to hold the plasma in the space away from the walls. The intense neutron radiation produced still gives the walls an absolute battering though, transmuting the various elements comprising the surface.

edit - wintertree beat me to it.

At the high temperatures required for fusion, atoms disintegrate into their constituent particles, forming that is called a plasma. The particles within this are charged and so can be confined using magnetic fields, keeping them away from mortal coils of metal, let alone any plastic. In principle. It is a tricky business, given the plasma densities required for fusion to occur.

Thanks for that reply, explained it really well.

As for solder.... A long time ago when I was an 18 yr old electrical apprentice I actually flicked a tiny blob of solder into my eye. Panicking and freaking out, thinking I'd be be blinded I went to the 1st aid room. Turned out it had solidified on contact with my eyeball. Some eye drops and a wipe and off it came.

As soon as the plasma touches the wall it cools and stops. The "interruptions" are like bulges in the plasma squeezing between the magnetic fields, and they do make the vacuum vessel jump. The one at JET is around 800tonnes and is suspended on bungy cord.

How do I know, I used to work at JET and seen first hand the melted inconel of the vessel wall where the plasma hit, and amongst other things I changed out the bungy cords that hold up the vessel. The shock absorbers used blutac  as a non Newtonian fluid.

ALS played around with Mast on the original build for a while

A couple of the major problems are maintaining a vacuum and cleaning out the plasma as the fuel is used. 

Whilst we're on the subject, does the structure of a fusion reactor become activated from all the neutrons whizzing about, in the same way a fission reactor does? We are always told that fusion is clean energy. 5000 tonnes of radioactive steel sounds like a bit of a decommissioning problem.

Not in the same way, ie lower activity. The later deuterium experiments at JET have made it more radioactive, but nothing like a fission reactor. Plus there's no high dose fuel rods left over. 

Not in the same way, ie lower activity. The later deuterium experiments at JET have made it more radioactive, but nothing like a fission reactor. Plus there's no high dose fuel rods left over.

Also decommissiong is planned from the design stage

I went to talk by someone from JET about sixteen or so years ago. He mentioned that selecting an appropriate metal from which to construct the containment vessel was a serious materials engineering challenge: for a (hoped-for) commercial reactor, it has to be able to withstand the neutron flux for long periods of continual use (say, at least a year) without significantly degrading.

OK interesting, thanks. 

As artif says, the neutron induced radioactivity is much lower level than for a fission reactor.  On the other hand, the mass of material that is activated is massively higher than for a fission plant.  It can all be planned for.

The neutron flux at the pressure vessel wall is one of the major problems yet to be fully solved.  Whatever lines the pressure vessel ("first wall") has to pass the heat made by the fusion reactions through to the heat exchangers behind it for conversion to power,) and has to withstand that radiated heat, stray bits of plasma and the insane neutron flux without spalling atoms in to the vacuum, which would salt the plasma, and without undergoing so much atomic transmutation or microscopic physical structural change that it can no longer serve its purpose.  Solving this keeps a lot of people busy, the materials physics is one of the key research areas for fusion power.

There are fusion reactions that emit power and don't produce neutrons, e.g. a proton and boron-11. Realising these are barely even flights of fancy for now, but if they do eventually work the nice thing is that something like 99% of the fusion power comes out as alpha particles (fully ionised helium nuclei), which can be caught and electrically neutralised, making the extraction of power a 95%+ efficient with the reactor acting like a 2 MV DC nuclear battery.  Which is a bit less embarrassing than spending all these decades and billions to make something so difficult and then hooking it up to steam turbogenerators with all the further losses they incur...

That blu-tak works overtime...  

Graph from the paper below.  1.7 MN transients via 8 suspension points.  That's the whole vessel (as then was, mass given as 240e3 Kg) jumping at about 1g.

https://scipub.euro-fusion.org/wp-content/uploads/2014/11/JETC94062.pdf

Works with superglue to.

Dont try it at home though kids!

That's the old style restraints they were changed after that, I first went there in '95.

There are fusion reactions that emit power and don't produce neutrons, e.g. a proton and boron-11. Realising these are barely even flights of fancy for now, but if they do eventually work the nice thing is that something like 99% of the fusion power comes out as alpha particles (fully ionised helium nuclei), which can be caught and electrically neutralised, making the extraction of power a 95%+ efficient with the reactor acting like a 2 MV DC nuclear battery.  Which is a bit less embarrassing than spending all these decades and billions to make something so difficult and then hooking it up to steam turbogenerators with all the further losses they incur...

This has always bothered me about nuclear power, we're still just making heat to turn a steam turbine (approx 35% efficient at best) Brunel would recognise most of the systems used in power generation today. There has to be a way of using the nuclear energy to directly generate power (if not, why aren't we investing billions into it). Currently were just making better kettles.

Quite a few companies are working on different approaches to achieving fusion that don’t rely on the magnetic confinement of a plasma.

First Light Fusion, based near Oxford, have a novel pulsed reactor design in which neutron energy is captured by a liquid lithium waterfall. Sounds a bit mad but it does try to address some major problems associated with tokamak designs, not least the method of breeding tritium as well as the longevity of the reactor vessel wall.

https://firstlightfusion.com/technology/power-plant

This recent video is also worth a watch to see some of the alternative approaches.

Why Nuclear Fusion is Closer Than You Think
youtube.com/watch?v=yNP8by6V3RA&

Not half as mad as General Fusion’s giant swirling ball of molten lead being hit by a bunch of steam driven pistons.

Oxford is really hotting up in the fusion world (pun intended).  

A lot of private money has gone in to a half dozen or so firms (not doing ITER style giant tokamak devices) in the last couple of years.  Make or break time for the leading candidates I think, the private money won’t keep pouring in at these levels without results.

Ah, that explains why the house jumps every time the Athena poster falls off.

Yep, totally bonkers.

Tokamak Energy is another Oxford based company worth keeping an eye on.

https://www.tokamakenergy.co.uk/technology/

Our path to demonstrate grid ready fusion power in the early 2030s
youtube.com/watch?v=6p8CH2ry9x4&

Moving closer to commercial fusion
youtube.com/watch?v=g2J30AeIuHc&

Some deep American pockets out there.

https://www.forbes.com/sites/christopherhelman/2022/01/02/fueled-by-billion...

If it's 'just' a few billion, those pockets don't have to be too deep. (I am just comparing that cost with, for example, the amount that was spent buying Twitter recently.) And of course, if someone does crack this, the payoff will be enormous.

It would be interesting to know how much is being (and has already been) spent overall on fusion research. Probably not as much as one might expect.

And in China.

youtube.com/watch?v=f9-NGW-hJAM&

Another good video.

Why Private Billions Are Flowing Into Fusion

https://m.youtube.com/watch?v=Dp6W7g9no0w&pp=QAFIAQ%3D%3D

But I think it will take a lot more than a few billion to demonstrate controlled fusion and then build the first commercial reactor. So many problems still to be solved.

You can demonstrate controlled fusion for about £15k if you're not scared of moderately high voltages and vacuum systems.  Kids do it in their basements in America for high school science fairs.  The problem is demonstrating a design that can exceed electrical break-even with controlled fusion...

ITER/DEMO suffer immensely from needing to have a very large plasma volume to hit break even, and from the downstream consequence of having a very low area/volume ratio meaning very high fluxes out of the plasma and through the vessel wall.   

A lot of the really hard problems downstream from that go away if you drop down to a device with a fusion region a few meters in size and an output power of ~300 GW.   Conventional tokamak devices are naff all use at that scale, but if you look at the commercially funded devices, they're much more in the 100-300 GW, "fusion in a volume a few meters wide" class.  

Edit: all these devices have their own problems, different for each device, and I’ve been really happy to see private money coming in to trying to solve them given the near-global stitch up of taxpayer funding by ITER.  It’s far too early in the day to back a single horse.

The issue of scale isn't just important for getting the materials science etc to work, something the size of the proposed DEMO is just too big and expensive for anyone to build them in earnest; look at the difficulties financing the  - much cheaper - new fission plants for the UK.   The first building to be completed at the ITER site was the project management office . It wasn't small...

Be interesting to see what news the next few years will bring.  Be embarrassing if someone cracks it before ITER is operational, it doesn't even have the tunnels to grow mushrooms in like the last abandoned science mega-project (Superconducting Super Collider in Texas). My hunch is on Tokamak Energy because I think they have more pipe smoking boffins on the staff than the others. Then there's the recent UK government funding for a post Mast-U spheromak.  Going to be hard for them to hire people given all the other activity?