# Why not particle accelerators for fusion?

1. Sep 25, 2005

### RayB

In the 1997 5th edition of the Halliday Resnick and Walker fund of Physics this question is asked about colliding two beams of deuterons directly toward each other as well as perhaps colliding deuterons with a steel target. p1116.
Why is this method not presently used.
Doesn't the ITER do something like this?

2. Sep 25, 2005

### ZapperZ

Staff Emeritus
The "luminosity" involved in such a collision produces interaction rate VERY low when compared to the standard fusion from plasma. If one tries to compensate this by injecting more ions, you will run into space-charge problems. Furthermore, you already will need not only energy to strip off the electrons off the deuterons, you also need a lot of RF power to accelerate something that heavy. See how much power a klystron requires to power all thos accelerating structures.

I see this as going the other way in trying to get more energy out than one put in.

Zz.

3. Sep 25, 2005

### Staff: Mentor

Neutral beam injection has been used for heating plasmas (in addition to ohmic and RF heating), and in fact, it appears that hot ion plasmas, i.e. where the ion temperature exceeds the electron temperature, seems to be the direction in which fusion reactors, e.g. ITER Tokamak, are going.

4. Sep 25, 2005

### RayB

Yes maybe the Iter was initiated after the textbook. Also the RO papers seem to indicate that more energy in terms of the number of emitted particles and their range was given off than used in producing the acceleration of protons. Is anyone working on this or interested in working on this that you know of?

5. Sep 26, 2005

### Geoff St. Germaine

I believe that in ITER the ion temperature will be comparable to the electron temperature, but still lower (8 keV versus 9 keV, or something). This is way better than either of the small tokamaks I've worked on where typically $$T_e > 4 T_i$$.

Last edited: Sep 26, 2005
6. Sep 28, 2005

### RayB

1)How does the radiation of colliding particles which I thought were 3MeV near the D2 fusion or He binding energy in the Tokamak cyclatrons get conveyed to the water that produces the steam for the generator?
2)What is the damage and radioactivity produced in the target of an acclerator? Does the underlying steel plate eventually melt or disintigrate.
3)

7. Sep 28, 2005

### Staff: Mentor

1) Neutrons (assuming D-T fusion) would carry energy to the chamber (torus) containing the plasma, and thus heat the wall and structure outside the torus. Also, gamma radiation will heat the chamber wall, which much be cooled. The chamber wall may be cooled directly by water or perhaps liquid metal or a pressurized gas. The water cold be pressurized and then sent to a turbine.

2) Neutron and gamma radiation induces defects in the structural materials, and over time the materials will become brittle, unless the material is annealed - heated to the a temperature where lattice defects (dislocations) migrate to grain boundaries or coalesce into new grain boundaries - or recombine - as in the case of holes and interstitial atoms.

3) ???

8. Sep 28, 2005

### ZapperZ

Staff Emeritus
I'm VERY confused with this thread.... <surprise!>

We have "neutrons" running around in a "particle accelerator"??!! Just how exactly did we accelerate and control the motion of these neutrons in a particle accelerator?

Or are we colliding ionized D and T? If we are, why are we using a particle accelerator? We don't want it to be THAT high in collision energy, do we? The cross-section for causing fusion does NOT require a collision as high as that achieved in the Tevatron or RHIC. So then why are we talking about accelerators in the first place?

Zz.

9. Sep 28, 2005

### Staff: Mentor

The OP did describe colliding beams of d's, but then asked "Doesn't the ITER do something like this?"

Well ITER is a Tokamak. And Tokamaks use neutral beam heating. They do not use accelerators, because accelearators are generally inefficient, the beam current densities are low, and the energy losses from radiation would be considerable.

I think the accelerators in HR were provably GeV or TeV range, and that's way too high when one wants d in the keV range.

I haven't seen HR or HRW, so I can't really make an accurate statement on the initial problem statement.

As for neutrons - D + D -> T + p or He3 + n about 50/50 probability.

in D + T, the yield is He4 + n, with the n taking 14.1 MeV of the 17.6 MeV produced.

But these might be of more use -

http://www.iter.org/techpara.htm, where one will find

Volume-averaged ion temperature <Ti> = 8.1 keV
Volume-averaged electron temperature <Te> = 8.9 keV

and likely D-T will be the initial fuel.

http://www.iter.org/Conditions.htm#product

Last edited: Sep 28, 2005
10. Sep 29, 2005

### ZapperZ

Staff Emeritus
Well, that's basically what I wrote in my first response here. So the whole thread is rather moot since there hasn't been any justification yet on why an accelerator is needed.

Zz.

11. Sep 29, 2005

### Staff: Mentor

Subsequent responses are either an elaboration (perhaps at the risk of being pedantic) or response to other comments or questions.

ZZ, you make a good point about the space charge limitation. The plasma has to be neutral, so if one uses an accelerator for the + particles, somewhere one must introduce - electrons to balance the + charge.

ITER is a tokamak, which is a completely different system than colliding accelerators.

Last edited: Sep 29, 2005
12. Sep 29, 2005

### ZapperZ

Staff Emeritus
If they wish to do that, I don't want to be anywhere NEAR the thing. :)

The acceleration scheme will be humoungously difficult. You are using the same accelerating structure to accelerate the light as hell electrons, and the heavy as hell ions. I don't even want to think about how you're going to set the magnets to make sure the space-charge and emittance of each type of particle beam are minimize. You also have each type of particles going in the opposite directions, with the ions having a much longer time to get up to speed. So you basically have separated those two in an accelerator.

If using an accelerator can produce efficient fusion, we would have done it by now, and I would have already written a research proposal to do this! :)

Zz.

13. Sep 29, 2005

### Staff: Mentor

Zz, I whole-heartedly agree with you. However, those not entirely familiar with the details of accelerators and tokamaks may not fully understand the physics and technical issues behind either method.

A typical particle accelerator would be overkill for fusion - like hitting a nail with a 5 kg sledgehammer (or actually bigger) - when a simple carpenter's hammer would suffice.

And you again raise some significant technical complications that simply make the use of typical particle accelerators for fusion impractical.

On the other hand, neutral beam injectors are indeed simple, low-energy particle accelerators - but certainly not in the same class as a synchrotron.

14. Sep 29, 2005

### NateTG

That's actually very close to what a fusor does. (Fusors are a simple type of fusion reactor that is sometimes used as a neutron source.)

It's actually not all that difficult to get fusion to take place, but very difficult to get more useful energy from the fusion than you use in order to get the fusion to occur in the first place. It's this second problem that makes the use of particle-accelerators for fusion power impractical.

A big part of the problem is that people don't seem to adequately distinguish between 'fusion reactor' (relatively easy) and 'fusion power plant' (very hard).

15. Sep 29, 2005

### ZapperZ

Staff Emeritus
Humm.. are neutral beam injectors really "particle accelerators"?

The facility next to my building is the Intense Pulsed Neutron Source (IPNS). While they do have an accelerator, that is only used to generate neutrons. After the neutrons are generated, you can't control much in terms of its momentum and energy. So these neutrons are neither accelerated, nor manipulated in terms of its direction. Both of these are necessary, by definition, in an accelerator.

Zz.

16. Sep 29, 2005

### Staff: Mentor

Hmmm . . . Yes.

If one considers that a neutral beam injector (NBI) has an ion source, an accelerating stage, then a neutralization stage - then, yes, an NBI is a particle accelerator, as would be a Van der Graff or Cockcroft-Walton accelerator.

Of course, I am speaking as much generically as with respect to physics involved.

R. Koch, "PLASMA HEATING BY NEUTRAL BEAM INJECTION", TRANSACTIONS OF FUSION SCIENCE AND TECHNOLOGY, VOL. 45, MAR. 2004, pp. 183-192.

A nice overview of neutral beam injection. Of course, the energies are relatively low - 50-200 keV, and it's a one shot deal (one stage electrostatic accelerator), as opposed to repeated kicks in a cyclotron or synchrotron, or other type of multi-stage accelerator.

17. Oct 2, 2005

### RayB

18. Oct 3, 2005

### Staff: Mentor

There is not such thing as inexpensive when it comes to 'nuclear-grade' components. That's because of the 'mandatory' (see 10CFR50, App. B) quality control and quality assurance requirements.

One does not simply buy an off-shelf part from just any supplier. The supplier must be qualified by the purchaser or the company which supplies the pump must be qualified by the ultimate user, i.e. licensee which owns/operates the nuclear reactor.

19. Oct 3, 2005

### RayB

20. Oct 3, 2005

### Staff: Mentor

1) in the neutralization stage, the hydrogen ions simply collect the electrons.

2) I imagine the D or T is simply feed into the ion source where it is ionized. Remember the NBI is connected directly to the Tokamak chamber which operates with a high vacuum.