Coating of electrodes

[Henning]

I’m just putting some threads together from talk-polywell.org that might help us to determine the best candidates for electrode materials.

They talk about x-ray reflection with diamonds: http://talk-polywell.org/bb/viewtopic.php?t=3271

In reference to: Bragg reflectivity of X-rays: At the limit of the possible

The diamond crystal, nearly defect-free in a volume of 5 x 5 x 1 mm3 , reflects more than 99% of hard x-ray photons backwards in Bragg diffraction, with a remarkably small variation in magnitude across the sample.

The reflecting x-rays might be too weak (ie. a DPF produces harder x-rays), or the band-width of x-ray wave-length might be to narrow, though.

As diamond is an insulator (except it is shined upon with UV light), it’s not a good candidate for an electrode. Maybe coat it again with beryllium?

There’s also a lengthy discussion of superconductors at room temperature by a diamond specialist: http://talk-polywell.org/bb/viewtopic.php?t=2137

If it all holds true, then electrode life may be prolonged.

[ikanreed]

How do you feel about a pie-in-the-sky response? It’s sounding like graphene has a lot of the properties you’re wanting. Absorbtion of high frequency radiation, excellent electrical conductivity, resistant to chemical changes like rusting, and durable. Too bad mass production of the stuff hasn’t started anywhere.

[jamesr]

I doubt any kind of coating would be better than just solid beryllium electrode. The interface between the bulk and the surface would always be a weakness. I suspect by alloying the beryllium with other metals you may be able to improve some properties (at the expense of others) in order to optimise the electrical, physical and x-ray absoption characteristics.

Such as a beryllium-copper alloy with ~60% Cu by weight (ie mix of delta-Be2Cu and beta-Be phase grains) with other minor (<<1%) components such as Indium or Manganese. Then fine tweak the grain structure between surface and bulk by heat treating, giving a smooth transition from bulk to surface.

[solrey]

What about boron doped diamond? The electrical properties can be fine tuned by adjusting the amount of boron doping used. Could a doped diamond CVD thin/thick film be applied to a graphite or niobium core for DPF electrodes? Boron doping results in a p-semiconductor while doping with something like phosphorous or nitrogen produces a n-semiconductor. Would those properties be useful in the FF-1 electrode configuration?

http://122.249.91.209/myukk/free_journal/Download.php?fn=NDFCT487_full.pdf&ei=7wLETp-6POeciAKK0ZWADA&usg=AFQjCNFnFzBfh7YXkvdkT0YxMWtpdLUzjA&cad=rja

Electrically conductive chemical vapor deposited (CVD) diamond has shown almost zero electrode wear, even at short pulse
durations, which is extremely important for maintaining the electrode shape throughout the EDM process.

http://www.scientific.net/KEM.257-258.535

http://www.electrochemsci.org/papers/vol2/2050355.pdf

[jamesr]

Interesting paper but for EDM they have a peak current of 6A at 90V. Not really in the same ballpark as 2MA at 45kV.

[asymmetric_implosion]

Why is everyone hung up on Be and C-based materials? Be is a mess of a material. It is toxic and can produce neutrons when hit by hard x-rays. The cost is a real problem too. Non diamond, C-based materials are susceptible to hydrogen etching and they generally have a poor electrical and thermal conductivity compared to metals. Based upon published studies about 20% of the bank power ends up on the anode when you have a highly electrically conductive metal and must be removed as heat. For a high repetition rate application like power generation, a modest resistivity material like graphite will only increase the heat load. As the electrodes get hot they tend to emit impurities or form other materials such as BC4 (insulator) in the presence of boron.

Studies were conducted nearly twenty years ago on electrode materials for plasma focus devices and the “best” materials were operated at up to 80 Hz (80 times a second) for 5 million shots. The winner was Molybdenum. It is a higher Z material which increases the x-ray dose a bit but the x-ray can generally be shielded with less than 0.5″ of lead or tungsten. The lead or tungsten shield is far less costly than a Be electrode.

It’s great folks are thinking outside the box from the typical copper electrode, but this was thoroughly studied in past experiments considering many of the key factors for fusion yield. Why reinvent the wheel when the wheel is well studied and it works?

[jamesr]

I agree, if Be can be avoided that would be great.

However, for a device hoping for good fusion yields you need to be very careful that no high-Z ions can be eroded off and pollute the hottest part of the plasma as they will cause rapid cooling. But more importantly the pinch X-ray flux will be orders of magnitude higher than any ‘low power’ studies done before, and we need to be able to recover the X-rays in the onion. Which means the anode must be as transparent to them as possible, to avoid shadowing a large portion of the collecting area.

Having said that I wasn’t aware Molybdenum had been studied before – do you know the source paper(s)?

[asymmetric_implosion]

jamesr wrote: I agree, if Be can be avoided that would be great.

However, for a device hoping for good fusion yields you need to be very careful that no high-Z ions can be eroded off and pollute the hottest part of the plasma as they will cause rapid cooling. But more importantly the pinch X-ray flux will be orders of magnitude higher than any ‘low power’ studies done before, and we need to be able to recover the X-rays in the onion. Which means the anode must be as transparent to them as possible, to avoid shadowing a large portion of the collecting area.

Having said that I wasn’t aware Molybdenum had been studied before – do you know the source paper(s)?

General materials survey

R. K. ROUT, A. SHYAM and V. CHITRA “EFFECT OF ELECTRODE MATERIALS ON THE NEUTRON EMISSION FROM A PLASMA FOCUS” Ann. nucl. Energy, Vol. 18, No. 6, pp. 357-358, 1991

Moly in a high rep rate source

R. Petr, A. Bykanov, J. Freshman, D. Reilly, and J. Mangano “Performance summary on high power dense plasma focus x-ray lithography point source producing 70nm line features in AlGaAs microcircuits” Rev. Sci. Instrum. Vol 75 No 8. pp 2551-2559 2004

To start, the paper by Rout et al shows that a very tiny fraction of the high Z atoms make it into the plasma but it doesn’t specify where the atoms are. Petr et al showed long lifetime operate of a Ne based plasma focus for ~5 million shots with moly and very little anode erosion.

R. Verma, P. Lee, S. Lee, S.V. Springham, T.L. Tan, R.S. Rawat and M. Krishnan, “Order of magnitude enhancement in neutron emission with Deuterium-Krypton admixture operation in miniature plasma focus device” App. Phy. Lett. 93 2008

B. L. Bures, M. Krishnan, R. Madden and F. Blobner “Enhancing Neutron Emission From a 500-J Plasma Focus by Altering the Anode Geometry and Gas Composition” IEEE Trans. Plasma Sci. Vol. 38 pp 667-671 2010

To further complicate the matter, high Z may not be a bad thing. People intentionally introduce high Z gases with the fuel gas to improve neutron yield. The high Z gases appear to improve compression of the pinch in small quantities (less than 10% by mass). Measurements with X-ray cameras do not show any contribution from a deuterium pinch. If the pinch was contaminated with high Z impurities they would by highly visible.

The final consideration is the x-ray contribution to the total yield. Even if the x-ray yield increases by orders of magnitude, the production of fusion charged particles will increase by orders of magnitude. The fusion contribution typically trumps the x-ray contribution by nearly 100x. If direct energy conversion is applied to the charged particles, something like 70% can be converted to electricity. For a 100 MJ per shot (fusion gain) system, that means nearly 69 MJ of energy derived from direct charged particle conversion. The x-rays are likely to be converted to electricity by heat then turbine. The x-rays make up ~1 MJ and can be converted at best 40%. Therefore, the x-rays provide ~400 kJ of electricity at best. One could claim some sort of exotic photovoltaic, but the conversion efficiency is not going to exceed 40%. By these arguments, the x-rays need to be shielded and not converted; a tiny fraction of the radiated power that is converted very inefficiently is not significant. Therefore, a refractory metal is the best choice for an anode.

[jamesr]

Thanks for the links. I haven’t read them all, but in R. Petr et al’s paper their ‘high power’ is 1.5kJ input yielding 0.08% of X-rays (ie 12J).

Using the figures from Eric’s 2007 Tech Talk (http://www.youtube.com/watch?v=O4w_dzSvVaM 36m30s in) he had:
Peak I 2.0MA
Gross input: 13.1kJ
X-Ray/Input: 81%
Beam/Input: 98%
(Beam+X-Ray)/Input: 1.79

so 0.81*13.1 = 10.6kJ ie. ~1000x as much X-ray energy as these lithography devices. One or two orders of magnitude more and maybe erosion resistant materials can cope with a reasonable lifetime, But we are talking about 3 orders of magnitude higher fluxes. The end of the anode has got to absorb only a tiny fraction of the X-rays passing through it if it is going to survive. Basically that means Beryllium is the only candidate that comes close.

[asymmetric_implosion]

jamesr wrote: Thanks for the links. I haven’t read them all, but in R. Petr et al’s paper their ‘high power’ is 1.5kJ input yielding 0.08% of X-rays (ie 12J).

Using the figures from Eric’s 2007 Tech Talk (http://www.youtube.com/watch?v=O4w_dzSvVaM 36m30s in) he had:
Peak I 2.0MA
Gross input: 13.1kJ
X-Ray/Input: 81%
Beam/Input: 98%
(Beam+X-Ray)/Input: 1.79

so 0.81*13.1 = 10.6kJ ie. ~1000x as much X-ray energy as these lithography devices. One or two orders of magnitude more and maybe erosion resistant materials can cope with a reasonable lifetime, But we are talking about 3 orders of magnitude higher fluxes. The end of the anode has got to absorb only a tiny fraction of the X-rays passing through it if it is going to survive. Basically that means Beryllium is the only candidate that comes close.

I refer you to comments made in another thread (https://focusfusion.org/index.php/forums/viewthread/973/P15/#9513) about this. I did a little back of the envelope math and it turns out that W is not so bad when one considers a few things.

[FrankOlson]

I was listening to the Oxford presentation. Pardon my layman’s knowledge in plasma physics perhaps, but as I understood, when the copper electrodes were being used, up to 1/3 or so of the plasma was impure and made up of copper from the electrode. So with tungsten, the purity issue is mostly resolved. What if the electrode, or a portion of it, was electroplated with the metalloid B11? I know there are other processes to make metals adhere to each other, but know what I mean. If a plasma is to be composed of impurities and some of those impurities are metal, why not have the metal impurity be the fuel at the same time? I read that boron 10 has not so good electrical conductivity until it’s heated to certain temperatures. I assume B11 would have relatively the same properties, so that’s not so good. Plausible?

[Author]

Posts