Cascade Tube Particle Accelerator

[fatality314]

Building a Particle Accelerator

Hi,

I just stumbled upon these forums and joined with the hopes that I could seek some help from some more experienced people. Thank you for your time.

http://bigthink.com/ideas/19050

http://bt.pa.msu.edu/pub/papers/steeremsc/steeremsc.pdf

http://scienceworld.wolfram.com/physics/Cyclotron.html

http://www.science-project.com/_members/science-projects/1971/08/1971-08-fs.html

About 5-6 months ago I ran across an article describing an amateur physicist building a working and fully functioning "particle accelerator." I say this because what he built was not research grade but merely experimental. This piqued my interest as I have an intense interest in particle physics and cosmology. After reading of this man I began research on low-energy accelerators and their according uses. I have researched many types of accelerators including Linear, Cyclotronic, Sychotronic, and Betatronic. Of these Linear and Cyclotronic appealed to me the most due to their construction simplicity(relatively speaking) and efficiency/power, respectively.

I am a sophomore in high school, age 15. I am not just peaking my head at the surface of this and being naive. I have spent a good portion of 2-4 months researching this endeavor and its hazards and possibilities. I would like the comments to not tell me it is unsafe or too technical because I have researched well enough to know my way through this.

I have decided on a Linear accelerator for my endeavor due mainly to reasons pertaining to resources and money. I would love to build the same thing as this:

http://www.thecyclotronkids.org

But I simply do not have the money needed for that project. Instead I have opted for a design roughly based on this:

http://www.science-project.com/_members/science-projects/1971/08/1971-08-01.jpeg

http://www.science-project.com/_members/science-projects/1971/08/1971-08-fs.html

I plan on modifying this design to make it slightly more modern and powerful though.
This design seems simple enough for my first foray into design although I would love to go to new heights. The cyclotron that I would use if I had more money and resources would be one capable of proton pair-production. This leads me to the questions that I would like the community to input on. I can understand most of the questions that I ask but feel as if I don't COMPLETELY understand them as I feel I should.

Questions:

1. Proton pair production occurs with a particle with energy 2.045 MeV, correct?

2. Would this energy be plausible to attain with a smaller Linear Accelerator, as in the one in the design, if the accelerator was correctly modified?

3. If so what would the desired modifications?

4. What would you estimate to be the MeV potential of the accelerator as shown in the design, because I can't seem to find the right equations nor specifications as to it?

5. How would I find and/or make an "ion gun" as shown in the design?

6. What would you estimate the cost to be with ample safety measures and the maximum MeV potential applied to the design in the picture?

7. If I can not approach energies necessary for proton pair production what would be the most interesting experiments possible with the equipment shown for a budding particle physicist or cosmologist?

8. If you have any additional information that you feel I may need, please tell me. Thanks!

I sincerely thank you all for your help and if you have any questions I would be more than happy to answer. The reason for so many questions is that I would love to be exactly sure of everything before I start construction. Here are some extra resources:

http://www.science-project.com/_members/science-projects/1971/08/1971-08-fs.html

http://scienceworld.wolfram.com/physics/Cyclotron.html

http://www.science-project.com/_members/science-projects/1971/08/1971-08-01.jpeg

http://bigthink.com/ideas/19050

http://www.youtube.com/watch?v=PNJGqjB9uj8&feature=related

http://www.thecyclotronkids.org

http://bt.pa.msu.edu/pub/papers/steeremsc/steeremsc.pdf

http://www.slac.stanford.edu

Once again I deeply thank everyone and please respond!

P.S. If you find something significantly wrong with the design and whatnot please warn me. Although I do not see anything fundamentally wrong it would be greatly appreciated if I were told of specific hazards that I am not aware of. I say this as I do not want their to be significant hazards, as in this:

https://www.physicsforums.com/showthread.php?t=451654&highlight=particle+accelerator

 

[Bob S]

I think you are interested in electron-positron pair production (threshold about 1.02 MeV) and not proton-antiproton pair production (about 5400 MeV threshold in the Lab). Is that correct?

Because the electron-positron pair production is via electron beam --> bremsstrahlung --> photon photoproduction of pairs, the suitable electron beam energy is really closer to about 4 MeV. See the second plot in Fig 27.14 on page 22 of

http://pdg.lbl.gov/2009/reviews/rpp2009-rev-passage-particles-matter.pdf

for pair production in lead. In particular look at the curve labeled k_nuc.

Also look at Fig. 27.17 on page 25. For lead, the probability of pair production is about 50% for a 4 MeV photon.

Bob S

 

[fatality314]

Thanks!

I knew that it would produce positrons but I accidentally put proton instead of electron.

I was thinking of using a tantalum or tungsten target. Would lead be a better target?

Do you have any experience with BGO or bismuth germinate as a detector of pair production? I have heard that it is reliable and is cheap for gamma ray detection.

As I have significant monetary constrictions do you think that it would be possible for me to lessen the energy to around 2.5 Mev per say and just run it more often until I achieve pair production. I really do not care about efficiency as much as I care about actually achieving pair production and detecting it.

Any other information at all regarding this subject would be greatly appreciated. Thanks!

 

[fatality314]

Would bremsstrahlung decrease the energy of the electron beam that much over a distance around 3-7 feet? (I am not sure of the length of the cascade tube yet)

 

[Bob S]

Tantalum and tungsten are nearly as good as lead. Their Z's are slightly less than lead, but they are much better for high-power beams; they don't melt.

I have not used BGO, but have used sodium iodide a lot. The most common NaI is about 3" dia by 3" long, followed by a 10-stage photomultiplier. The best positron signature is the 0.511 MeV annihilation gamma at rest. You will need a pulse height analyzer.

See the 0.511 MeV line in the first sodium-22 spectrum in

http://community.middlebury.edu/~PHManual/Photos/gamma/fig3.html

This was taken from http://community.middlebury.edu/~PHManual/gamma.html

Bremsstrahlung occurs only when the electrons hit a target, preferably a high-Z target like lead. This can be an end window in a vacuum system. You will need a vacuum of about 10^-8 torr.

Bob S

 

[fatality314]

So, if I accelerated an electron at 0.511 MeV specifically, it has a higher chance of producing a pair than at other comparable energies?

How exactly would I go about making it that specific energy? Would that be where the pulse height spectrum analyzer comes in? Would the pulse height analyzer control the beam strength or simply detect it? How would I set the analyzer up to work?

I am sorry for so many questions but this is absolutely fascinating for me and I learn so much for my project. Much thanks!

 

[Bob S]

The threshold for pair production is 1.02 MeV, not .511 MeV, because both an electron and a positron have to be created at the same time. To get a sense of the probability of creating a pair from a photon, look at the graphs on page 22 and 25 of

http://pdg.lbl.gov/2009/reviews/rpp2009-rev-passage-particles-matter.pdf

The probability increases monotonically for photon energies above 1.02 MeV.

The photon (bremsstrahlung) spectrum of a 2-MeV electron hitting a tungsten target is a smooth spectrum with relatively few photons above 1 MeV. So 2-MeV electrons will produce only a few pairs.

The pulse height analyzer is used to to collect and store the signals from the BGO crystal (and photomultiplier) and is useful in observing the gamma ray spectrum resulting from positron annihilation at rest. The 0.511-MeV annihilation gamma is a unique signature of a stopping positron. See

http://community.middlebury.edu/~PHManual/Photos/gamma/fig3.html

Bob S

 

[fatality314]

Thanks. That helped a lot in understanding that aspect of pair production.

Another question though; based on the cascade tube accelerator design in this picture:

http://www.science-project.com/_members/science-projects/1971/08/1971-08-01.jpeg

Would it be possible for this design to accelerate the particles to at least the threshold energy? If so could you please point me in the right direction as to helpful modifications that could be made. Keep in mind that this design is merely an approximation. If you do not think that it would be possible with that design could you give me advise as to a better or more efficient design that would not be excessively expensive. (i.e. less than ~$750)

 

[berkeman]

You will need a vacuum of about 10^-8 torr.

Bob -- his article mentions construction intended for 10^-5 torr, and doesn't look to be able to support lower than that. Is that a problem?

Also, I'd be worried about implosion issues with the glass fittings that are described in the article he is following:

http://www.science-project.com/_members/science-projects/1971/08/1971-08-fs.html

.

 

[fatality314]

Thanks berkeman,

I just saw that also. Like I said I am not going exactly by this design but saw it as more of a starting point.

Could I just upgrade the quality of the tubing and fittings to make it withstand at least 10-6 torr. If I did that would it ensure safety and a better vacuum or would it just be easier to get a smaller diffusion pump and risk the quality of the beam/particles. Is a 10-8 torr ABSOLUTELY necessary or is it just something that would be nice to have for quality? Please correct me if I am wrong in any of this.

P.S. Keep in mind that I would be fine if I could get ~1.75 MeV out of the accelerator. I just want to create positrons and detect them. I have the detecting part down but the create part is still in progress. I don't need to be able to create pair's every time I run the machine. I just need to produce and detect it one time and I would be ecstatic. Anything more is just "gravy."

 

[fatality314]

Is there an alternative to the Van de Graaff generator that would be cheaper and more powerful for use in this linear/cascade tube accelerator?

 

[Bob S]

Bob -- his article mentions construction intended for 10^-5 torr, and doesn't look to be able to support lower than that. Is that a problem?

Also, I'd be worried about implosion issues with the glass fittings that are described in the article he is following:

http://www.science-project.com/_members/science-projects/1971/08/1971-08-fs.html

A beamline pressure of 10^-5 torr corresponds to a mean free path (or a 1/e beam flux attenuation length) of about 8 meters. See calculation in thumbnail. This is ok for a single-pass accelerator like this project, but not for a small cyclotron (multi-pass) which has a much longer beam path length in vacuum.

The 1½" diameter ID pyrex glass tube has to hold off 15 psi radial compression, and if the pyrex tube wall is ⅛" thick, the azimuthal wall compression is 15·1.5/(2·0.125) = 90 psi. This is very safe. Compare to the glass neck on a cathode ray tube. The brass (or aluminum) electrodes should be designed to provide a lip to mechanically support the ID of the pyrex tube, provide alignment, and a good vacuum sealing surface.

Bob S

 

[Bob S]

Is there an alternative to the Van de Graaff generator that would be cheaper and more powerful for use in this linear/cascade tube accelerator?

See pictures and discussion of the 750-KeV Cockroft Walton generator in the Fermilab article

http://www-bd.fnal.gov/public/proton.html

This is an alternative to the Van der Graaff, but probably more expensive.

Bob S

 

[fatality314]

I just came across this:

http://www.instructables.com/id/900000-volt-Van-de-Graaff-Generator-using-cheap-p

This seems perfect for my project. It has a 900 Kv potential and is actually cheaper to build than if I bought a 250 Kv Van de Graaff. I already have many of these parts and if I don't I could salvage some. I think I could build this 900 Kv Van de Graaff for around $100.

I have questions about two things though:

1. Is it possible for the generator to be too powerful? (i.e. will it weaken the integrity of the accelerator as a whole?)

2. It uses a "gazing ball" as it's head. How would I be able to get this:

http://www.science-project.com/_members/science-projects/1971/08/1971-08-02.jpeg

Inside of the gazing ball without causing possible structural problems?

Any help would be appreciated. Correct me if I am wrong in any of this. Thanks!

 

[Bob S]

These two tables

http://home.earthlink.net/~jimlux/hv/spherev.htm

imply that you need a 100-cm dia sphere to get to 900 kV. A gap spacing of 8" implies a voltage of about 500 kV.

The two cross section plots in the thumbnail imply that you need ~ 200 keV to get full neutron yield for a D-T (deuterium-tritium) reaction (5 barns at ~ 100 keV); ~ 1 MeV to get low yield (~ 0.1 barn) for D-D.

( 1 barn = 1 x 10^-24 cm² per atom)

Bob S

 

[fatality314]

So, the Van de Graaff generator in the described link only has a potential of around half of what is advertised? How could he be so off in his calculations?

If I were to use a stainless steel bowl with a diameter of 3 feet (~100cm.) would the generator achieve 900 kv or would I have to upgrade the entire machine for it to reach that potential?

I would like to use this design for the Van de Graaff generator because it seems concise and frankly is better than buying a kit or a pre-made one. I would like to know if the design modifications mentioned in the previous paragraph would be enough for it to reach ~900 Kv.

P.S. I am trying to make the Van de Graaff generator to be able to give enough power for pair production to occur when accelerating the particles.

 

[fatality314]

What would be other interesting experiments that I could do?

 

[fatality314]

Edit:

I am no longer aiming for pair production as it seems far off and I would rather not be working in those energies with me being only a high school sophomore. I would appreciate any other ideas on what I could do with the project ( i.e. possible and interesting experiments)

 

[Vanadium 50]

I would like the comments to not tell me it is unsafe or too technical because I have researched well enough to know my way through this.

I am no longer aiming for pair production as it seems far off

OK, so now you know that it was too difficult. You thought it wasn't. but it was. How sure are you that you are correct that you have covered all the safety issues?

 

[fatality314]

It is not really due to safety as much as it is to money. Which is a very large reason that I am trying to downsize this project and make it more realistic. I understand the safety precautions fully but I feel that after some more research that it is unrealistic and I would rather not waste money on something too far off that would also be dangerous. I would rather just make it smaller and cheaper whereas I could focus more on safety and get the adequate protection. If I were to do something more expensive, I would have less money to spend on safety for something that requires more. That is what I meant.

By the way, any recommendations as to an interesting or insightful experiment would be greatly appreciated. Thanks!

 

[Bob S]

I would like to make several comments on the this ion beam accelerator, just in case someone does build this 1971 Amateur Scientist design by Larry Cress proposed in the OP.

1) These very low momentum beams are affected by transverse magnetic fields such as the Earth's magnetic field. The steering and deflection is caused by the Lorentz force F = q v x B on a moving charged particle. For a laboratory experiment and example of the Lorentz force on an electron beam, see http://phoenix.phys.clemson.edu/labs/cupol/eoverm/

For a proton being uniformly accelerated in a 2.5-meter long accelerating column to 250 KeV, the beam in a 1-Gauss transverse magnetic field will be deflected by 1.1 cm (see equation in above url). This effect can be minimized by aligning the accelerating tube along the Earth's field lines.

For an electron being uniformly accelerated in a 2.5-meter long accelerating column to 250 KeV, the beam in a 0.1-Gauss transverse magnetic field will be deflected by 4.9 cm. This deflection would be very hard to correct, even if the accelerating column is aligned along the Earth's field lines. (However, unlike protons, electron sources are ubiquitous).

2) The copper tubes in the electrode assemblies in the accelerating tube will focus and defocus the beam as it gets accelerated, due to the shape of the electrostatic fields between the electrode assemblies. This focusing/ defocusing is very difficult to calculate.

3) The vacuum in the regions between all the electrode assemblies, and between the dome and the first electrode, has to be pumped out via the port near the target end. Each of the copper tubes in the electrode assemblies is a restriction, and limits the pumping capacity to the regions above it. The problem is especially acute between the dome and the first electrode. If a proton hits a residual gas molecule, it is likely to pick up an electron and become a neutral hydrogen atom, which then needs to be pumped out. Any buildup of hydrogen gas in this region will further impede the accelerating protons, and eventually "plug" the accelerating column. So the ion gun has to be made as efficient as possible. Suppose the ion gun is only 20% efficient in generating a 20-microamp ion beam. this means that the ion gun is injecting only about 1 nanogram of dry hydrogen gas per second. The injected gas rate has to be metered very sparingly.

Bob S

 

[fatality314]

Thanks Bob that was very helpful!

Any more equations or links to journals or designs pertaining to low-energy linear accelerators would be extremely helpful. I expect to spend many more months studying the math and vacuum system involved in this endeavor among other things.

I am trying to gain as much knowledge as possible before actually starting construction. I want things to work on paper before I begin construction. I am only an amateur.

What would be suitable analyzers or detectors that I could use to measure things such as magnetic field, particle speed/energy (eV), alpha/beta decay, gamma rays, etc.

Also any help on the design and specifications for an "ion gun" would be helpful.