Nuclear Fusion

Tags:
1. Jul 13, 2013

|mathematix|

After long thinking I decided that I want to do either engineering or science.
I am not really interested in one specific field in science but I am more interested in what field is more useful, active and revolutionary.
I became very interested in quantum computing but I lost interest since it doesn't really seem necessary at the moment outside research. Current technology will satisfy people's normal needs for decades (I think).
I'm very interested in nuclear fusion and nanotechnology/material science. Both fields really seem to have a great future in terms of research and funding and I feel like society will be forced to accept nuclear fusion as the future source of energy. There are still a few people who associate fusion reactors with nuclear weapons but they will realise that they are wrong if they get educated a little more.
I am currently more interested in nuclear fusion than nanotechnology because I have always found it so fascinating and kind of 'too good to be true' that we can make this much, almost infinite energy from matter.
My questions are:
1) What are currently the main problems facing nuclear fusion? I read that there were some problems with the interaction between the reactor core (not sure if it is called that) and the hot plasma so materials that can withstand such energies needed to be developed. Other problems I read are funding. Society doesn't seem interested in nuclear power. Will this change in the near future?
The reason why I am asking about the difficulties is to have a better idea of what kind of research will be more useful to the development of nuclear fusion.
2) Which degree would be better to do if I wanted to get into this field? Engineering (nuclear or energy) or science (physics)? Which one will be more important when the technology become implemented in around 20 years from now?
3) Are there good books that explain the science behind plasma fusion and modern fusion reactors?
4) What is the current source of energy to heat the gas and turn it into plasma and power the magnets? Is it renewable?
5) Is the job availability good or bad?

Thank you!

2. Jul 13, 2013

Staff: Mentor

"There is a world market for maybe five computers" - it is easy to underestimate the potential applications of new technologies.
I think quantum computing will get a significant market in the future.

Even if commercial fusion power plants become reality, this will take 40+ years. Before 2050, there are just non-commercial experiments, so funding really depends on the results of ITER and smaller experiments.

Fusion reactors use deuterium and (if they really fuse stuff) tritium. Both can be used in/for nuclear weapons.

I think it is money. It has been shown that larger reactors have a better performance, so I expect that ITER will be much better than existing experiments - but ITER is really expensive, a commercial reactor will have to be bigger and much cheaper at the same time. That is possible, but it is still unknown how much electricity from fusion will cost.
Sufficient tritium breeding has to be demonstrated, too.

Depends on the part you want to work on.
20 years? No. For commercial power plants, certainly engineering, but that is something for the next generations.
I am sure there are, but I don't know them.
Grid power. The mixture depends on the country you consider.

For physics/engineering, it is fine, and it does not depend so significantly on the specific topic you choose. You can consider it, but I would not make a choice based on that.

I would propose to wait a bit with a specific decision then. A BSc in physics gives you enough time to find a favorite area, and allows you to choose it afterwards.

3. Jul 14, 2013

|mathematix|

Thank you a lot for the reply!
After the failed cold fusion and sonoluminescence fusion I feel like governments and private investors were disappointed and decided to invest in something more reliable like solar and wind. Hopefully ITER makes some good progress because the technology seems very promising. Using solar and wind to power the magnets and heat the plasma can probably reduce the cost of the project.
I read that Lockheed Martin are developing a superior reactor compared to ITER so hopefully it works.
I don't really think the technology will be implemented in 50 years. Quantum computing was also estimated to take about 50 years (may be less) but Dwave made the first QC (many are still skeptic about whether it quantum computes) but I think we need a new source of energy urgently so if ITER or Lockheed don't make good and promising progress, nuclear fusion may become very far away and solar/wind will be used instead. If they do make good progress, I think more investments will be made and would take less than 50 years.
I was looking at future technologies to see which will be the most important and revolutionary and when I found that some progress is being made in nuclear fusion I thought that this is exactly what I want to do but it seems like a very risky field to get into.
In 4 months I will start my undergrad and the degree I will do will be research focused engineering/science so for now I need to make sure I get a 3.8 GPA to get into a good grad school and decide about what I want to do in like second year (I think I will need to major in second year).

4. Jul 14, 2013

Staff: Mentor

I don't see how you can reduce project costs by using more expensive electricity sources.
Anyway, electricity costs are not the main costs for ITER, construction and manpower are the dominating parts of the price. At 50MW non-stop (probably above the required power), 1 cent/kWh difference just saves 5 million €/year or 100 million € in 20 years. This has to be compared to the total project costs of 15 billion €.

Quantum computers and fusion power plants have completely different timescales - fusion power plants need to be big, so they are expensive and need years of development and construction. You cannot build a new prototype each month.

5. Jul 15, 2013

kinkmode

First of all, I'm coming from a background in magnetic confinement fusion, not inertial. Most of the following probably doesn't apply for inertial/laser fusion.

The main problems in fusion, in my opinion, are the materials issue you alluded to (often referred to as the first wall or plasma-materials interactions), funding, and the tritium cycle.

Any energy that you put into the plasma is eventually going to come out. The more energy you put in (and generate due to fusion), the more energy that comes out. Developing techniques and materials to mitigate and deal with this power flow is a pretty big issue.

Funding is self explanatory.

The tritium fuel cycle is a sticky issue. There are ideas on how to deal with it, like lithium or lithium salt blankets that will breed more tritium using the fusion-product neutron, but we haven't done much testing on that yet. It needs to be a closed loop, either directly on site, or with some extra processing, but the fusion plant needs to ultimately be able to breed enough tritium to be self sustaining. We are a long way from that. Other issues like tritium retention in the plasma facing surface are also tricky issues, but hopefully we will come up with satisfactory ways to deal with it.

Of course, there are a lot of physics issues to be dealt with still, but headway is being made on them. Things like learning more about and dealing with ELMs, mitigating disruptions, learning more about H-mode, improving non-inductive current drive methods, dealing with diffusion, etc.

As far as books, no, not a lot of good books dealing with fusion reactors in my opinion. You first need to learn plasma physics, which often mention fusion, but don't deal exclusively with it. See Chen, Bellan, Krall & Trivelpiece, Hazeltine & Waelbroeck, and/or Goldston. Some books are a little more fusiony than others. IF you can find it, Ideal Magnetohydrodynamics by Freidberg is a great book that delves into MHD from a very fusion relevant point of view. Once you have the basics of plasma physics and MHD, if you really want a good but expensive overview on how a tokamak works, I'd recommend picking up Wesson's Tokamaks. There's a third edition that is out that is up to date, but I feel that if you can get through the 2nd edition of it, you'll know quite a bit about tokamaks, and maybe save a few bucks too.

Degrees: I wouldn't worry so much about what the degree is in. Just go to a program that is strong in fusion research. You'll find that depending on the school/program, your degree will be in different things. At Princeton, it's in Astrophysics. At Wisconsin, it can be in physics, engineering physics, or electrical engineering (I think). At MIT, I think the degree is in nuclear engineering. Go to a good program that is active in the community, and people won't care what your degree is in; you'll graduate, most likely with a relevant postdoc/permanent position already lined up, and be just fine.

Job availability: In my opinion, if you want to stay in the field of fusion research and are willing to do what is necessary for it, like moving, taking a post doc or two, possibly living out your days in places like Oak Ridge or Los Alamos, the job prospects are quite good. Most all of my friends are still employed in fusion research post graduate school. Many have permanent positions. A good handful of them have decided to leave magnetic confinement fusion and move over to the ICF side at LLNL or Sandia; all of them are well paid and enjoy their work. Job prospects outside of the field are pretty dim, outside of a few companies and locations. A few friends are in industry, but I think most (all?) are in CA.

Lastly, cost: Magnets in a tokamak are a huge portion of the cost, often the most expensive component. Not in powering them. Just building them. I just did some googling, and I saw a couple talks/papers that put the ITER magnet system at 25% of the total cost of ITER (not sure what 'total cost of ITER' that is - it's been a moving target). Still, whether using $10 billion or$20 billion as your total figure, you are talking about spending \$2.5-5 billion alone on magnet materials and construction.

As far as operation costs, at the research stage, it's almost always in personnel, not in electric costs. ITER might shift that a bit, since they are going to be dealing with some pretty difficult situations, like tritium, beryllium first walls, etc.

Lastly, if you are really interested in fusion, I would recommend doing at least one summer of research under a program like NUF:

http://science-education.pppl.gov/NUF/Overview.html

Not only will it give you some valuable first hand experience on a fusion project, if you play your cards right, you'll make some contacts in the field which will help you with your grad school prospects. I have a number of friends who either got permanent jobs as engineers in fusion labs out of college or who entered fusion related Ph.D. programs after doing the NUF program.

6. Jul 16, 2013

|mathematix|

Unfortunately, I live in Australia and I can't do any programs here.

The only place that does nuclear fusion research in Australia is the Australian National University. ANU has a very strong focus on research and theory and it is ranked in top 30 in the world, I think.
This is a link to the nuclear fusion research section http://h1nf.anu.edu.au/.

My plan is to work hard during undergrad and gain some good research skills as research is a compulsory part of the degree then get a good GPA, recommendations from professors then apply for a good European or American university but there is still the chance that I won't get the scholarship so it is a big decision to make.

The Australian government spends very little money on scientific research and are strongly against any form of nuclear (they like to sell the uranium instead as we have a lot of it).

I don't know if I should just do what every one else in Australia does and follow the money or follow my passion and do something meaningful.

7. Jul 16, 2013

kinkmode

I went to school with a lot of foreign students. As far as I know, they were supported just like the American ones in terms of tuition being covered and a stipend on top of that. So if it's what you really want to do, it's possible that you might be able to get your graduate degree paid for in the US. I can't really speak to how difficult it is to get in for foreign students though.

Good luck!

8. Jul 16, 2013

|mathematix|

Thank you!
PhB (https://studyat.anu.edu.au/programs/4660HPHB;overview.html) allows me to do graduate medicine with no admission tests if I pass so I think this is a safe option.
I will do PhB (major in chem and phys) and look into scholarships and how to maximise my chance of getting one but if I fail to get one I might do medicine because working in research in Australia is very bad, not only in terms of money but the research itself is on small, boring projects.
I know a guy from UNSW who got a scholarship from Cambridge (100k p.a. for uni fees and living expenses) but he managed to get his masters research published. Another guy is very good at maths and topped his class at USYD but got rejected by all American and European universities so it is pretty difficult.

9. Jul 16, 2013

kinkmode

My one friend from grad school from Australia was a math guy too. I think he had a master's degree already when he got into the University we went to. Actually, thinking back on it, a lot of the foreign students had master's degrees already.