How would you get to grips with nuclear fusion?
June 13, 2010 11:51 AM Subscribe
Nuclear fusion physics and tokamak reactors: where to start?
Assuming that a layman with 'high-school' level scientific knowledge wished to fully understand the complex physics underlying a tokamak fusion reactor, and had five years in which to do so, what would be the essential building blocks of physics necessary to begin the long journey upward? An image that springs to mind is a pyramid of square blocks, each successive level allowing advance to the next. Another assumption (possibly mistaken) is that a full physics degree would involve units and modules not relevant to the specific subject of desired study. Somehow, this seemed a little too specific and prosaic for the Wiki Reference Desk.
Assuming that a layman with 'high-school' level scientific knowledge wished to fully understand the complex physics underlying a tokamak fusion reactor, and had five years in which to do so, what would be the essential building blocks of physics necessary to begin the long journey upward? An image that springs to mind is a pyramid of square blocks, each successive level allowing advance to the next. Another assumption (possibly mistaken) is that a full physics degree would involve units and modules not relevant to the specific subject of desired study. Somehow, this seemed a little too specific and prosaic for the Wiki Reference Desk.
Just to be clear, the maths, the classical mechanics and the electrostatics are at the base of the pyramid. Then you need to pursue the plasma physics and nuclear physics and finally you combine them to understand how a Tokamak works.
Note though, that you will have to cover the majority of an undergraduate physics curriculum just to understand the basics. I just noticed that I haven't included special relativity, which is basically required to understand any modern physics at all. I think you can get away with leaving quantum out entirely, unless you really want to understand how nuclear binding energy actually works.
This may not prepare you for how actual tokamaks work though, because dusty plasmas also require statistical physics.
Still, you definitely don't need General Relativity. So that's something.
posted by atrazine at 12:18 PM on June 13, 2010
Note though, that you will have to cover the majority of an undergraduate physics curriculum just to understand the basics. I just noticed that I haven't included special relativity, which is basically required to understand any modern physics at all. I think you can get away with leaving quantum out entirely, unless you really want to understand how nuclear binding energy actually works.
This may not prepare you for how actual tokamaks work though, because dusty plasmas also require statistical physics.
Still, you definitely don't need General Relativity. So that's something.
posted by atrazine at 12:18 PM on June 13, 2010
Don't mean to dominate the thread here, but could you tell us a little more about how much mathematics you've done? I mean, I just blithely put in there "whatever you need for the calc", but that includes quite a lot of trigonometry, complex numbers, and other things. I'm not very good at gauging what "high-school" scientific knowledge actually means in the US.
posted by atrazine at 12:20 PM on June 13, 2010
posted by atrazine at 12:20 PM on June 13, 2010
Best answer: atrazine has given a good high level overview. Here are some more resources:
One approach would be to look at this curriculum for a graduate program in fusion engineering at UC Berkeley. Each of the course descriptions includes prerequisite courses and skills. As you drill your way down you'll eventually get to basic physics, engineering, and math courses.
Just to get you started, you'll definitely need a good understanding of calculus and differential equations, including multivariable calculus (i.e., what colleges typically call Calculus I, II, III and Differential Equations). You'll also need an understanding of the basic physics behind electricity and magnetism, as would be discussed in the first year of university physics courses. That will get you most of the way to understanding Maxwell's equations.
Following a university pace that would take you about a year if you do the math and physics simultaneously.
posted by jedicus at 12:22 PM on June 13, 2010
One approach would be to look at this curriculum for a graduate program in fusion engineering at UC Berkeley. Each of the course descriptions includes prerequisite courses and skills. As you drill your way down you'll eventually get to basic physics, engineering, and math courses.
Just to get you started, you'll definitely need a good understanding of calculus and differential equations, including multivariable calculus (i.e., what colleges typically call Calculus I, II, III and Differential Equations). You'll also need an understanding of the basic physics behind electricity and magnetism, as would be discussed in the first year of university physics courses. That will get you most of the way to understanding Maxwell's equations.
Following a university pace that would take you about a year if you do the math and physics simultaneously.
posted by jedicus at 12:22 PM on June 13, 2010
last comment I promise
Nobel Laureate G. t'Hooft has a site where he collects all the stuff that an autodidact might want to learn to become a theoretical physicist, obviously you don't need all of this, but you can use it to find free online resources for what you do need.
posted by atrazine at 12:23 PM on June 13, 2010
Nobel Laureate G. t'Hooft has a site where he collects all the stuff that an autodidact might want to learn to become a theoretical physicist, obviously you don't need all of this, but you can use it to find free online resources for what you do need.
posted by atrazine at 12:23 PM on June 13, 2010
Response by poster: Prompt, thorough, and precisely what I needed. My heartfelt thanks to you both!
posted by the_observer at 12:25 PM on June 13, 2010
posted by the_observer at 12:25 PM on June 13, 2010
Best answer: The only thing that I don't think is really necessary in a physics degree to understand tokamak fusion reactors is quantum mechanics. You'll need a basic understanding of QM to understand some of the approximations, but the details are largely irrelevant to the plasma physics.
Other than that, I'm afraid I think a physics degree is pretty much what you're looking for. Of course, you could learn everything you'd learn as a physics major on your own, but for most people, it would be a lot harder.
Classical mechanics and electrodynamics are the basic building blocks of plasma physics. A bit of relatively comes in handy, and a general understanding of modal decomposition of systems is crucial, but you would get that in a good intro CM or E&M course.
So yeah, if you were going to do this yourself, I'd do the following:
Learn multivariable calculus, differential equations, linear algebra (3-5 courses)
Learn classical mechanics (1-2 college courses)
Learn electrodynamics (2-3 college courses)
Learn about "waves and oscillations", which is often a course by itself (1 college course)
Learn a bit about relativity (1/2 - 1 college course)
Learn about basic QM (1/2 - 1 college course)
Learn plasma physics (2 - infinity grad courses)
Study the early literature about tokamaks, read plasma physics textbooks (1-3 grad courses)
Read current papers about tokamaks (1-infinity grad courses)
Total estimated equivalent learning: 12 - 19 semesters of college work
If you did the equivalent of four semester-courses per year, you'd have a good understanding of tokamak physics in 3-5 years.
I think it takes an unusual person to be able to do that sort of thing in a self-directed manner, but if you're that sort of person, maybe it'll work.
posted by Salvor Hardin at 12:26 PM on June 13, 2010
Other than that, I'm afraid I think a physics degree is pretty much what you're looking for. Of course, you could learn everything you'd learn as a physics major on your own, but for most people, it would be a lot harder.
Classical mechanics and electrodynamics are the basic building blocks of plasma physics. A bit of relatively comes in handy, and a general understanding of modal decomposition of systems is crucial, but you would get that in a good intro CM or E&M course.
So yeah, if you were going to do this yourself, I'd do the following:
Learn multivariable calculus, differential equations, linear algebra (3-5 courses)
Learn classical mechanics (1-2 college courses)
Learn electrodynamics (2-3 college courses)
Learn about "waves and oscillations", which is often a course by itself (1 college course)
Learn a bit about relativity (1/2 - 1 college course)
Learn about basic QM (1/2 - 1 college course)
Learn plasma physics (2 - infinity grad courses)
Study the early literature about tokamaks, read plasma physics textbooks (1-3 grad courses)
Read current papers about tokamaks (1-infinity grad courses)
Total estimated equivalent learning: 12 - 19 semesters of college work
If you did the equivalent of four semester-courses per year, you'd have a good understanding of tokamak physics in 3-5 years.
I think it takes an unusual person to be able to do that sort of thing in a self-directed manner, but if you're that sort of person, maybe it'll work.
posted by Salvor Hardin at 12:26 PM on June 13, 2010
*bit of relativity
posted by Salvor Hardin at 12:27 PM on June 13, 2010
posted by Salvor Hardin at 12:27 PM on June 13, 2010
Response by poster: Oh, and as it happens, I'm not from the US but assumed that much of MeFi's population was, hence the quotes around 'high-school' I completed Physics at A-Level, but my last Maths was at GCSE so that'll take a lot of brushing up and refreshing.
posted by the_observer at 12:29 PM on June 13, 2010
posted by the_observer at 12:29 PM on June 13, 2010
Something else worth pointing out - you specify the tokamak system. It's not the only game in town. There are folks out there working on taking the Farnsworth–Hirsch Fusor concept and making a practical energy producer out of it.
posted by Kid Charlemagne at 12:54 PM on June 13, 2010
posted by Kid Charlemagne at 12:54 PM on June 13, 2010
There's a small technical university in my home town that has a Faculty of Energy Systems and Nuclear Systems, including bachelor's programs in Nuclear Engineering and Nuclear Power. There are two nuclear plants close by and it seems like they have very close ties. Here's their website, you may find something of interest there.
posted by PercussivePaul at 1:01 PM on June 13, 2010
posted by PercussivePaul at 1:01 PM on June 13, 2010
You mentioned physics, not engineering, but some of the physics is only meaningful in the context of the engineering of the Tokamak device. A certain percentage of the energy will be carried away in Bremsstahlung radiation, for example, but the exact amount depends on the design of the device itself.
posted by It's Never Lurgi at 1:01 PM on June 13, 2010
posted by It's Never Lurgi at 1:01 PM on June 13, 2010
Response by poster: Thanks for the additional sources, they've all been noted and logged. Fortunately, (regarding a couple of messages I've been sent) I'm in my twenties, I've got experience of lengthy projects consisting of thousands of smaller steps, and I don't easily get bored or distracted (particularly when it's a subject as interesting as this).
posted by the_observer at 1:06 PM on June 13, 2010
posted by the_observer at 1:06 PM on June 13, 2010
I'll just add that if you were actually going to attempt to build such a thing you would need a whole set of skills completely separate from everything that's been listed already. At a minimum you'd need the electrical engineering background to design the power electronics that control the field coils, which would mean the equivalent of several semesters worth of general electronic theory followed by some exposure to the specialization of power electronics and probably some control systems theory. And of course you'd need a whole slew of technical skills such as circuit layout and assembly, machining and metalworking, etc.
posted by Rhomboid at 2:03 PM on June 13, 2010
posted by Rhomboid at 2:03 PM on June 13, 2010
This thread is closed to new comments.
Then, there will be two "streams" converging to understanding what happens in a Tokamak, these are plasma physics and particle/nuclear physics.
Calculus: start with single variable integration & differentiation, work up through multivariable, 2 & 3D, vector calculus, and differential equations
Other math: Basic algebra, matrices, vectors, whatever you need as a basis for the calc.
Electrostatics
Basic classical mechanics
For Plasma:
Electromagnetism: To include Maxwell's laws and common solutions in integral and vector forms
Some fluid mechanics might help but is not essential
Plasma physics itself
Nuclear physics:
Understand the basics of nuclear binding energy and collision cross sections
Advanced knowledge of high energy/particle physics is not required
What you could mostly skip from a standard undergraduate program:
Virtually all of quantum mechanics
Advanced particle physics / standard model
Gravitation
posted by atrazine at 12:13 PM on June 13, 2010