Math/physics in nuclear engineering?

[creepypasta13]

I graduated last year with BS degrees in physics and applied math. I also took heat transfer and fluid mechanics courses which I enjoyed. I'm looking into various areas to complete a phD in and nuclear engineering is one of them, but I know little about it. I had some questions about how much physics and math is involved in the theoretical grad courses in a phD nuclear engineering program:

1. How much quantum mechanics and E&M is used? I thought quantum mech was ok but didn't really like E&M as much.
2. How much heat transfer and CFD is used? What areas of ME besides those is also used in nuclear eng?
3. How much math is involved? Is there heavy use of linear algebra, numerical analysis, probability/stats, PDEs, and other math?

Basically my main concerns with applying to a phD program in nuclear engineering are:
1. I haven't taken any classes in it so I don't know if I'll like it.
2. Do nuclear engineering graduates find jobs easily outside of nuclear engineering? such as mechanical/aerospace industries?
3. Is a security clearance required for jobs in the nuclear industry?

 

[Astronuc]

1. How much quantum mechanics and E&M is used? I thought quantum mech was ok but didn't really like E&M as much.

QM and E&M may be used in nuclear engineering, but that the depends on the particular application or area of work. QM might be used in the development of finely detailed nuclear cross-sections. E&M would be used in radiation shielding and radiation effects on materials. With respect to the interactions of charged particles (betas and alphas), neutrons and photons with matter, much work has already been accomplished, and much of that work was done in the early years (50s, 60s and 70s) by physicists or engineers.

2. How much heat transfer and CFD is used? What areas of ME besides those is also used in nuclear eng?

A lot of nuclear engineering involves ME, particularly with respect to heat transfer and CFD, which are both used extensively in the design of core and reactor components, as well as reactor core simulation. In addition, transport theory of neutrons and photons are used.

3. How much math is involved? Is there heavy use of linear algebra, numerical analysis, probability/stats, PDEs, and other math?

All of those. Nuclear engineering is mathematically/computationally intensive.

Basically my main concerns with applying to a phD program in nuclear engineering are:
1. I haven't taken any classes in it so I don't know if I'll like it.
2. Do nuclear engineering graduates find jobs easily outside of nuclear engineering? such as mechanical/aerospace industries?

It really depends on the person, one's skills and the course work. Some of my school mates went into other fields. Proficiency in problem solving and applied mathematics is transferable.

3. Is a security clearance required for jobs in the nuclear industry?

A security clearance is not necessary, but some jobs will require a background check. Security clearance would only be required if one is work on nuclear weapons or military nuclear power systems, e.g., naval reactors. Safeguards work requires one submit to a background check - similar to one for a security clearance. If one becomes involved in nuclear or certain plants operations, the one is subject to fitness for duty scrutiny.

 

[creepypasta13]

QM and E&M may be used in nuclear engineering, but that the depends on the particular application or area of work. QM might be used in the development of finely detailed nuclear cross-sections. E&M would be used in radiation shielding and radiation effects on materials. With respect to the interactions of charged particles (betas and alphas), neutrons and photons with matter, much work has already been accomplished, and much of that work was done in the early years (50s, 60s and 70s) by physicists or engineers.

A lot of nuclear engineering involves ME, particularly with respect to heat transfer and CFD, which are both used extensively in the design of core and reactor components, as well as reactor core simulation. In addition, transport theory of neutrons and photons are used.

All of those. Nuclear engineering is mathematically/computationally intensive.

It really depends on the person, one's skills and the course work. Some of my school mates went into other fields. Proficiency in problem solving and applied mathematics is transferable.

A security clearance is not necessary, but some jobs will require a background check. Security clearance would only be required if one is work on nuclear weapons or military nuclear power systems, e.g., naval reactors. Safeguards work requires one submit to a background check - similar to one for a security clearance. If one becomes involved in nuclear or certain plants operations, the one is subject to fitness for duty scrutiny.

thanks for your help! So if it depends on the person, how do grad students decide to enter nuclear eng programs since most schools don't offer any nuclear eng undergrad classes?

I asked about the security clearance because I just my job due to getting my interim denied and the company didn't want to wait for months or a year to get the final clearance. I'm aware this could close ALOT of doors for me in the aerospace/defense industry, so I'm now considering other areas, and nuclear sounds pretty interesting

 

[Astronuc]

thanks for your help! So if it depends on the person, how do grad students decide to enter nuclear eng programs since most schools don't offer any nuclear eng undergrad classes?

I went through a nuclear engineering program, which was and still still around as a stand alone program. Other universities like Univ. of Florida, GaTech, NCState (Raleigh), Univ. of Tenn., Univ of Illinois, Univ. of Michigan, Univ. of Wisconsin, MIT, Oregon State, all have independent nuclear engineering program, while others like PennState and RPI have nuclear engineering programs associated with Mechanical or other engineering programs. Even PSU and RPI offer graduate NE programs.

Probably students get into nuclear through research, or a professor, or someone they know.

A hot area right now is multiphysics simulation, which is a skill used in many engineering and scientific fields.

 

[creepypasta13]

I've done some reading on it and it seems like numerical simulations of neutron and radiation transport is interesting. What kind of math/physics is involved with that? I tried to find out by looking at professor's publications, but all of them require payment.. you can only see the abstract for free

 

[Astronuc]

Neutron and Photon transport is frequently simulated with Monte Carlo methods.

Monte Carlo Lectures. LA-UR-05-4983 by Forrest Brown.
http://mcnp-green.lanl.gov/resources.html

http://www.engr.utk.edu/nuclear/TIW/BDG7.html
University of Tenn has a course Case Studies in Neutron Transport Theory. Note that the instructor/professor has a background in Aerospace and Mechanical Engineering.

Dr. B. D. Ganapol is professor in the Aerospace and Mechanical Department at the University of Arizona and a Research Professor in the Nuclear Engineering Department at the University of Tennessee. He is a noted expert in analytical methods of solution to the transport equation, in particular, the generation of benchmark solutions for code verification. He is currently a consultant at Los Alamos National Laboratory, Idaho National Laboratory and Oak Ridge National Laboratory and also consults for NASA on satellite remote sensing of vegetation.

Time dependent neutron transport equation
http://en.wikipedia.org/wiki/Neutron_transport

Here is an example of the underlying mathematics in neutron transport theory
Mathematical topics in neutron transport theory: new aspects By M. Mokhtar-Kharroubi
http://books.google.com/books?id=gRnd5FXrOPIC&printsec=frontcover

Otherwise, solves https://netfiles.uiuc.edu/mragheb/www/NPRE%20402%20ME%20405%20Nuclear%20Power%20Engineering/Neutron%20Diffusion%20Theory.pdf (use save target as)

 

[creepypasta13]

yeah the physics involved in the diffusion eq, neutron transport eq, and other PDEs in nuclear eng look really interesting. I love solving PDEs. Whats turning me away from NE is that I want to have no part with most 'practical' things like engines, heat exchangers, turbines, etc. Is knowledge of that stuff really necessary for those doing research in neutron transport simulations? I could see myself simulate the neutron transport in reactors.

The one 'practical' part of engineering that I like is the aerodynamics of missiles. But the physics involved isn't as interesting as that of neutron transport.

I thought about theoretical physics, but it gets too theoretical into stuff like relativity, which I can't stand.

So I'm still confused as to what to focus on for grad school

 

[law&theorem]

I went through a nuclear engineering program, which was and still still around as a stand alone program. Other universities like Univ. of Florida, GaTech, NCState (Raleigh), Univ. of Tenn., Univ of Illinois, Univ. of Michigan, Univ. of Wisconsin, MIT, Oregon State, all have independent nuclear engineering program, while others like PennState and RPI have nuclear engineering programs associated with Mechanical or other engineering programs. Even PSU and RPI offer graduate NE programs.

Probably students get into nuclear through research, or a professor, or someone they know.

A hot area right now is multiphysics simulation, which is a skill used in many engineering and scientific fields.

In thermo-hydraulic-physics coupling field, which univercity or national lab is the advanced among them ? And which is the best respectively in thermo, hydraulic and reactor physics ?

 

[Thermalne]

I do not know if my university (Tennessee) is the top of the top, but the professor I worked under does research in Thermo hydraulics.

http://www.engr.utk.edu/nuclear/ruggles.html

Although if you're looking for the top Nuclear Engineering program, the University of Michigan at Ann Arbor is number one.

 

[law&theorem]

I do not know if my university (Tennessee) is the top of the top, but the professor I worked under does research in Thermo hydraulics.

http://www.engr.utk.edu/nuclear/ruggles.html

Although if you're looking for the top Nuclear Engineering program, the University of Michigan at Ann Arbor is number one.

Ok, thank you.
And some professor in my univercity said: Argonne National Lab is the best ,especially in the field of fast breeder reactor, right ?

 

[Thermalne]

Ok, thank you.
And some professor in my univercity said: Argonne National Lab is the best ,especially in the field of fast breeder reactor, right ?

They might have a lot more research projects and topics going on related to Fast Breeder Reactors, but there really isn't the "best."

 

[nlsherrill]

Neutron and Photon transport is frequently simulated with Monte Carlo methods.

Monte Carlo Lectures. LA-UR-05-4983 by Forrest Brown.
http://mcnp-green.lanl.gov/resources.html

http://www.engr.utk.edu/nuclear/TIW/BDG7.html
University of Tenn has a course Case Studies in Neutron Transport Theory. Note that the instructor/professor has a background in Aerospace and Mechanical Engineering.

Time dependent neutron transport equation
http://en.wikipedia.org/wiki/Neutron_transport

Here is an example of the underlying mathematics in neutron transport theory
Mathematical topics in neutron transport theory: new aspects By M. Mokhtar-Kharroubi
http://books.google.com/books?id=gRnd5FXrOPIC&printsec=frontcover

Otherwise, solves https://netfiles.uiuc.edu/mragheb/www/NPRE%20402%20ME%20405%20Nuclear%20Power%20Engineering/Neutron%20Diffusion%20Theory.pdf (use save target as)

After glancing at that book I was surprised to see that it seemed like a real, hardcore, proof based math book. I have to ask(and sorry OP for taking over), and this will probably sound silly, but is there such an area as 'mathematical nuclear engineering" were people just work on new mathematics for advances in nuclear reactors? Or would this just lie within the realms of mathematical physics or something?

 

[Thermalne]

After glancing at that book I was surprised to see that it seemed like a real, hardcore, proof based math book. I have to ask(and sorry OP for taking over), and this will probably sound silly, but is there such an area as 'mathematical nuclear engineering" were people just work on new mathematics for advances in nuclear reactors? Or would this just lie within the realms of mathematical physics or something?

I doubt there is that much "new" mathematics. If you could solve the http://en.wikipedia.org/wiki/Navier–Stokes_equations that could lead to some kind of ramifications.

I believe Astronuc would be better fit to answer this.

 

[Astronuc]

After glancing at that book I was surprised to see that it seemed like a real, hardcore, proof based math book. I have to ask(and sorry OP for taking over), and this will probably sound silly, but is there such an area as 'mathematical nuclear engineering" where people just work on new mathematics for advances in nuclear reactors? Or would this just lie within the realms of mathematical physics or something?

Yes - there is a big initiative within DOE at the moment - and it basically involves multiscale computational physics.

The Nuclear Energy Advanced Modeling and Simulation (NEAMS) Program
http://www.nuclear.energy.gov/AdvModelingSimulation/program.html

Part of that effort involves advanced simulators of LWR systems.
http://www.nuclear.energy.gov/AdvModelingSimulation/casl.html
http://www.casl.gov/
http://westinghousenuclear.mediaroom.com/index.php?s=43&item=230
http://michigantoday.umich.edu/2010/06/story.php?id=7790

There is a lot going on and it's rather complicated but it involves a computational system - hardware and software - including advanced nuclear methods, computational fluid dynamics, and thermomechanical simulation systems - all of which are more intimately coupled. The ultimate system would be capable of simulating steady-state and transient system behavior over a broad range of systems from LWRs to FBRs to Gas Reactors to advanced Gen IV systems.

This post is somewhat relevant -
https://www.physicsforums.com/showthread.php?p=3050512#post3050512

See also -
https://hpcrd.lbl.gov/scidac09/talks/Knoll_scidac09.pdf

• Multiphysics simulation is the next frontier in computational nuclear
engineering.
• Jacobian-free Newton-Krylov methods and physics-based preconditioner
allow tightly coupled multiphysics simulations.

Computational nuclear engineering = computational physics

 

[Astronuc]

More taste of the mathematics. One finds similar math in aerospace, mechanical and electrical engineering - particuarly in power systems. It's matter of solving a system of non-linear time-dependent PDEs.

Jacobian-Free Newton Krylov Discontinuous Galerkin Method and
Physics-Based Preconditioning for Nuclear Reactor Simulations

http://www.inl.gov/technicalpublications/Documents/4074917.pdf

International Conference on Reactor Physics, Nuclear Power: A Sustainable Resource

HyeongKae Park
Robert R. Nourgaliev
Richard C. Martineau
Dana K. Knoll


Jacobian-free Newton–Krylov methods: a survey of approaches and applications
D.A. Knoll and D.E. Keyes
http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.83.8466&rep=rep1&type=pdf
Journal of Computational Physics 193 (2004) 357–397

NEWTON-KRYLOV METHODS FOR COUPLING TRANSPORT WITH CHEMISTRY IN POROUS MEDIA
http://proceedings.cmwr-xvi.org/getFile.py/access?contribId=315&sessionId=12&resId=0&materialId=paper&confId=a051
LAILA AMIR AND MICHEL KERN

A Jacobian-Free Newton-GMRES(m) Method with Adaptive Preconditioner and Its Application for Power Flow Calculations
http://www.ecs.csun.edu/~bruno/IEEEpapers/01664943.pdf
Ying Chen and Chen Shen, Member, IEEE
IEEE TRANSACTIONS ON POWER SYSTEMS, VOL. 21, NO. 3, AUGUST 2006

Proceedings of Computational Fluid Dynamics 2008

 

[creepypasta13]

So if I like solving the neutron transport eq analytically, but I want to have no part with most 'practical' things like engines, heat exchangers, turbines, etc in NE, would going into NE be a bad decision?

 

[gmax137]

So if I like solving the neutron transport eq analytically, but I want to have no part with most 'practical' things like engines, heat exchangers, turbines, etc in NE, would going into NE be a bad decision?

I dunno. But, I can tell you there's plenty of nuclear engineers that don't know the difference between a feedwater heater and an air compressor. Not that that's a good thing, but there it is.

On the other hand, many of the nukes working for the power companies at the plants do end up learning quite a bit about the rest of the plant. Turns out it's pretty interesting; also those companies tend to breed cross-trained employees. If you never want to see the plant, you'd be better of working at a fuel vendor or even more so, in academia.

 

[creepypasta13]

I dunno. But, I can tell you there's plenty of nuclear engineers that don't know the difference between a feedwater heater and an air compressor. Not that that's a good thing, but there it is.

On the other hand, many of the nukes working for the power companies at the plants do end up learning quite a bit about the rest of the plant. Turns out it's pretty interesting; also those companies tend to breed cross-trained employees. If you never want to see the plant, you'd be better of working at a fuel vendor or even more so, in academia.

yeah I could care less about the difference between a heater and compressor. My ultimate goal is to work in academia, but I realize the chances of that are slim.

Since I don't really care about nuclear power plants and care more about working on missiles and stuff defense contractors work on, I thought AE was better for me than NE. But OTOH, the actual physics involved in NE is more interesting, at least the neutron transport equation seems much cooler than say the navier-stokes equations

 

[Tony R]

There are some good post grad masters course which will allow you to get a good understanding of nuclear engineering, broaden your degree and can be the preparation for a PhD programme.

Look on nuclear liaison website for a list of all nuclear courses in UK -http://www.nuclearliaison.com/nl-courses

or for a specific MPhil course in Nuclear Energy to consider http://www-diva.eng.cam.ac.uk/mphil_nuclear/

 

[edgepflow]

So if I like solving the neutron transport eq analytically, but I want to have no part with most 'practical' things like engines, heat exchangers, turbines, etc in NE, would going into NE be a bad decision?

I don't think so. Most graduate NE programs are into theory and you could probably avoid most classes associated with electrical power generation. I graduated in nuclear engineering and after working 4 years at a nuclear power plant went back to school and earned a masters in mechanical engineering. So for me the practical side was my interest but again NE departments offer plenty of theory.

 

[Tony R]

I think you will find that some of the courses understand just the point you make and have a mix of theory and practice - based strong links to industry and labs.
Nuclear engineering and energy are now much more about how to make reactors work efficiently than just interesting places to do theoretical analysis and studies.

I might be biased, but look at the new masters in nuclear energy at Cambridge - it has its share of theory, practice and business - which you mix and match to suit yourself.

 

[Astronuc]

yeah I could care less about the difference between a heater and compressor. My ultimate goal is to work in academia, but I realize the chances of that are slim.

Since I don't really care about nuclear power plants and care more about working on missiles and stuff defense contractors work on, I thought AE was better for me than NE. But OTOH, the actual physics involved in NE is more interesting, at least the neutron transport equation seems much cooler than say the navier-stokes equations

Vendry ors use their own proprietary core design methodology which is based on multi-group diffusion theory. There is a lattice code which determines the macroscopic cross-sections for the each assembly design as a function of exposure (burnup) and hot and cold conditions, and for BWR fuel, void fraction. The lattice codes feed into core simulation codes. Aside from the codes developed by the fuel suppliers, there are independent codes, e.g., CASMO-SIMULATE by Studsvik-Scandpower. For other applications, there is a Monte-Carlo transport code, MCNP which was developed by Los Alamos. MCNP can be used for a variety of nuclear design applications.

Transport theory and NS are both interesting to solve, particularly as one focuses on smaller scales. NS is a subset of CFD. CFD is quite interesting especially if one includes fluid-structure interaction (FSI).

There is a move now to push into the mutli-physics realm on varying scales which combine, nuclear (transport theory), CFD and thermo-mechanical analyses in an integrated multi-physics system. Basically the math is much the same in the sense that one solves coupled systems of partial differential equations, which involve diffusion/transport of neutrons, and generally, mass, momentum and energy (including stress-strain states) in a range of materials.

The nuclear side of missiles is handled through the NNSA with the design and analysis performed at specific national labs. Otherwise, aerospace/defense companies provide rockets and missiles which use chemical propulsion systems. CFD, structural mechanics, control theory, guidance-navigation are used.

 

[QuantumPion]

So if I like solving the neutron transport eq analytically, but I want to have no part with most 'practical' things like engines, heat exchangers, turbines, etc in NE, would going into NE be a bad decision?

Not at all. Nuclear engineering curriculum is a combination of core/neutron physics and T&H, and depending on the university, probably has a number of electives to choose from in undergraduate study. If you primarily are interested in practical applications of nuclear technology, nuclear engineering would be a more productive avenue than pure physics in my opinion. You'll learn how to make approximations and solve the transport equation for realistic systems, as opposed to solving the schroedinger equation in 1-D for a single hydrogen atom .