October 4, 2011
Outline
• Discuss the fusion power flow chart to determine the fusion gain required by any given inertial fusion scheme in order to produce net power on the grid
– Compare PJMIF against laser ICF
• Explain the fusion burn configuration for PJMIF
• Explain how this is achieved in PJMIF → a discussion of the implosion scheme
• Explain how the target is magnetized
• The plasma guns
• Concluding remarks
The Fusion Power Flow Cycle
The PJMIF Fusion Burn Configuration
The PJMIF Fusion Burn Configuration
PJMIF needs to accomplish two things
A spherical chamber with one or more sets of plasma guns
Two sets of jets are launched by the plasma guns
The target, the imploding liner, and the magnetization of the target
The dynamics of the converging liner
• The pressure of implosion is provided by the momentum flux density, rv2, of the liner (commonly called its ram pressure).
• As the liner converges towards the center, its density increases rapidly inversely as the square of the radius (inverse square law), so does its ramp pressure (ram pressure amplification, A)
• The ram pressure amplification is limited by
– the self heating of the liner, creating self internal pressure that works against its inward motion towards the center.
– Hydrodynamic instabilities, such as the Rayleigh-Taylor instability
Assessment of the potential of PJMIF
• We have used analytical and lumped parameter models to assess the potential of PJMIF
• Assessments are not unanimous among all authors
• Generally there are more favorable assessments
• We have reasonably advanced computer codes for modeling the implosion dynamics of PJMIF
– 1D Radiative Hydrodynamics codes (RAGE of LANL)
– 3D Smoothed Particle Hydrodynamics for study of asymmetric effects, preliminary studies of Rayleigh-Taylor instabilities
Ref: Thio et al. (1999); Cassibry let al. (2009); Awe et al. (2011);
Hsu et al.(2011); Parks (2008); Samulyak (2010)
An illustrative computer modeling example
Lf1d-thio-ab-54: 30 MJ liner energy
1D Lagrangian plasma dynamics with ideal gas EOS, fusion burn, fixed-parameter alpha energy re-deposition, no radiative transport.
Shown are interfaces #52, #39, #27, #24, #22, #12, #3
Target and liner jets merge at a radius of 60 cm
Target is structured – inner and afterburner layer
Liner has kinetic energy of 30 MJ
An analytic model is used to compute the flow down to a radius of 8 cm
The 1D computer code then takes over and simulates the detailed dynamics and fusion burn
An illustrative computer modeling example
Liner convergence and implosion dynamics
• The implosion dynamics is sensitive to the details of the physics assumptions:
– Radiative transport
– Plasma interpenetration
• Our computer codes at present make reasonable assumptions about these physics details, but are not exact, and have not been validated against experiments
• They serve to give qualitative expectations of the potential of the concept and scoping of the magnitudes of the key parameters.
Liner Convergence and Implosion Dynamics: What remains to be done
• The equation of state, the radiation properties (opacities), the collisionality, the growth rates of instabilities need to be accurately measured, modeled, or characterized using a combination of experiments and computational modeling
• This physics needs to be captured in the most advanced hybrid particle-in-cell and smoothed particle hydrodynamics codes, complete with radiative magnetohydrodynamics, fusion burn physics with particle energy deposition.
• The codes need to be validated against further experiments so that they can be relied upon as being predictive rather than hind-casting.
• Apply the codes systematically to search for and design the most optimum set of initial conditions and implosion trajectory.
The company that pays for this development owns the know-how and the computer codes
• The combination of computer model development and experimental validation, leading to predictive computer models and using
them in engineering design is common in the development of advanced technology today.
• This is a peta-flop-scale supercomputing effort.
• Supercomputing modeling is a key enabling tool for our program
– A state-of-the-art supercomputing cluster based on GPU parallel computing is planned for our proposed program.
Target Magnetization
• Our favorite technique for target magnetization is the use of lasers, but we are open to other possibilities.
• The basic principle is based on the beating of two plasma waves to produce a beat wave which is optimally coupled to the electron thermal motion, resulting in a net acceleration of the electrons in the direction of the plasma beat wave.
• The plasma waves are excited by electromagnetic waves in the form of laser beams.
• Prior to the firing of the lasers, a pre-initial magnetic field is set up to guide the accelerated electrons.
Target Magnetization: What has been done and what remains to be done
• The principle has been demonstrated in the microwave regime in plasmas with densities several orders of magnitude less than what we plan in PJMIF.
• The use of lasers to produce plasma beat waves to drive currents in dense plasma is a new and innovative technology to be developed on the program. The technology might have other scientific and industrial applications.
• Advanced 3D particle-in-cell (PIC) plasma code with accurate laserplasma interaction physics and EM field calculation need to be developed.
• Experiments need to be conducted to validate the code, so that it has a predictive capability.
• The computer codes and the lessons learned in developing the technique is the technological know-how to be owned by the company, and which cannot be acquired in any other way.
• The codes need to be applied to search for and design the most optimal way of establishing an initial seed magnetic field in the target plasma.
The Key to PJMIF is the ability to produce plasma jets of high density, high Mach number and high velocity
How are these plasma jets produced?
Conventional mode of operating pulsed plasma guns
What generates the accelerating force in an electromagnetically driven plasma
40+ years of research based on this snow-plow mode of accelerating the plasma
• Plasma guns were researched since the 1960’s mainly by NASA (or its predecessor) and the Air Force as pulsed plasma thrusters (PPT) for space applications.
• The snow-plow mechanism was formulated by Marshall Rosenbluth in a 1952 paper to explain the acceleration of plasma in a theta pinch.
– Researchers then apply it to accelerate plasma in plasma guns instead.
• PPTs have been flown by NASA and the Air Force on many satellites for station keeping
• Very well developed technology
However, there was an inherent performance barrier of the conventional PPTs
• Performance improvements on these PPTs grind to a halt in the mid-1990’s
• Thio reviewed the field and concluded the performance barrier was rooted in the pre-fill and the snow-plow mechanism of accelerating the plasma (Thio, 2000)
There were two major flaws in the snow-plow mechanisms
• Critical ionization velocity barrier
– When the current sheet reaches a velocity sufficient to cause ionization of the neutral atoms in front of it, the neutral atoms are ionized, and energy is taken out of the current sheet, slowing it down
• The snow-plow was found to be leaky
– Not all of the neutral atoms get entrained by the current sheet
• Some leak past the current sheet
• As bad as 10% to 20%
– Neutrals left behind become seeds for arc re-strike behind the main plasma.
• The restrike arc draws current away from the main current sheet, causing a reduction in the force of acceleration.
A paradigm shift is required: No pre-fill, no snow-plow, no surface flashover
• To produce plasma jets with the required density, velocity and Mach number required for PJMIF, a new mode of operating the plasma gun was proposed by Thio (2000) with the following prescription:
(2) Inject the plasma to be accelerated with sufficient density supersonically
(3) Turn on the voltage across the injected plasma, producing a controlled current sheet over the back of the conducting plasma – no surface flashover.
– The plasma has a well defined initial condition, making controlled acceleration of the plasma slab possible
(4) The electrodes are shaped so that the plasma is accelerated as a compact slab
What happens when the electrodes are not shaped?
MHD shaping of the electrodes stabilize the acceleration of the plasma as a slab
Engineering implementation of the new plasma gun concept
We have now routinely produce plasma jets with these devices
For expediency, in the LANL PLX experiment, we are using plasma railguns with cylindrical nozzle
What remains to be done for the Plasma Guns
• The overall objective is to deliver the required density, Mach number, velocity and degree of purity, and electricto- kinetic efficiency
– Learn to inject two or more gaseous materials into the same gun
• Demonstrate the ability to attain the required degree of purity
• Demonstrate the attainment of the required jet density and mass
• Demonstrate the attainment of the required jet Mach number (temperature)
• Learn to operate the gun repetitively and demonstrate its lifetime between maintenance
• Advanced 3D hybrid, 2-fluid, Hall PIC-MHD
• Development will be assisted by the development of predictive computer codes for modeling the gun.
Concluding Remarks
• PJMIF is an attractive fusion concept with relatively low cost scientific experimental analysis and research & development path
•Can PJMIF produce economic fusion power?
– The resolution of these uncertainties and learning curves is the objective of Phase I of the program, and will provide a definitive answer to the question
– There is no guarantee what the answer will be at this point. It requires a minimum amount of scientific experimental research, pure and simple.
– By the same token, the group that pays for Phase I will own all the computer codes and the engineering experience and lessons, which collectively is the technological know-how.
– This technological know-how might still be valuable in spin-off applications if the answer is “No”, or it might be worth billions if it is “Yes”. To the best of our knowledge, we believe the answer is likely to be “YES”.