![Arxiv:2103.10794V1 [Physics.Plasm-Ph] 19 Mar 2021 ΨN, N ≥ 124–26 Or Even an Exponential Relationship, P ∝ Synchronously](https://data.docslib.org/img/3a60ab92a6e30910dab9bd827208bcff-1.webp)
The effect of rigid electron rotation on the Grad-Shafranov equilibria of a class of FRC devices C.P.S. Swanson1 and S.A. Cohen2 1)Princeton Fusion Systems, Plainsboro, NJ, USA 2)Princeton Plasma Physics Laboratory, Princeton, NJ, USA (Dated: March 22, 2021) Rigid electron rotation of a fully penetrated Rotamak-FRC produces a pressure flux function that is more peaked than the Solov’ev flux function. This paper explores the implications of this peaked pressure flux function, including the isothermal case, which appear when the temperature profile is broader than the density profile, creating both benefits and challenges to a Rotamak-FRC based fusion reactor. In this regime, the density distribution becomes very peaked, enhancing the fusion power. The separatrix has a tendency to become oblate, which can be mitigated by flux conserving current loops. Plasma extends outside the separatrix, notably in the open field line region. This model does not apply to very kinetic FRCs or FRCs in which there are significant ion flows, but it may have some applicability to their outer layers. I. INTRODUCTION The Grad-Shafranov equation has also been used to model FRC equilibria with pressure flux functions that do not include 32,33 Rotamak-FRCs are plasma physics experiments in which rigid rotor effects. electron current is driven by an externally-imposed rotat- In Section II, we will derive the flux functions to be in- ing magnetic field (RMF).1 Interest in this configuration of serted into the Grad-Shafranov solver that are required by the plasma arises from its favorable properties for scaling into a condition of rigid electron rotation. In Section III, we will dis- nuclear fusion reactor, particularly a compact one.2–5 The fa- cuss the likely values of the free parameters that are defined vorable properties include: High plasma b, maximizing the in Section II, the likely relative peakedness of the density and plasma pressure for a given magnetic field; low internal field, temperature profiles. In Section IV, we will discuss the appli- allowing high-temperature and advanced fuels; and a simple, cability of this model to experiments and reactors, and sketch compact, and efficient method of heating and current drive in alterations that may be required. In Section V, we discuss the the form of the RMF system. RMF current drive dates to the solver which produces self-consistent MHD equilibria from 1960s.6 There were several Rotamak-FRC experiments oper- the equations in Section II. In Section VI, we will discuss the ating in the 1990s and 2000s.5,7–10 An existing example is the results of these MHD equilibria. In Section VII, we will con- Princeton Field-Reversed Configuration 2 (PFRC-2) experi- clude with a discussion of these results and their effect on the ment at the Princeton Plasma Physics Laboratory (PPPL).11 future design of Rotamak-FRC based fusion reactors. In fully penetrated Rotamak-FRCs, the current drive is as- sumed to be due to electrons rotating in synchrony with the applied external RMF while the ions are stationary.1,9,12–15 II. THE PRESSURE FLUX FUNCTIONS PRODUCED BY RIGID ROTATION There is also speculation that FRCs that are not driven by RMF will also rotate synchronously due to collisional effects.16 RMF-synchronous electron rotation has been ob- In this section we will derive the pressure flux function to 19,20 served in experiments17 and PIC (kinetic) simulation.18 substitute into the Grad-Shafranov equation. The Grad- Shafranov model has the pressure flux function as a free pa- A simplified (no Bt ) Grad-Shafranov model is often used to predict and reconstruct the MHD equilibria of these plasma rameter, but as we will see here, the condition of rigid electron configurations.19,20 While many modern analyses assume a rotation implies a functional form. Solov’ev20 pressure flux function, assuming P ∝ Y (pres- The central assumption of this model is that RMF-driven sure linear in flux),21–23 it has been known since 1982 that current (Equation 3) is the diamagnetic current (Equation 1), the Solov’ev linear pressure flux function is the least steep that is that ~j × ~B balances ~∇P. The net effect of RMF is that pressure flux function consistent with rigid rotation, and that it forces density to migrate across field lines until the pressure more realistic flux functions have a higher power law, P ∝ profile is such that the diamagnetic electron velocity rotates arXiv:2103.10794v1 [physics.plasm-ph] 19 Mar 2021 YN, N ≥ 124–26 or even an exponential relationship, P ∝ synchronously. This model also includes an isotropic pres- eY=Y0 .16,27–31 The well-known Rigid Rotor 1-D radial pres- sure and no ion flows. This model is of limited applicability sure profile P(r);B(r) implicitly assumes this pressure flux to reactor concepts in which kinetic effects dominate and ion function. flows are strong. Where these steeper pressure flux functions have been used to numerically generate equilibria, it has predominantly been for the purpose of fitting to experimental measurements. In A. Diamagnetic current condition, MHD equilibrium this context, the implications of this steeper pressure flux function on fusion reactor design have not been explored in We assume an ideal axisymmetric MHD equilibrium, and detail. that the plasma pressure at any given point is a function only 2 of the enclosed flux, P(Y(r;z)). The equilibrium equation is temperature vary as a power law with the pressure whose ex- simplified when it is assumed that there is no toroidal mag- ponents sum to 1: netic field. m The equilibrium condition is ne=n0 = (P=P0) (8) ~j ×~B = ~∇P (1) 1−m T=T0 = (P=P0) (9) By assuming axisymmetry, no toroidal field, and an where m is a number between 0 and 1, and P0 = n0T0 are isotropic pressure flux function P(Y), this equation becomes: the pressure, density, and temperature at some arbitrary point. m = 0 corresponds to the constant-density case, where varia- jf = 2pr¶YP (2) tion in temperature is responsible for the variation in pressure. m = 1 corresponds to the isothermal case, where variation in where f is the toroidal or azimuthal direction, P = SnT is density is responsible for the variation in pressure. the plasma pressure, r is the radial coordinate, Y(r;z) = The behavior for = 1 must be treated differently from the R r 0 0 m 0 dr 2prBz(r ;z) is the enclosed flux at point (r;z), ¶YP is the behavior for 0 ≤ m < 1. derivative of the plasma pressure P with respect to the mag- netic flux Y. Y has not been normalized by 2p as is some- times the custom. C. The case of 0 ≤ m < 1 Equation 2 is an intermediate step in the derivation of the 19,20 Grad-Shafranov equation. This case, encompassing all situations except the isother- mal, was explored by R.G. Storer in 1982 and 1983.24,25 The special case of m = 1=2 was explored in detail by I.J. Donnelly B. Rigid rotor current condition, penetrated RMF et. al. in 1987.26 Equations 4 and 8 have the solution:24 The central assumption of current drive in a fully- 1 penetrated Rotamak-FRC is that the current is due to all elec- P ∝ (Y − Y0) 1−m (10) trons rigidly rotating in synchrony with the applied RMF. The electron current is: m ne ∝ (Y − Y0) 1−m (11) jf = enerw (3) 1 − m where w is the angular rotational frequency of the applied T = ew(Y − Y0) (12) RMF. In order to compute P(Y), we will substitute Equation 2p 3 into Equation 2: where Y0 is the value of the flux at the plasma-vacuum bound- ary, which may or not be the separatrix. Y = 0 at the separa- 2pr¶YP = enerw (4) trix. Several interesting features are apparent: If instead the current drive were due to ions rotating against Steep power law: Equation 10 is what is substituted into stationary electrons, an ion momentum term would have to be Equation 2 to find the Grad-Shafranov equilibrium. This is a added to the Grad Shafranov equation. These results still hold power-law function, P ∝ YN, N = 1 ≥ 1. The least steep (with subscripts e;i transposed) in the regime that m r2w2 1−m i exponent is N = 1, the Solov’ev function, which corresponds T . i to m = 0. This analysis indicates that the Solov’ev solution We will now discuss the relationship between density, tem- is only valid for rigid electron rotation when the density is perature, and pressure: constant, and only the temperature varies. If the density is allowed to vary at all, N > 1 and the pressure flux function n T + n T = nT = P (5) i i e e becomes more steep. We will find in Section VI that a steep where function of flux results in a pressure profile that is peaked at the magnetic axis. The specificity of T: T, as specified in Equation 12, does ne = ni = n (6) not have a multiplicative free parameter as P;n do in Equations and 10 and 11. It is always linear to the flux, and the constant 1−m of proportionality is always 2p ew. For T to reach a large T = Ti + Te (7) thermonuclear value, w and Y must be large enough. This analysis breaks down when m = 1. This case is dis- are valid when Z = 1, the ion charge state is 1. cussed in Section II D. The m = 1 case is not T = 0 as Equa- To proceed, we must make an assumption of the rela- tion 12 would imply; rather it is T constant.
Details
-
File Typepdf
-
Upload Time-
-
Content LanguagesEnglish
-
Upload UserAnonymous/Not logged-in
-
File Pages10 Page
-
File Size-