Skip to content

MHD and frozen-in flux

On large scales and slow timescales we can treat the plasma as a single conducting fluid. This is magnetohydrodynamics (MHD).

Ohm's law and the induction equation

For a fluid moving with velocity \(\mathbf{v}\) and electrical conductivity \(\sigma\), the simplest Ohm's law is

\[ \mathbf{E} + \mathbf{v}\times\mathbf{B} = \frac{\mathbf{J}}{\sigma}. \]

Combine this with Faraday's law \(\partial_t \mathbf{B} = -\nabla\times\mathbf{E}\) and Ampère's law \(\mu_0\mathbf{J} = \nabla\times\mathbf{B}\) (displacement current neglected) to get the induction equation:

\[ \frac{\partial \mathbf{B}}{\partial t} = \underbrace{\nabla\times(\mathbf{v}\times\mathbf{B})}_{\text{advection}} + \underbrace{\eta\,\nabla^2\mathbf{B}}_{\text{diffusion}}, \qquad \eta = \frac{1}{\mu_0\sigma}. \]

Here \(\eta\) is the magnetic diffusivity (units m\(^2\)/s).

Magnetic Reynolds number

Compare the two terms using a length scale \(L\) and velocity \(U\):

\[ R_m = \frac{|\nabla\times(\mathbf{v}\times\mathbf{B})|}{|\eta\nabla^2\mathbf{B}|} \sim \frac{UL}{\eta}. \]

When the velocity is the Alfvén speed,

\[ v_A = \frac{B}{\sqrt{\mu_0 \rho}}, \]

this ratio is called the Lundquist number:

\[ S = \frac{L v_A}{\eta}. \]

Alfvén's frozen-in theorem

When \(R_m \gg 1\), diffusion can be dropped and

\[ \frac{\partial \mathbf{B}}{\partial t} = \nabla\times(\mathbf{v}\times\mathbf{B}). \]

One can show that the magnetic flux through any surface moving with the fluid is then constant. Field lines behave as if they are frozen into the plasma: two fluid elements on the same field line stay on the same field line forever.

How frozen is it?

In the solar corona, \(L\sim10^7\) m, \(v_A\sim10^6\) m/s and \(\eta\sim1\) m\(^2\)/s, so

\[ S \sim \frac{10^7\times10^6}{1} = 10^{13}. \]

The diffusion time \(\tau_D = L^2/\eta \sim 10^{14}\) s is millions of years. In ideal MHD, the topology of the coronal field can never change.

Yet solar flares release stored magnetic energy in minutes. Something must break the frozen-in condition. That something is magnetic reconnection.

Practice problems

Try each problem before opening its solution. Click an option to check a concept question; type a number and press Check for a numerical one (forms like 6.2e7 or 6.2x10^7 both work).

Concept questions

Problem 1

What does \(R_m\gg1\) imply?

  • The magnetic field diffuses through the plasma quickly
  • Field lines move with the plasma (frozen in)
  • The plasma is collisionless
  • The magnetic field is weak

\(R_m\) compares advection to diffusion. When it is large, diffusion is negligible and the field is carried with the flow.

Problem 2

If the system size \(L\) doubles, how does the magnetic diffusion time \(\tau_D = L^2/\eta\) change?

  • Doubles
  • Quadruples
  • Halves
  • Unchanged

\(\tau_D\propto L^2\). Large systems diffuse very slowly, which is why astrophysical fields are so well frozen in.

Problem 3

Which speed appears in the Lundquist number \(S = Lv/\eta\)?

  • The flow speed
  • The Alfvén speed
  • The sound speed
  • The speed of light

\(S\) is the magnetic Reynolds number evaluated with the Alfvén speed \(v_A = B/\sqrt{\mu_0\rho}\), the natural speed of magnetic disturbances.

Show that…

Problem 4: the induction equation

Starting from \(\mathbf{E} + \mathbf{v}\times\mathbf{B} = \mathbf{J}/\sigma\), Faraday's law and Ampère's law (with uniform \(\eta\)), derive \(\partial_t\mathbf{B} = \nabla\times(\mathbf{v}\times\mathbf{B}) + \eta\nabla^2\mathbf{B}\).

Hint

Take the curl of Ohm's law and use \(\nabla\times(\nabla\times\mathbf{B}) = \nabla(\nabla\cdot\mathbf{B}) - \nabla^2\mathbf{B}\).

Proof

\(\mathbf{E} = -\mathbf{v}\times\mathbf{B} + \eta\mu_0\mathbf{J} = -\mathbf{v}\times\mathbf{B} + \eta\nabla\times\mathbf{B}\). Then \(\partial_t\mathbf{B} = -\nabla\times\mathbf{E} = \nabla\times(\mathbf{v}\times\mathbf{B}) - \eta\nabla\times(\nabla\times\mathbf{B})\), and with \(\nabla\cdot\mathbf{B} = 0\) the last term is \(+\eta\nabla^2\mathbf{B}\).

Problem 5: the ratio of timescales

Show that the diffusion time divided by the Alfvén crossing time equals the Lundquist number: \(\tau_D/\tau_A = S\).

Hint

\(\tau_D = L^2/\eta\) and \(\tau_A = L/v_A\).

Proof

\(\tau_D/\tau_A = (L^2/\eta)/(L/v_A) = Lv_A/\eta = S\).

Numbers

Problem 6

What is the Alfvén speed in the corona for \(B = 0.01\) T and a proton density \(n = 10^{15}\) m\(^{-3}\)? Give it in km/s.

Solution

\(\rho = nm_p = 1.67\times10^{-12}\) kg/m³, \(v_A = B/\sqrt{\mu_0\rho} = 6.9\times10^6\) m/s.

Problem 7

How long would it take a coronal field with \(L = 10^7\) m and \(\eta = 1\) m²/s to diffuse away? Give it in years.

Solution

\(\tau_D = L^2/\eta = 10^{14}\) s \(\approx 3.2\times10^6\) years.

Problem 8

A laboratory plasma has \(L = 0.1\) m, \(v_A = 5\times10^4\) m/s and \(\eta = 5\) m²/s. What is its Lundquist number?

Solution

\(S = Lv_A/\eta = 0.1\times5\times10^4/5 = 1000\).

Problem 9

The Spitzer magnetic diffusivity is roughly \(\eta\approx10^9\,T^{-3/2}\) m²/s (with \(T\) in K). What is \(\eta\) in the chromosphere, where \(T = 10^4\) K?

Solution

\(\eta\approx10^9/(10^4)^{3/2} = 10^9/10^6 = 1000\) m²/s: a thousand times more diffusive than the corona.