a simple derivation of magnetic reconnection, part 4
I have created a
web page that summarizes the derivation so far, with links to parts 1, 2, and 3.
To review:
- Part 1 and its erratum used one of Maxwell's equations to derive the magnetic field B around a current-carrying rod.
- Part 2 expressed that magnetic field in both cylindrical and Cartesian coordinates.
- Part 3 demonstrated that the magnetic field around four current-carrying rods reproduces
- Part 3 also showed that magnetic lines can begin or end at a neutral point, which I later proved in more detail.
In this part 4, I will use
the experiment I've been suggesting to Michael Mozina to illustrate magnetic reconnection and to prove that the topology of the magnetic field changes during a variation of that experiment. As Yamada et al explain in their appendix, that change in the topology of the magnetic field is what we mean by magnetic reconnection.
[size=+1]
The experimental setup[/size]
In a well-shielded laboratory, we assemble four long conducting rods parallel to the vertical (z) axis of a Cartesian coordinate system whose y axis points north and whose x axis points roughly east. We place each rod exactly 1 meter east, west, north, or south of the coordinate origin, and run 1000 amperes of current through each rod. That current runs upward (positive z) in the east and west rods, but downward (negative z) in the north and south rods. By Maxwell's equations (as was shown in parts 1 through 3), the magnetic field for the 4x4 meter square centered on the origin looks like this:
As before, the colors in that graph show the magnitude of the magnetic field, ranging from white (most intense) to black (least intense). The gray lines are the magnetic field lines that run through points of the form <±1,y,0> or <x,±1,0> for x,y∈{±1.0, ±0.9, ±0.8, ±0.7, ±0.6, ±0.5, ±0.4, ±0.3}.
The four rods' individual magnetic fields cancel each other out at the origin, so the magnetic field is zero there. Within the xy plane at z=0, the origin is the only neutral point of this particular magnetic field
B4.
[size=+1]
Magnetic reconnection at the neutral point[/size]
Without changing the 1000 ampere current within the east and west rods, let's vary the current within the north and south rods between 999.9 and 1000.1 amperes. That tiny variation in current causes magnetic reconnection at the neutral point:
[qimg]http://www.cesura17.net/~will/Ephemera/Nerdliness/MR/mr100.gif[/qimg]
When that animated GIF is running, you'll see two arrows flash briefly when the current within the north/south rods returns to its original value of exactly 1000 amperes. At that infinitesimal instant, the two magnetic field lines that are almost touching divide into four field lines that begin or end at the neutral point. That division and subsequent merger of magnetic field lines can only occur at a neutral point.
The next animation starts with the original magnetic field (
B4), slowly reduces the north/south rods' current to zero, and then increases that current back to the original 1000 amperes:
[qimg]http://www.cesura17.net/~will/Ephemera/Nerdliness/MR/mr00.gif[/qimg]
Wikipedia's
animation of magnetic reconnection shows essentially the same phenomenon, but is rigged to run forever without any change in the magnetic field. My animation above shows what magnetic reconnection actually looks like in a theoretically simple magnetic field that's easy to reproduce experimentally.
[size=+1]
Mathematical proof of topological changes[/size]
Reducing the north/south rods' current to zero yields a magnetic field for which
- Every field line either begins at the neutral point, ends at the neutral point, or loops back on itself.
- Given any 3 magnetic field lines that loop back on themselves, at least one of the 3 field lines lies within the loop formed by another of the 3 field lines.
Those are topological properties of the magnetic field lines, which means those properties are preserved by any
homeomorphism of the magnetic field lines.
The original magnetic field has the first of those two topological properties, but does not have the second property. For the original magnetic field, you can have 4 magnetic field lines that loop back on themselves without any of the 4 loops being contained within any of the other loops (in that set of 4).
That implies that the topology of the magnetic field lines must change as the current within the north/south rods is increased from zero to 1000 amperes.
For reasons that should be clear from the links above, that change in the topology of the magnetic field lines is called magnetic reconnection [Yamada et al 2010].
[size=+1]
Conclusion to part 4[/size]
We have now completed a mathematical proof that magnetic reconnection is a direct consequence of Maxwell's equations. For the most part, that proof used nothing beyond freshman-level vector calculus and electromagnetism. At the very end of the proof, when we proved that the topology changes, we needed a few additional facts about topology: the definition of homeomorphism and topological properties, and the fact that one of the properties listed is a topological property.
We have proved that magnetic reconnection can occur in vacuo, at a neutral point spatially separated from all currents, and with a negligible
E field.
In part 5, I will explain the two main points of [Dungey 1958] by adding a little plasma to the magnetic field
B4. It is not really possible to understand Dungey's paper without first understanding the neutral point and magnetic field lines of
B4 or similar fields in a vacuum.
[size=+1]
References[/size]
[Dungey 1958] J W Dungey.
The neutral point discharge theory of solar flares. A reply to Cowling's criticism. Proceedings of
Electromagnetic Phenomena in Cosmical Physics, edited by Bo Lehnert. International Astronomical Union number 6, Cambridge University Press, page 135.
[Yamada et al 2010] Masaaki Yamada, Russell Kulsrud, and Hantao Ji. Magnetic reconnection.
Reviews of Modern Physics volume 82, January-March 2010, pages 603-664.