People often ask me about colourful pictures. Below is a selection which I find suitable for talks to the general (and not so general) public. They have been created a long time ago together with Christoph Schlichter. By clicking at this Link you can also download all of them in one 1.6 MB gzipped tar file, in ps and jpg format, together with a little README. And yes.. I am indeed planning to upgrade them sometime in the far future. (As of 1 September 2005 there will be an update in the very near future). Feel free to use them as they suit you (hopefully acknowledging the source).
This is the famous action density distribution between two static colour sources (e.g. infinitely heavy quarks) from a 1994 simulation of SU(2) gauge theory on a 32^4 lattice. Nowadays one should be able to produce similar results in a finite amount of time on a laptop. Those days we used a so called Connection Machine CM-2. The lattice spacing corresponds to 0.08 fm and the sources are separated by 18 such lattice units. The mesh is not equidistant since in the orthogonal direction we have determined the distribution at integer distances as well as at multiples of sqrt(2). The colour scale encodes the relative statistical errors (and is obviously correlated with the amplitude of the signal). As the lattice spacing goes to zero the two peaks will diverge. The original reference is Observing.... Please note that there are differences between the (final) PRD version and the eprint.
This is what happens when the sources are pulled apart. In real QCD with sea quarks the ``string'' would break around r = 1.1 fm. Obviously the field does not look like an electric dipole field but more ``stringy''.
One way of obtaining electric field lines is the Maximal Abelian Projection (MAP). The above example is obtained in SU(2) gauge theory at beta = 2.5 and with a separation of 12 lattice spacings, corresponding to 1 fm. The original references (with pictures of monopole currents etc) are A Ginzburg-Landau....
The same as the first figure at very slightly coarser lattice spacing (beta=2.5 instead of 2.5115) but after MAP.
The same as above for the energy density.
Again the action density after MA projection, this time for a separation of 16 lattice units or 1.3 fm.
The photon contribution to the action density. This resembles almost exactly an electric dipole field.
The respective monopole contribution giving rise to the linear confining potential. The first of these three figures is almost exactly the sum of photon and monopole contributions. Note that the normalisation of the vertical axis varies.
