J.R. Kirtley et al., Nature 373, 225 (1995).
Here is a beautiful false color scanning SQUID image (red=high B, blue=low B) from John Kirtley, Chang Tsuei and Mark Ketchen of IBM showing the magnetic field in four split YBCO rings. Only the central ring is composed of three junctions, such that in a d-wave superconductor, the three crystal interfaces would lead to a net phase shift of 180 degrees around the ring. This leads to spontaneous currents creating a 1/2 quantum of flux in the ring, as observed.

Has side-by-side images of 2 tricrystal ring samples with two geometries, the left which should show the 1/2 flux quantum effect in the center ring for a d-wave superconductor, the right which should not for any gap function with inversion ( x = -x, y = -y ) symmetry. The left does, the right doesn't. Also from Nature paper.

Shows the d-wave symmetry tricrystal ring sample, cooled down in a field of about 15 mG, such that the control 0-junction ring has 3 flux quanta in it, the 2 2-junction rings each have 2 flux quanta in them, and the 3-junction ring in the center has 1 1/2 flux quanta in it. This image was included in the conference proceedings for the Miami conference.

This image of magnetic field variations was taken with a Scanning Hall Probe Microscope. A tiny sensor of magnetic field is moved over the surface of a sample, allowing a computer to "see" the magnetic field. The spots correspond to individual magnetic l ines or vortices that are formed in a superconducting film of LaSrCaCuO. The net magnetic flux from each spot is a fundamental constant can be deduced directly from such an image. It agrees with nature's single magnetic flux quantum of hc/2e. This illus trates that magnetic flux in superconductors is "quantized".

Superconducting Flux Lattice