Page: 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18



Figure 8. Vt vs. frustration for chip H7#43. The threshold voltage varies periodically with frustration. The peaks at f=-0.87, -0.3 and 0.28 are due to a second SQUID loop size. The threshold decreases with increasing temperature but the peaks are still present at 100 mK.

Figure 8 shows the periodic dependence of Vt at four different temperatures. Higher temperature gives lower Vt , but the periodic behavior is still clear at 100 mK. However, there are a few peaks that appear with a period of 0.57 frustration units. These peaks are most likely due to a second effective loop size at the top and bottom of the array, and in a zero-bias resistance (R0) measurement on chip H9#44, extra peaks appear at the same frustration values (see figure 14). They look a little shifted to the left, and even more so in the R0 diagram, but this could just be due to measurement errors or too few sampling points.

Some histograms show more than one peak (see figure 9). Multiple peaks appear at temperatures lower than 100 mK, but other than that there is no clear pattern. In figure 9 the second peak is always smaller than the first, but that is not a general rule. Triple peaks have even been observed. The distance between the peaks, Vt, is sometimes hard to determine, but it seems that it is dependent on magnetic field. In figure 10, Vt has been measured at two temperatures and seven magnetic fields, spanning a frustration slightly less than 1. It is clear that Vt changes with B, but whether or not it is periodic cannot be deduced, nor if the period is the same as the mean Vt period, 1.6 mT.

10

<===Previous page ooo Next page ===>

My homepage.