Since the last update we've been quite busy, just because the experiment is going well at this point. Since the last update we've continued our program of controlled cooling towards Tc, stopping at selected temperature points along the way. By this time we've completed slow ramps to temperatures 100mK, 56mK, 30mK, 18mK, 10mK, 5.6mK, 3mK, and 1.8mK above the critical temperature. For the last ramp we've gone at our slowest cooling rate yet of .53mK/hr.

At most of these temperatures we've been collecting correlogram data, and we have some of those to show you here. These are all single examples of dual correlograms (each correlogram contains both forward scattering and backscattering data), and we take about 15 dual correlograms at each temperature. In all cases the faster-decaying curve is the backscattering correlogram (at a 168-degree scattering angle), and the slower-decaying curve is the forward-scattering correlogram (at a 12-degree scattering angle). As we get close to Tc, you'll notice that the curves decay ever more slowly, most noticably in the forward-scattering direction (this represents the phenomenon known as critical-slowing down).

This first curve shows a dual correlogram at Tc + 100mK:

And next we have a dual correlogram at Tc + 56mK:

Already you can see the difference between the decay times of the two forward-scattering correlograms (not the time when the curve reaches its half-amplitude, and notice that the horizontal scale is logarithmic--yes, these curves are single exponentials!).

And one more example, this being a dual correlogram taken at Tc + 30mK:

Here we see additional significant slowing down in the forward scattering curve.

Interference Phenomena

And now for the tricky one. This is a real pretty graph coming up, but I don't know whether my mind is clear enoguh at this late hour to explain it. Let's try though.

Our sample of xenon is a 100-micron thin disk of fluid, contained between two very thick windows. We measure the transmission of the fluid (to get to a value for its turbidity) by measuring the ratio of the intensity of the laser light before it enters the cell to the intenstiy of the light after it's been through the cell. The latter value is always less. In part it is less because of any inherent opacity of the fluid (the interesting bit, which increases as we approach Tc), because of opacity of the windows (a constant contribution), and because of an interference effect which arises because each surface of each window (four surfaces) reflect small amounts of light.

These bits of beam reflected from the window surfaces interfere with each other and make a predictable reduction in the overall transmission of the cell. Normally this would be a constant contribution and of no interest.

However, Zeno is far from normal. It turns out that as we change the temperature of the sample cell it contracts (going towards Tc) by minute amounts (about 1/2 a wavelength of light for each Kelvin temperature change!), and as this spacing between the windows change, the contribution to the transmission of the cell due to the interference effet changes, and our transmission measurements are quite sensitive enough to see the effect.

To demonstrate it, here is the interference curve that we have measured so far on orbit:

Rather than being a single sinusoid, the curve is more complicated because there are 4 window surfaces involved and the resulting intereference possibilities are greater. Those little wiggles at the right edge of the curve, by the way, are the actual contributions from the fluid turbidity, exhibiting transient behavior during temperature changes.

Now, the turbidity measurements near Tc of the sample are an important measurement for Zeno, and the changes we are looking for there are but a fraction of this overall interference effect. However, this curve also represents changes over a nearly 8-Kelvin interval, whereas our most important measurements take place over a fraction of a Kelvin. So, in some ways, the effect is more or less constant over our range of intereste, but it is very important that we know where on this interference curve the data are taken.

This curve is the best calibration curve over the largest range that we've ever obtained on this instrument, so we'er pretty happy with it. We're also busy modelling the effect and the model now just about reproduces the entire curve in detail. Overall, a good report.

And your scribe humbly hopes that he's explained it clearly enough for you to get some sense of what he's talking about.