X-ray PMTs reveal more than 60 keV electron energies, triple pulse structure

Posted by Rezwan on Aug 09, 2010 at 09:16 AM
Tools: Print | | (5) Comments | Discuss In Forums

The three X-ray PMTs have now become functional, thanks to a large reduction of their noise pickup.  Our two visiting students helped by replacing the aluminum tape we had used for an electromagnetic shield with ordinary aluminum foil, which we had used in Texas.  While the foil looks less neat than the tape, it had far higher conductivity and that is what makes a good shield. 

One recalcitrant PMT’s noise was tracked down by Dr. Subramanian to a faulty cable which was replaced.  Some noise still travels through the grounding system, but since it takes an indirect route, it arrives too late to interfere with either the X-ray or neutron signals.

Two of the PMTs have copper filters, one 0.5 mm and the other 3 mm, while one is unfiltered.  This allows us to look at the ratios between the signals and thus measure the energy of the X-rays.  This in turn gives us an idea of the energy of the electrons which produce them.

By comparing the signals in some shots where the filters were the same on all PMTs, we have accomplished an initial cross-calibration of the instruments.  Since the signals were so strong they went beyond the instrument’s linear regime, where output is proportional to input, we also had to measure the saturation curve, or how the sensitivity declined with increasing input.  While more work is needed on this, our initial results are extremely interesting and encouraging.

We analyzed in detail one shot, number 6 on July 6, where the current and gas pressure were correctly matched (680 kA and 7 torr).  In Figure 2, we show the calculated average electron energies plotted against time, and the signals from the three PMTs.  The purple line is the average electron energy in keV calculated from the 3 mm signal’s ratio to the unfiltered signal, and the green line is the same thing calculated from the 0.5 mm signal.  The bold red line is the signal from the unfiltered instrument, the total X-ray signal, in volts, and the thin black and orange lines are the signals from the 0.5 mm-filters PMT and 3mm filtered one respectively.  These signals are corrected for saturation but are not to scale.  The red one is actually much larger than the other two at all times.  The time is in nanoseconds.  (For reference, 100 keV is the equivalent of a temperature of 1.1 billion degrees C.)  The total neutrons from this shot are 1.8x10^10 and the peak X-ray power is about 0.5 MW.

Average electron energies (in keV) over time (in nanoseconds).

Several things are notable from this data.  First, there is a difference in the two estimates of electron temperature.  This is not a contradiction, as the energy estimates are based on the assumption of a Maxwellian or random distribution of energies, which is probably not the case. The data shows that there is, after 20 ns or so, a high energy “tail’ to the distribution, fast electrons that are not entirely mixed with the others.  This is probably evidence of the electron beam electrons themselves.
Second, there is a characteristic three-peak structure which we see in many of the shots now observed with our improved instrumentation.  A small peak appears first, followed by two much larger ones.  Our data show that it is the third peak that produces the neutrons.  The total X-ray signal shows that this third peak is by far the largest and must be produced by a high-density plasma when the plasmoids are produced.  What is interesting is that electron energy seems to drop by a factor of two as the plasmoid forms in this shot, perhaps signaling the transfer of energy from the electrons to the ions, which then undergo fusion reactions.  Much more data is required to see if this pattern is universal, but it is a tantalizing clue that this same, “ready, set, go” three-pulse pattern is seen in the X-rays emitted from some large solar flares, which we know to be one of the natural models of plasmoid formation.

Third, the energies shown here are very high, more than 60 keV at the height of the pulse.  Interestingly, we can use the PMT’s data on the neutron pulse, together with the FTF and NTF neutron detectors’ data to determine the average ion energy, which turns out to be, within calculated errors, the same 70 keV.  From these numbers, we can calculate the product n^2V (where n is density and V is volume) for both the electrons and the ions.  Both numbers are the same, 1.1+-0.1x10^35/cm^3.  This is a strong indication that the X-rays and neutrons come from the same plasma, the plasmoid.  It is also a good indication of how our instruments are designed to supplement and confirm each other, so that every measurement we take will be checked by at least two instruments.