Gas gain in proportional counters

Gas gain measurements were made for nitrogen, xenon and x~non with 2,3 dimethyl-2-butene (DMB) as an additive. Two different methods of measuring gas gain were examined, pulse matching methods and current comparison methods. A preliminary investigation was made into a new approach to the pulse matching method. Previous authors had shaped their test pulses to the same shape as their detector pulses or used a calibration factor to correct for the difference between the shape of the test pulses and the detector pulses. These approaches would be valid only if the detector pulses were always the same shape. The pulses from the detector vary in shape depending upon where the primary ion pairs are formed in the detector sensitive volume. The new approach' I have investigated shapes the detector pulses and the test pulses to the same shape so that the variation in the detector pulse shape becomes irrelevant. This method of measuring gas gain was tested using P-10 and 148Gd as an alpha particle source and gave similar results to gas gains determined by current comparisons. The measurements of gas gain finally used to obtain the data were all performed using the current comparison technique. When the electrons were collected at the central wire and 55Fe was used as the radiation source the ion saturation curves for xenon showed no definite plateaus, only points of inflexion. I found that ion saturation currents could be determined by fitting the observed currents and associated operating voltages to an expression proposed by Johnson. This expression also enabled me to determine the voltage region where gas gain began. The gas gain characteristics of nitrogen and xenon were measured for a range of gas densities. From these measurements it was shown that the gas gain of a cylindrical proportional counter operating in the proportional region is a function only of the reduced field strength at the surface of the anode. Aoyama has proposed an expression for the first Townsend coefficient and he has been able to show that previously proposed expressions are only special cases of his expression. I was able to obtain an excellent fit to Aoyama's expression using my gas gain data for xenon and a value for one of the parameters determined by Kowalski. It is widely held that the mechanism of gas gain in proportional counters is primarily by electron impact. Ionization may also occur in binary mixtures by the non-metastable Penning effect, and the metastable Penning effect. A binary mixture of xenon and DMB was used to investigate these effects. It was found that the gas gain in xenon could be greatly increased by the addition of a small quantity of DMB. A broad peak in gas gain appeared when the concentration of additive was between ~ 0.4% and ~ 0.75% and the molecular number density of the gas filling was 2.7 x 10 [to the power] 25 m-3 (~ 1 atm). When the gas density was reduced to 1.4 x 10 [to the power] 25 m-3 (~ 0.5 atm) the peak became sharper and better defined and reached its maximum when the concentration of additive was ~ 0.43%. Lowering the density to 7 x 10 [to the power] 24 m-3 (~ 0.26 atm) resulted in a very sharp well defined peak centred on an additive concentration of ~ 0.23%. The gas gains used to define these peaks were of the order of 10[ to the power] 5, whilst in pure xenon at identical anode potentials the gas gain was of the order of 10. The mechanisms proposed for these increased gas gains were the non-metastable Penning effect and the metastable Penning effect in the case of high density mixtures, the Pennning effect in the' case of the intermediate density mixtures and for the low density mixtures it was suggested that doubly excited xenon molecules were ionizing the additive. The W-values of the low density xenon plus DMB mixtures were also measured to 5% accuracy. The trend in the mean values of these measurements indicates that the Penning effect is occuring and the minimum value in the W-value occurs at ~ 0.4% which suppports the above proposed mechanisms for the increased gas gains. Detectors filled with mixtures of xenon plus large concentrations of DMB developed double full energy peaks in a relatively short amount of time using the 55Fe source. Passing a large current through the anode removed these double peaks demonstrating that the cause was uneven deposits on the anode. These deposits were probably due to polymerization of free radicals formed in the electron avalanche. Small percentages of DMB did not result in double peaks but drops in gain were observed when the detector was subjected to prolonged usage. The absence of double peaks is attributed to the diffuse nature of the avalanches. If the Penning effect or ionization of the additive by UV photons is taking place the avalanche activity would always completely surround the anode giving rise 'to uniform coatings around the entire circumference of the anode.