GENERAL PROCEDURES FOR RADIOASSAY 



175 



Standard mounts, which are used for end-window counters, have shelves 

 with definite geometrical relationships between them, usually so arranged 

 that each shelf has one-half the geometry of that above it. The geom- 

 etry, which primarily depends on the distance between source and counter 

 and the size of the source, is an important factor in the over-all sensitivity. 



Instrumental Counting Losses. When two particles arrive at a Geiger 

 counter tube within a time interval less than \hc resolving time of the 

 tube, they will be counted only as one, giving low results. This is called 

 the coincidence loss and increases as the counting rate increases. There 

 may also be losses in the electronic circuits and the mechanical register, 

 but these are not usually important if a high enough scahng factor is 

 employed. 



The over-all instrumental counting losses can be estimated in a prac- 

 tical manner by preparing a series of sources of known radioisotope con- 

 tent relative to each other. These samples should be measured under 

 identical conditions and should not differ between themselves in self- 

 absorption due to varying mass. The "true" value at each counting 

 rate can be calculated from the observed value at the low counting rate 

 and the known dilution. This is illustrated in Table 5-2 for a typical 



Table 5-2. Determination of Counting Loss in Commercial Scaler 



AND End-window Counter 



commercial scaler and end- window counter; the measurements were made 

 with P^'. It will be noted that the correction at 10,000 counts/min was 

 only 3.5 per cent and, of course, was smaller at the lower counting rates. 

 As a rule of thumb, a correction of 0.5 per cent per 1000 counts/min is 

 often used. At counting rates lower than 10,000 counts/min, then, the 

 coincidence correction can usually be neglected. Proportional counters 

 have the advantage that no such corrections are needed until counting 

 rates approach 100,000 counts/min. The data in Table 5-2 actually 

 represent a calibration curve, and, as will be pointed out in the next 

 section, it is possible to use a calibration procedure that minimizes still 

 further the need for coincidence corrections. 



