An innovative method for dead time correction in nuclear spectroscopy

All nuclear spectroscopy systems, whether measuring charged particles, X-ra ys, or gamma-rays, exhibit dead time losses during the counting process due to pulse processing in the electronics. Several techniques have been emplo yed in an effort to reduce the effects of dead time losses on a spectroscop y system including live time clocks and loss-free counting modules. Live ti me extension techniques give accurate results when measuring samples in whi ch the activity remains roughly constant during the measuring process (i.e. , the dead time does not change significantly during a single measurement p eriod). The loss-free counting method of correcting for dead time losses, a s introduced by HARMS and improved by WESTPHAL (US Patent No. 4,476,384) gi ve better results than live time extension techniques when the counting rat e changes significantly during the measurement. However, loss-free counting methods are limited by the fact that an estimation of the uncertainty asso ciated with the spectral counts can not be easily determined, because the c orrected data no longer obeys Poisson statistics. Therefore, accurate analy sis of the spectral data including the uncertainty calculations is difficul t to achieve. The Ortec((R)) DSPECPLUS(TM) implements an improved zero dead time method that accurately predicts the uncertainty from counting statist ics and overcomes the limitations of previous loss-free counting methods. T he uncertainty in the dead-time corrected spectrum is calculated and stored with the spectral data (Patent Pending). The GammaVision-32((R)) analysis algorithm has been improved to propagate this uncertainty through the activ ity calculation. Two experiments are set up to verify these innovations. Th e experiments show that the new method gives the same reported activity and associated uncertainties as the well-proven Gedcke-Hale live time clock. I t is thus shown that over a wide range of dead times the new ZDT method tra cks the true counting rate as if it had zero dead time, and yields an accur ate estimation of the statistical uncertainty in the reported counts.