Journal of Radioanalytical and Nuclear Chemistry, Vol. 271, No.1 (2007) 145–150 0236–5731/USD 20.00 Akadémiai Kiadó, Budapest © 2007 Akadémiai Kiadó, Budapest Springer, Dordrecht Automatic activation analysis* G. P. Westphal,** F. Grass, H. Lemmel, J. Sterba, P. Schröder, Ch. Bloch Atominstitut der Österreichischen Universitäten, Stadionallee 2, A-1020 Vienna, Austria (Received April 13, 2006) Automatic activation analysis (AAA) is rendered possible by a unique neutron activation analysis facility for short-lived isomeric transitions based on a fast rabbit system with sample changer and sample separation, and an adaptive digital gamma-spectrometer for very high counting rates of up to 10 6 cps. The system is controlled by a computer program performing irradiation control, neutron flux monitoring, and gamma-spectrometry with real-time correction of counting losses, spectra evaluation, nuclide identification and calculation of concentrations in a fully automatic procedure. As spectrometry is done by means of hundreds of sequentially measured pairs of concurrently recorded loss-corrected and non-corrected spectra, concentrations are derived from an optimally weighted average of all individual occurrences in this sequence of spectra which also enable the separation of isomeric transitions with coinciding energies but different half-lives such as 116m2 In (162.4 keV, T 1/2 = 2.2 s) and 77m Se (162.2 keV, T 1/2 = 17.4 s). To clear up repeatedly voiced misconceptions concerning the errors of loss-free counting our findings of 1978 and 1981 are reiterated, namely that the counting error of a peak in a corrected spectrum may be derived consistently from the error of the same peak in the respective non-corrected spectrum and from the error of weighting factors in the corresponding region of interest, according to the principle of propagation of errors. Experimental proof is provided for conditions of stationary as well as rapidly varying counting rates and spectral shapes. Introduction The fast rabbit system A simple but highly effective pneumatic transport system is made of an aluminum in-core part connected to inexpensive plastic tubing, not at last to avoid expensive bending of aluminum tubes. 1 Pressure comes from an industrial compressed air generator or, if available, from a central compressed air supply. Transport time at a pressure of 5 bars is 150 to 300 ms depending on sample size. Argon exhaust is collected in plastic balloons wherefrom it is removed by the Triga reactor’s beam tube ventilation system. An automatic sample changer with sample from transport rabbit separation has been already described elsewhere. 2 The sample changer as well as the rabbit system may be controlled either manually or by the computer. The spectrometry system As a necessary prerequisite for high count-rate gamma-spectrometry a high-purity germanium detector with a transistor reset preamplifier is the main component of the spectrometry system. Pulse processing is performed by the preloaded digital filter 3 (PLDF) developed at the Atominstitut which alternatively may be replaced by (less powerful) plug-compatible digital filters or analog filter-ADC combinations from Canberra. By automatically adapting the noise filtering time to individual pulse intervals, the preloaded digital filter 3 combines low- to medium-rate resolutions comparable * To the memory of Vincent P. Guinn. ** E-mail: westphal@ati.ac.at to those of high-quality Gaussian amplifiers with throughput rates of up to 100 kc/s, and high-rate resolutions superior to those of state-of-the-art gated integrator systems. In contrast to commercially available digital filters, the PLDF in its new implementation performs pulse shortening as well as pole zero cancellation in the analogue domain to render possible the use of cheap monolithic converters such as the AD9240 by analog devices. By feeding to the ADC very short signals as near as possible to the current pulses originating in the detector, pulse pileup and the consequent degradation of system resolution is completely avoided, even at elevated counting rates. At the same time, much better use may be made of the dynamic range of the converter as it is no longer used up by pulse pileup. Additional benefits are the much simpler digital circuit and conventional pole zero cancellation instead of digital pole zero cancellation. To further optimize the signal to noise ratio, a pseudo-rectangular pre-filter with a pulse duration of 1.5 μs has been chosen which makes possible a maximum throughput rate of 92 kc/s after pileup rejection. A specially designed parallel interface connects the digital filter to a fast PC. At the same time it serves for the generation of weighting factors for real-time correction of counting losses, with digital dead-time extension 1 or according to the recently developed method of computed pileup correction (CPC). 4 A software implementation of a loss-free counting multi-channel analyser is programmed in Assembler and C++ and, contrary to most commercial products, it runs in the MS-DOS environment, in order to save the computing power for real-time calculations. Storing the spectra immediately into the multi-megabyte memory of a Pentium type PC, it offers programmable rabbit