In large samples. the gamma-ray count rate of a prompt gamma neutron activation analysis system is a multi-variable function of the elemental dry composition, density, water content and thickness of the material. The experimental calibration curves require tremendous laboratory work, using a great number of standards with well-known compositions. Although a Monte Carlo simulation study does not avoid the experimental calibration work, it reduces the number of experimental calibration standards. This paper is part of a feasibility study for a PGNAA system for on-line continuous characterisation of cement raw material conveyed on a bell [Oliveira, C., Salgado, J. and Carvalho, F. G. (1997) Optimisation of PGNAA instrument design for cement raw materials using the MCNP code, J. Radioanal. Nucl. Chem. 216(2), 191-198; Oliveira, C., Salgado, J., Goncalves, I. F., Carvalho, F. G. and Leitao, F. (1997a) A Monte Carlo study of the influence of geometry arrangements and structural materials on a PGNAA system performance for cement raw materials analysis. Appl. Radiat. Isot. (accepted); Oliveira, C., Salgado, J. and Leitao, F. (1997b) Density and water content corrections in the gamma count rate of a PGNAA system For cement raw material analysis using the MCNP code. Appl. Radiat. Isot. (accepted).]. It reports on the influence of the density, mass water content and thickness on the calibration curves of the PGNAA system. The MCNP-4A code (Briesmeister, 1993), running in a Pentium-PC and in a DEC workstation, was used to simulate the PGNAA configuration system. (C) 1998 Elsevier Science Ltd. All rights reserved.
