The paper presents an improved approach for the measurement of Cu-dispersion in catalysts using the oxidation of the Cu surface atoms by N2O. During a single easy-to-perform experiment, the degree of surface and bulk oxidation of copper can be separated due to a continuous measurement of evolved N-2, which leads to a more accurate measurement of the copper surface. The measurement is made at ambient pressure with a constant flow of gas through a fixed bed of catalyst particles by switching the feed gas composition from pure helium to 2% N2O in helium. The effluent gas passes through a freeze trap with liquid nitrogen, which selectively removes any unreacted N2O from the gas. Nitrogen evolved by the oxidation reaction is not removed in the N2O trap and a binary mixture of nitrogen diluted in helium carrier gas is then measured accurately directly in a calibrated thermal conductivity detector (TCD). Five methanol synthesis catalysts (Cu/ZnO/Al2O3), which differ in terms of the copper surface area are studied by this method. All recorded TCD-signals consist of two features: first, a significant peak appearing immediately after the switch from He to N2O in He, which is attributed to a fast surface reaction. Secondly, the peak is followed by a long tail, which is due to slow formation of N2 due to a diffusion-limited oxidation of bulk copper atoms. During the tailing part of the experiment, the N2O-reactions is under differential condition and the produced N2 is continuously removed by the He carrier gas. By the new experimental procedure we can separate the two parts of the N-2-signal, i.e. a rapid surface oxidation and a slow solid state diffusion-limited bulk oxidation. As a result, we can - based on a mathematical correction for the influence of diffusion-make a more accurate measurement of the copper surface area. Measurements made at different temperatures show-as expected-an increase in bulk oxidation rate due to the temperature dependence of oxygen diffusion in the upper oxide layer. For precision it is recommended, though, to measure under conditions where the influence of bulk diffusion is small, that is below 90 degreesC. The method is fast, simple, sensitive and versatile and only requires a thermal conductivity cell for the analysis. (C) 2004 Elsevier B.V. All rights reserved.