An application of computational fluid dynamics (CFD) to the characterisation and optimisation of a cyclonic spray chamber for CP-AES

This paper describes the application of computational fluid dynamics (CFD) to the modelling of the performance of a cyclonic spray chamber with flow spoiler designed to provide good efficiency with rapid wash-out time. The modelling is carried out under two flow regimes, one corresponding to typical sample introduction conditions and one to low sample uptake. Velocity distributions, pressure distributions and mass flows are simulated, the latter employing the measured primary aerosol distributions as inputs to the computation. The dynamic performance of chambers is also modelled and the signal profiles produced are shown to correspond to those derived directly from measurement. Six different chambers are studied corresponding to a chamber with no spoiler, a single spoiler in four different positions, and a chamber with 3 spoilers. The position of the spoiler is shown to be critical in determining the flow field in the chamber and hence the efficiency and washout time. It is demonstrated that chamber geometry has maximum impact under low uptake conditions where the limited dispersion primary aerosol can effectively follow the gas flow lines. Computation indicates that a chamber with optimised spoiler, operating under low uptake conditions, provides the best combination of efficiency and response time. This was confirmed by experiment in which the performance of 3 different chambers was compared, with the optimised design providing detection limits 1.5– 19 better than those obtained with the chamber supplied with the ICP-AES instrument. A chamber with 3 spoilers indicates the formation of a low-volume