The Philips QL-lamp: modelling and comparison with experiments

Abstract

Summary form only given. The Philips QL lightsource is a low pressure inductively coupled RF discharge. The main advantage of this system is the fact that it contains no electrodes so that the lifetime is longer. A characteristic feature is that the induction coil with a ferrite core is located at the centre of the lamp. The plasma simulation model developed at the Eindhoven university of technology is used to study the mercury-argon discharge of the QL-system. Basic assumptions of this model are: (1) steady state, (2) axial symmetry, (3) nonlocal thermal equilibrium, and (4) the plasma can be described by a hydrodynamical approach. The model than calculates distributions of the plasma and electron density, electron and heavy particle temperature (T/sub h/). The electro-magnetic field is calculated self-consistently, using the vector potential formalism, in the discharge as well as in the ferrite core. With the electron density (n/sub e/) and electric conductivity distributions the plasma resistance and self-induction coefficient can be determined. By calculating the geometric coupling coefficient the impedance of the primary circuit, which can be measured, is also computed. In the model calculations the electron kinetics is governed by mercury and argon is purely a buffergas, i.e. it determines the transport coefficients. A comparison is made between the experimental and calculated primary impedance, and radial profiles of electron density and heavy particle temperature. It seems that the calculations are quite sensitive for the data for ionisation and radiative losses (escape of the resonant 253.6 nm). With the current model assumptions and data for ionisation the calculated results for n/sub e/, T/sub h/, and primary impedance agree within 30% with the experimental results. It is expected that a better agreement is obtained when the effect of diffusion of mercury atoms is taken into account.