Investigation of the cathodic region of a fluorescent lamp

Abstract

In this work the absolute intensity of barium lines during dimming operation in the vicinity of fluorescent lamp electrodes is investigated using the method of atomic emission spectroscopy. Investigations of fluorescent lamps (FL) are often focused on the electrodes, since the lifetime of the lamps is typically limited by the electrode lifetime and durability. In general, a commercial electrode system consists of a tungsten coil coated with a work function reducing emitter mix of alkali oxides, such as BaO, SrO and CaO. One of the main issues with dimming is a reduction in lamp life when the additional heating of the electrodes is not optimal, causing not-optimal electrode temperatures. There is a need for design rules for dimming to reach "good" lamp life, based on "good" additional heating. Such dimming design rules can be derived from measurements of the absolute intensities of both neutral (Ba I) and ionised (Ba II) barium lines, as these are key parameters to describe the evaporation and the sputtering of the emitter material. During steady state operation free barium is produced, which transports to the surface by diffusion through the coating mass. Barium escapes from the emitter during the course of lamp operation due to sputtering (primarily during starting and dimming), where the discharge is sustained by secondary electron emission from the (cold) electrode, and due to evaporation (primarily during steady-state), where the discharge is sustained by thermionic emission from the (hot) electrode. In the first type of experiment, the atomic emission diagnostic is used for the detection and measurement the intensities of the neutral (Bal-553.5 nm) and ionised (Ball-455.4n m) barium. The FL is dimmed for a range of discharge currents and auxiliary coil heating currents. It is seen that there is a threshold discharge current in which the behaviour of barium intensity with respect to current is markedly different. In a second type of experiment, we can estimate the relative intensities for Ba I and Ba II from the relative area under their line of absolute intensities. From these two measurements we show that the Ba loss can very easily be reduced by appropriate auxiliary coil heating.