C Ruhrmann

Space-resolved characterization of high frequency atmospheric-pressure plasma in nitrogen, applying optical emission spectroscopy and numerical simulation

Averaged plasma parameters such as electron distribution function and electron density are determined by characterization of high frequency (2.4 GHz) nitrogen-plasma using both experimental methods, namely optical emission spectroscopy (OES) and microphotography, and numerical simulation. Both direct and stepwise electron-impact excitation of nitrogen emissions are considered. The determination of space-resolved electron distribution function, electron density, rate constant for electron-impact dissociation of nitrogen molecule and the production of nitrogen atoms, applying the same methods, is discussed. Spatial distribution of intensities of neutral nitrogen molecule and nitrogen molecular ion from the microplasma is imaged by a CCD camera. The CCD images are calibrated using the corresponding emissions measured by absolutely-calibrated OES, and are then subjected to inverse Abel transformation to determine space-resolved intensities and other parameters. The space-resolved parameters are compared, respectively, with the averaged parameters, and an agreement between them is established.

Investigating the gas phase emitter effect of caesium and cerium in ceramic metal halide lamps in dependence on the operating frequency

The work function and with it the temperature of tungsten electrodes in HID lamps can be lowered and the lifetime of lamps increased by the gas phase emitter effect. A determination of the emitter effect of Cs and Ce is performed by phase resolved measurements of the electrode tip temperature Ttip(), plasma temperature Tpl() and particle densities N() by means of pyrometric, optical emission and broadband absorption spectroscopy in dependence on the operating frequency. The investigated HID lamps are ceramic metal halide lamps with transparent discharge vessels made of YAG, filled with a buffer gas consisting of Ar, Kr and predominantly Hg and seeded with CsI or CeI3. In the YAG lamp seeded with CsI and CeI3 as well as in a YAG lamp seeded with DyI3 (corresponding results can be found in a preceding paper) a gas phase emitter effect is observed in the cathodic phase due to a Cs, Ce or Dy ion current. In the YAG lamp seeded with CsI the phase averaged coverage of the electrode surface with emitter atoms decreases and the electrode temperature rises with increasing frequency, whereas the emitter effect of Ce and Dy is extended to the anodic phase, which leads to a decreased average temperature Ttip() with increasing frequency. This different behaviour of the averaged values of Ttip() for increasing frequency is caused by the differing adsorption energies Ea of the respective emitter materials. In spite of the influence of Ea on the coverage of the electrode with emitter atoms, the cathodic gas phase emitter effect produces in the YAG lamps seeded with CsI, CeI3 and DyI3 a general reduction in the electrode tip temperature Ttip() in comparison with a YAG lamp with Hg filling only.

Simulation of the hot core mode of arc attachment at a thoriated tungsten cathode by an emitter spot model

Recently, a constricted attachment of an atmospheric pressure low-current argon arc in the centre of the flat end face of a thoriated tungsten cathode was observed and spectroscopically analysed. Its diameter of 0.6mm and its length of the free standing part of 10mm are the typical dimensions of electrodes for high-intensity discharge lamps. This paper gives a physical interpretation of the axially symmetric arc spot by a simulation of its properties with a cathodic sheath model which takes into account a reduction in the work function above a critical temperature of the cathode surface by a thorium ion current. At first the optical observation and spectroscopic investigations are recapitulated. Then, an overview is given on the essential elements which are needed to simulate the cathodic arc attachment on a hot electrode. A simulation of a central cathode spot with these elements gives results which are far away from the experimental findings if a constant work function φ is used. Therefore, a temperature-dependent work function φ(T )is introduced. This φ(T )transitions from 4.55 to 3eV above temperatures of the order of 3000K. With this emitter spot model a constricted arc attachment is obtained by simulation in the centre of the flat end face of the cathode in accordance with experiment. For currents below iarc,max ≈ 15.5A, two spot solutions with different cathode falls are found. They form a current‐voltage‐characteristic consisting of two branches which extend from a turning point at iarc,max to lower currents. For iarc >i arc,max, only a diffuse mode of cathodic arc attachment is obtained. It is shown by a comparison with measured data for iarc = 7.5,10,12.5 and 15A that the solution with the lower cathode fall is observed experimentally.

Combining emission and absorption spectroscopy at rare earth spectral lines: plasma temperature measurements in ceramic metal halide lamps

Presently, most high intensity discharge (HID) lamps contain mercury to generate a high pressure buffer gas and thereby an appropriate power input into the arc. Due to its toxicity, the replacement of Hg is of particular interest in recent research on HID lamps. Up to now, the emission coefficient of an atomic Hg double line is widely used to determine the plasma temperature Tpl in HID lamps. Tpl is needed to calculate the total density of atoms and ions of elements inside these lamps. A combination of optical emission and broadband absorption spectroscopy allows us to evaluate Tpl independently of Hg emission lines. The method is required for a determination of Tpl if the Hg line intensity within the investigated lamp is too low, is superimposed by other lines or if environmental-friendly Hg-free lamps are developed.Within this work, phase-resolved plasma temperatures are determined in front of the electrode of Hg-containing MH lamps by emission spectroscopy at atomic Hg lines. Above all, temperatures are measured by a combination of emission and absorption spectroscopy at atomic rare earth lines, namely Dy and Tm. A comparison of Tpl determined by both methods agree within an error margin of <10%. Total phase-resolved rare earth atom densities are obtained by means of the measured ground state densities and Tpl. The combination of emission and absorption spectroscopy is also applied to the bulk plasma of lamps where the intensity of the Hg emission lines is too low for plasma temperature measurements or Hg is absent. It provides the partial rare earth pressure and by comparison with thermodynamic data cold spot temperatures within the lamps.

Investigating the dependence of the temperature of high-intensity discharge (HID) lamp electrodes on the operating frequency by pyrometric measurements

Phase-resolved temperature distributions are determined along a rod-shaped tungsten electrode, by which an ac arc is operated within a model lamp filled with argon. Switched dc and sinusoidal currents are applied with amplitudes of several amperes and operating frequencies being varied between 10 Hz and 10 kHz. The temperature is deduced from the grey body radiation of the electrode being recorded with a spectroscopic measuring system. Phase-resolved values of the electrode tip temperature Ttip and of the power input Pin are determined comparing the measured temperature distributions with the integral of the one-dimensional heat balance with these parameters as integration constants. They are supplemented by phase-resolved measurements of the sum of cathode and anode fall called the electrode sheath voltage. If a switched dc current is applied it is found that both quantities are within the cathodic phase only marginally higher than for a cathode being operated with a dc current. Ttip and Pin start to decrease for low currents and to increase for high currents at the beginning of the anodic phase. But with increasing operating frequency the deviations from the cathodic phase are reduced until they cannot be resolved for frequencies of several kHz. A more pronounced modulation, but the same tendencies, is observed with a sinusoidal current waveform. For 10 kHz a diffuse arc attachment with an almost phase-independent electrode tip temperature, which deviates only marginally from that of a dc cathode, and an electrode sheath voltage proportional to the arc current is established with both current waveforms.

Investigation of the flickering of La2O3 and ThO2 doped tungsten cathodes

Short-arc lamps are equipped with tungsten electrodes due to their ability to withstand a high thermal load during operation. Nominal currents of more than one hundred amperes lead to a cathode tip temperature near the melting point of tungsten. To reduce the electrode temperature and, thereby, to increase the maintenance of such lamps, ThO2 or tentatively La2O3 are added to the electrode material. They generate a reduced work function by establishing a monolayer of emitter atoms on the tungsten surface. Emitter enrichments on the lateral surface of doped cathodes are formed. They are traced back to transport mechanisms of emitter oxides in the interior of the electrode and on the electrode surface in dependence of the electrode temperature and to the redeposition of vaporized and ionized emitter atoms onto the cathode tip by the electric field in front. The investigation is undertaken by means of glow discharge mass spectrometry, scanning electron microscope images, energy dispersive x-ray spectroscopy, ...

Investigating the outer-bulb discharge as ignition aid for automotive-HID lamps

This work considers the ignition process of mercury-free high-intensity discharge lamps used for car headlights. These lamps have to run-up fast. This is achieved with a high xenon pressure of about 15 bar (cold) in the inner bulb. The high filling-gas pressure causes an increased ignition voltage compared with lower-pressure lamps used in general-lighting applications. In this paper the possibility is investigated to reduce the ignition voltage by optimizing a dielectric-barrier discharge (DBD) in the outer bulb working as ignition aid. A special outer bulb was built up allowing gas exchange and adjustment of the gas pressure. For diagnostic purposes different electrical and optical methods are used, namely the recording of ignition voltage, ignition current and light emission by a photo-diode signal on nanosecond time scale as well as short-time photography by a intensified charge-coupled device camera. It was found that the DBD mainly generates a potential distribution within the lamp which supports ignition by an increase in the E-field in front of the electrodes and the wall. It is shown that this effect is distinctly more effective than UV radiation potentially emitted by the DBD.

The gas phase emitter effect of lanthanum within ceramic metal halide lamps and its dependence on the La vapor pressure and operating frequency

The gas phase emitter effect increases the lamp lifetime by lowering the work function and, with it, the temperature of the tungsten electrodes of metal halide lamps especially for lamps in ceramic vessels due to their high rare earth pressures. It is generated by a monolayer on the electrode surface of electropositive atoms of certain emitter elements, which are inserted into the lamp bulb by metal iodide salts. They are vaporized, dissociated, ionized, and deposited by an emitter ion current onto the electrode surface within the cathodic phase of lamp operation with a switched-dc or ac-current. The gas phase emitter effect of La and the influence of Na on the emitter effect of La are studied by spatially and phase-resolved pyrometric measurements of the electrode tip temperature, La atom, and ion densities by optical emission spectroscopy as well as optical broadband absorption spectroscopy and arc attachment images by short time photography. An addition of Na to the lamp filling increases the La vapor ...

The effect of active antennas on the hot-restrike of high intensity discharge lamps

The ignition voltage of high intensity discharge (HID) lamps with mercury as the buffer gas may rise from 3 kV for the cold state up to more than 15 kV for a hot lamp. By coating a lamp burner with an electrically conductive layer, which operates as an active antenna, the ignition voltage of HID lamps can be significantly reduced. An active antenna connected to one of the lamp electrodes transports the potential from this electrode to the vicinity of the opposite electrode and generates an enhanced electric field inside the burner. On applying a symmetrically shaped ignition pulse, a weak pre-discharge within the first half-cycle produces free charge carriers initiating ignition of the lamp within the subsequent second half-cycle. The authors present a set-up for electrical and optical investigations of hot-restrike in HID lamps. The ignition voltage is measured for two different polarities as a function of the cooldown time. An analysis of its reduction is given. Furthermore, the pre-discharge is investigated by means of short-time photography. It is demonstrated that a negative polarity of the active antenna within the first half-cycle and a positive polarity within the second one is the most effective succession.

Antenna induced hot restrike of a ceramic metal halide lamp recorded by high-speed photography

The hot restrike is one of the biggest challenges in operating ceramic metal halide lamps with mercury as buffer gas. Compared to a cold lamp, the pressure within a ceramic burner is two orders of magnitude higher during steady state operation due to the high temperature of the ceramic tube and the resulting high mercury vapour pressure. Room temperature conditions are achieved after 300 s of cooling down in a commercial burner, enclosed in an evacuated outer bulb. At the beginning of the cooling down, ignition voltage rises up to more than 14 kV. A significant reduction of the hot-restrike voltage can be achieved by using a so called active antenna. It is realized by a conductive sleeve surrounding the burner at the capillary of the upper electrode. The antenna is connected to the lower electrode of the lamp, so that its potential is extended to the vicinity of the upper electrode. An increased electric field in front of the upper electrode is induced, when an ignition pulse is applied to the lamp electr...