Filtern
Dokumenttyp
Sprache
- Englisch (7) (entfernen)
Schlagworte
- Ellipsometry (3)
- Gas sensing (3)
- Gas releases (2)
- SPR effect (2)
- Dispersion (1)
- Doped tin oxide (1)
- Gas Dispersion (1)
- Gas sensor (1)
- Gold layer (1)
- Heavy gas (1)
Heavy gases in large quantities are used worldwide in various industries. Past incidents, such as
the liquefied gas disaster in Viareggio (2009) have shown that these materials are difficult to
handle in a safe manner. According to the German Hazardous Incident Ordinance (StörfallVO
2000 - 12. BImSchV) plant operators with 'extended responsibilities' must produce a report in
which they verify that, in the event of an unintentional gas release, the surrounding area will not
be aversely affected. The essential elements of this report include calculations of both released
mass flow and gas dispersion. Using models such as the VDI guideline 3783 (state of the art in
Germany) plant operators are able to predict the characteristics of likely gas dispersions.
Within the framework of the reported experimental trials, the release process of heavy gases
from standard gas cylinders was investigated. Using the results of this investigation a calculation
methodology has been developed which is able to predict the mass flow of a gas-phase release.
As input parameters only the geometry of the gas cylinder, the material properties of the
respective fluid as well as the environmental parameters such as velocity of approach and air
temperature are required. In the course of modeling various approaches for nucleate boiling have
been tested for their applicability. Both the calculation methodology and a comparison between
the calculated and experimental results will be presented.
Within the same framework of experimental trials, the dispersion process of the aforementioned
heavy gases was also investigated. The dispersed gas cloud in this case was considered as originating from a continuous point source under ambient conditions. For the various trials
concentrations both in the heavy gas (≥ 1 vol.-%) and in the neutral gas field (≤ 3000 ppm) were
measured. In the immediate area of the release the length, width and height of the heavy gas
cloud were evaluated. The experimental results, comparison calculations and the measurement
techniques will be presented.
Heavy gases in large quantities are used worldwide in various industries. Past incidents, such as the liquefied gas disaster in Viareggio (2009) have shown that these materials are difficult to handle in a safe manner. According to the German Hazardous Incident Ordinance (StörfallVO 2000 - 12. BIMSchV), plant operators with extended responsibilities must produce a report in which they verify that , in the event of an unintentional gas release, the surrounding area will not be aversely affected. Essential elements of this report are calculations of both the released mass flow and the gas dispersion. Using models such as the VDI guideline 3783 (state of the art in Germany) plant operators are able to predict the characteristics of likely gas dispersions. The presented experimental investigations were carried out at the BAM better understand heavy gas dispersion with high gas concentrations in the air (≥ 1 Vol.-%), as well as concentrations with approximately neutral density characteristics (≤ 3000ppm) in order to test the accuracy of the VDI guideline. The starting point for experimental trials was the heavy gas releases resulting from pipeline, vessel or standard gas bottle leaks, with mass flows of between 20 and 100 g s-1. Investigations on the gas-phase release process focused on the unsteady mass flow associated with releases from standard gas bottles. The experimental results will be used as comparative parameters for future simulations. The goal of these simulations is to develop a model with which calculation of the unsteady mass flow, based on the material characteristics for any gas, is possible.
A gas sensor system based on the surface plasmon resonance (SPR) effect in a 44 nm Au layer was studied. The usability of this sensor system was investigated for several analyte gases and gas mixtures which are interesting due to their flammability, toxicity or greenhouse effect. The SPR effect of gold was measured in nitrogen and compared with the gaseous alkane series which caused a shift of the position of the SPR effect. For in situ measurements, the phase shift Δ at the SPR angle (ψ minimum) in the analyte gas mixtures was measured for investigated gases in comparison with nitrogen. With this method, detection limits of different gases could be determined, which lie in a range interesting for safety engineering. Furthermore the aging effect of the gold layers which leads to loss in sensitivity of the sensor was investigated.
An ellipsometric gas sensor based on the surface Plasmon resonance (SPR) effect of ~ 43 nm thick gold layers was investigated. To protect the gold layer from contamination and to improve the detection limits, the gold layers were top-coated with 5–6 nm thick layers of organic a-C:H or with inorganic metal oxides TiO2 or ZrO2. The additional layers increased the long-term stability, whereas the metal oxide layers protect better than a-C:H. Furthermore, the additional layers decreased the detection limits by one order of magnitude in case of a-C:H and two orders of magnitude in case of the oxides. It could be shown that the detection limits also depend on the kind of preparation (sol–gel process or physical vapour deposition) of the additional layers.
In this study, thickness related changes of the optical properties of doped tin oxide were studied. Two different sets of samples were prepared. The first set was doped with iron or nickel on silicon substrate with thicknesses of 29–56 nm, the second was iron doped on gold/glass substrate with 1.6–6.3 nm. The optical constants were determined by using spectral ellipsometry (SE) followed by modelling of the dielectric function with an oscillator model using Gaussian peaks. The analysis of the optical constants shows a dependence of the refraction and the absorption on the thickness of the doped tin oxide coating. In addition to the tin oxide absorption in the UV, one additional absorption peak was found in the near-IR/red which is related to plasmonic effects due to the doping. This peak shifts from the near-IR to the red part of the visible spectrum and becomes stronger by reducing the thickness, probably due to the formation of metal nanoparticles in this layer. These results were found for two different sets of samples by using the same optical model. Afterwards the second sample set was tested in the Surface Plasmon Resonance Enhanced Ellipsometric (SPREE) gas measurement with CO gas. It was found that the thickness has significant influence on the sensitivity and thus the adsorption of the CO gas. By increasing the thickness from 1.6 nm to 5.1 nm, the sensing ability is enhanced due to a higher coverage of the surface with the over coating. This is explained by the high affinity of CO molecules to the incorporated Fe-nanoparticles in the tin oxide coating. By increasing the thickness further to 6.3 nm, the sensing ability drops because the layer disturbs the SPR sensing effect too much.
In this work, thin iron doped tin oxide layers (3–6 nm) were investigated with regard to gas sensitivity by means of surface plasmon resonance (SPR) effect with ellipsometric readout. The experimental set-up was a Kretschmann configuration with gold as metal layer. The sensor system was exposed to different concentrations of methane in the low ppm range and the sensor response of an uncoated gold layer compared to a gold layer coated with iron doped tin oxide. The additional layer effects stability with regard to drift behavior of the sensor and an increase in sensitivity. Furthermore, the sensor is able to detect the toxic gas carbon monoxide in low concentration range (down to 0.5 ppm). The thin layers were investigated by X-ray photoelectron spectroscopy, secondary electron microscopy and spectroscopic ellipsometry. The possible mechanisms taking place on the surface are discussed.