A neural network based tool for semi-automatic code transformation

A neural network based tool has been developed to assist in the process of code transformation. The tool offers advice on appropriate transformations within a knowledge-driven, semi-automatic parallelisation environment. We have identified the essential characteristics of codes relevant to loop transformations. A Kohonen network is used to discover structure in the characterised codes thus revealing new knowledge that may be brought to bear on the mapping between codes and transformations or transformation sequences. A transform selector based on this process has been developed and successfully applied to the parallelisation of sequential codes.

Optical diagnostics of micro discharge jets

Micro plasmas operated at ambient pressure with dimensions of the confining geometry in the order of a few ten micrometers to a millimeter are actually in the focus of interest due to the broad regime of applicability they offer and due to a similarly broad range of open physical questions. Here we present optical measurements within the discharge core and the effluent region of an especially developed micro discharge jet. To get an understanding of the complex system of this discharge it is important to analyse transport phenomena of energy and particles within both parts of the discharge by various highly sophisticated diagnostics. As a consequence of the limited access and the dimensions of the micro discharge most of these diagnostics are optical. Here we present diagnostics applied to determine spatially resolved absolute atomic oxygen densities as the most reactive constituent of the effluent, density maps of ozone as final reaction product of the gas chemical chain induced by the discharge and phase resolved optical emission spectroscopy yielding insight into the excitation dynamics of the discharge. (C) 2007 WILEY-VCH Verlag GmbH & Co. KGaA. Weinheim.

Diagnostics on an atmospheric pressure plasma jet

The atmospheric pressure plasma jet (APPJ) is a homogeneous non-equilibrium discharge at ambient pressure. It operates with a noble base gas and a percentage-volume admixture of a molecular gas. Applications of the discharge are mainly based on reactive species in the effluent. The effluent region of a discharge operated in helium with an oxygen admixture has been investigated. The optical emission from atomic oxygen decreases with distance from the discharge but can still be observed several centimetres in the effluent. Ground state atomic oxygen, measured using absolutely calibrated two-photon laser induced fluorescence spectroscopy, shows a similar behaviour. Detailed understanding of energy transport mechanisms requires investigations of the discharge volume and the effluent region. An atmospheric pressure plasma jet has been designed providing excellent diagnostics access and a simple geometry ideally suited for modelling and simulation. Laser spectroscopy and optical emission spectroscopy can be applied in the discharge volume and the effluent region.

Determination of the degree of dissociation in an inductively coupled hydrogen plasma using optical emission spectroscopy and laser diagnostics

Gas temperature is of major importance in plasma based surface treatment, since the surface processes are strongly temperature sensitive. The spatial distribution of reactive species responsible for surface modification is also influenced by the gas temperature. Industrial applications of RF plasma reactors require a high degree of homogeneity of the plasma in contact with the substrate. Reliable measurements of spatially resolved gas temperatures are, therefore, of great importance. The gas temperature can be obtained, e.g. by optical emission spectroscopy (OES). Common methods of OES to obtain gas temperatures from analysis of rotational distributions in excited states do not include the population dynamics influenced by cascading processes from higher electronic states. A model was developed to evaluate this effect on the apparent rotational temperature that is observed. Phase resolved OES confirmed the validity of this model. It was found that cascading leads to higher apparent temperatures, but the deviation (~25 K) is relatively small and can be ignored in most cases. This analysis is applied to investigate axially and radially resolved temperature profiles in an inductively coupled hydrogen RF discharge.