Neural Net Based Hybrid Modeling of the Methanol Synthesis Process

A Hybrid modeling approach, combining an analytical model with a radial basis function neural network is introduced in this paper. The modeling procedure is combined with genetic algorithm based feature selection designed to select informative variables from the set of available measurements. By only using informative inputs, the model's generalization ability can be enhanced. The approach proposed is applied to modeling of the liquid–phase methanol synthesis. It is shown that a hybrid modeling approach exploiting available a priori knowledge and experimental data can considerably outperform a purely analytical approach.     

Stereoselective hybrid catalysts: new opportunities

The combination of biotransformations and metal‐assisted catalysis within hybrid systems represents a promising strategy to extend the repertoire of reactions enabled in a biomolecular environment. Both covalent catalytic entities and supramolecular assemblies have been created recently to mediate transformations otherwise difficult to perform in ‘classical’ biocatalysis. In this context, various biopolymers can serve as ligand spheres simultaneously acting as chiral inducers. In particular, modular approaches embedding novel metal‐based catalytic centers via an affinity interaction are highly appealing, as this technology may serve as a versatile platform to incorporate diverse reactivities in a biological matrix in a stereoselective fashion. Copyright © 2007 Society of Chemical Industry     

Is cluster policy useful for the energy sector? Assessing self-declared hydrogen clusters in the Netherlands

Research on science-based industries has shown that it is important for organisations to be active in interorganisational networks. Cluster policy has been developed as a means to stimulate the development of these networks and thereby the success rate of these industries. Cluster policy is however not a common policy instrument in the energy sector. In this paper, we focus on three self-declared clusters active in hydrogen-related R&D in the Netherlands and address several characteristics of these clusters. We conclude that cluster policy is a useful addition to existing energy R&D policies but that monitoring whether self-declared clusters actually function as clusters and what their contribution is to the overall system is pivotal in reaping the benefits of cluster policy.     

The Effect of Ozonation on the Size Fractions of Manganese

Manganese is a commonly found substance in groundwater in Finland. As a powerful oxidant, ozone can be used for the oxidizing of manganese even without raising the pH. The SFS (Finnish Standards Association) standard has set the accepted limit for soluble manganese to 0.45 μm. However, some research papers have used the limit of 30 kD (kilodalton) for soluble manganese. This research concentrates on the size fractions of manganese in four samples of untreated groundwater and in four samples of ozonized groundwater when treated with 0.45 μm, 0.20 μm, 100 kD, 30 kD, and 10 kD filters. In all tests, nearly all manganese contained in raw water penetrated all filters. There were slight variations in the flocculation of manganese in ozonized groundwater; nevertheless, hardly any reduction in manganese levels took place beyond 100 kD. After ozonation, there were two water samples which surpassed the manganese limit of 50 μgl^?1 set for domestic water when the filtration was 0.20 μm and another two samples when the filtration was 100 kD.     

Operational and safety analyses of a commercial PEMFC system

A commercial, transportable proton-exchange-membrane fuel-cell (PEMFC) system rated at 7 kW nominal power was the subject of an operational (reliability, availability, and maintainability) and process-safety analysis. The scope of the analysis included the system start-up to engine-run mode, the maintenance tasks, as well as the continuous power-production mode, carried out with a hazard and operability (HAZOP) study and a failure-scenarios' development. The failure and process-safety risks were categorized and the probabilities were assessed with a fault-tree analysis (FTA) using equipment-failure-rate data from the literature, if available. The results led to similar conclusions to those previously reported in similar studies, and the most important equipment/devices with respect to reliability and safety aspects were identified. While the most important uncertainty in the complete analysis comes from the generic equipment failure-rate data used, about three failures in the first year of operation are estimated together with an availability of about 98% (a point value figures, confidence interval analysis was also made). The process-safety analysis revealed four potential accident scenarios involving hydrogen releases, fires and explosions. Finally, detailed suggestions and recommendations for changes/improvements with respect to the reliability/availability, maintainability and safety aspects of the system analysed are presented.     

Liquid Chromatography–Light Scattering Detector–Mass Spectrometric Analysis of Digested Oxidized Rapeseed Oil

Rapeseed oil was oxidized chemically and thermally to produce two distinct oxidized oils. These oils, along with unoxidized oils, were subjected to an artificial digestion model to simulate the digestive processes in humans. Lipid digestion involves lipases that break down the intact triacylglycerol (TAG) molecules first to diacylglycerols, and eventually to sn-2-monoacylglycerols (MAG) and free fatty acids. A high performance liquid chromatography-evaporative light scattering detector-electrospray ionization-mass spectrometric (HPLC–ELSD–ESI–MS) method was developed to monitor the lipolysis and the presence of oxidized lipids. The HPLC–ELSD–ESI–MS analysis enabled the separation and detection of nearly all the lipid species present in the sample after TAG hydrolysis. The HPLC–MS analyses of digestion products revealed that oxidized triacylglycerols are hydrolyzed by the digestive enzymes in a manner similar to that of native, unoxidized molecules. Significant amounts of sn-1(3)-MAG were found in all the samples after lipolysis, however, more of these were found in unoxidized rapeseed oil samples than in the oxidized oils. Several oxidized molecules were identified with the aid of synthesized oxylipids. This novel method is scalable to small-scale preparative fractionation of oxidized lipid molecules from a complex digestion sample. Also, the fingerprint-like, diagnostic, MS profiles of oxidized oils, reference compounds, and digestion products may be a great aid in comprehensive analysis of lipid oxidation and lipolysis.