Carbon-coated ceramic membrane reactor for the production of hydrogen by aqueous-phase reforming of sorbitol.

Hydrogen was produced by aqueous-phase reforming (APR) of sorbitol in a carbon-on-alumina tubular membrane reactor (4 nm pore size, 7 cm long, 3 mm internal diameter) that allows the hydrogen gas to permeate to the shell side, whereas the liquid remains in the tube side. The hydrophobic nature of the membrane serves to avoid water loss and to minimize the interaction between the ceramic support and water, thus reducing the risks of membrane degradation upon operation. The permeation of hydrogen is dominated by the diffusivity of the hydrogen in water. Thus, higher operation temperatures result in an increase of the flux of hydrogen. The differential pressure has a negative effect on the flux of hydrogen due to the presence of liquid in the larger pores. The membrane was suitable for use in APR, and yielded 2.5 times more hydrogen than a reference reactor (with no membrane). Removal of hydrogen through the membrane assists in the reaction by preventing its consumption in undesired reactions.

Dealing with a local heating effect when measuring catalytic solids in a reactor with Raman spectroscopy.

In continuation of our previous work on the applicability of the G(R(infinity)) correction factor for the quantification of Raman spectra of coke during propane dehydrogenation experiments (Phys. Chem. Chem. Phys., 2005, 7, 211), research has been carried out on the potential of this correction factor for the quantification of supported metal oxides during reduction experiments. For this purpose, supported chromium oxide catalysts have been studied by combined in situ Raman and UV-Vis spectroscopy during temperature programmed reduction experiments with hydrogen as reducing agent. The goal was to quantify on-line the amount of Cr(6+) in a reactor based on the measured in situ Raman spectra. During these experiments, a significant temperature effect was observed, which has been investigated in more detail with a thermal imaging technique. The results revealed a temperature 'on the spot' that can exceed 100 degrees C. It implies that Raman spectroscopy can have a considerable effect on the local reaction conditions and explains observed inconsistencies between the in situ UV-Vis and Raman data. In order to minimize this heating effect, reduction of the laser power, mathematical matching of the spectroscopic data, a different cell design and a change in reaction conditions has been evaluated. It is demonstrated that increasing the reactor temperature is the most feasible method to solve the heating problem. Next, it allows the application of in situ Raman spectroscopy in a reliable quantitative way without the need of an internal standard.

Real time quantitative Raman spectroscopy of supported metal oxide catalysts without the need of an internal standard.

In continuation to the possibility of using a combined operando Raman/UV-Vis-NIR set-up for conducting qualitative Raman spectroscopy, the possibilities for quantitative Raman spectroscopic measurements of supported metal oxide catalysts under working conditions without the need of an internal standard have been explored. The dehydrogenation of propane over an industrial-like 13 wt% Cr/Al203 catalyst was used as a model system. During reaction, the catalytic solid was continuously monitored by both UV-Vis-NIR and Raman spectroscopy. As the dehydrogenation proceeds, the catalyst gradually darkens due to coke formation and consequently the UV-Vis-NIR diffuse reflectance and Raman scattered signal progressively decrease in intensity. The formation of coke was confirmed with TEOM, TGA and Raman. The measured Raman spectra can be used as a quantitative measure of the amount of carbonaceous deposits at the catalyst surface provided that a correction factor G(R(infinity)) is applied. This factor can be directly calculated from the corresponding UV-Vis-NIR diffuse reflectance spectra. The validity of the approach is compared with one, in which an internal boron nitride standard is added to the catalytic solid. It will be shown that the proposed methodology allows measurement of the amount of carbonaceous deposits on a catalyst material inside a reactor as a function of reaction time and catalyst bed height. As a consequence, an elegant technique for on-line process control of e.g. an industrial propane dehydrogenation reactor emerges.