Applied Catalysis A: General, Vol.561, 7-18, 2018
Modified Harrick reaction cell for in situ/operando fiber optics diffuse reflectance UV-visible spectroscopic characterization of catalysts
UV-visible (UV-vis) spectroscopy is a common, powerful, and affordable technique for the characterization of heterogeneous catalysts. Here, we present an improved design of the commercial ubiquitous Harrick Scientific high temperature reaction cell for use in diffuse reflectance (DR) UV-vis spectroscopy with fiber optics at very close proximity of a catalyst sample and with high time resolution. The cell possesses significant dead volume which was reduced by a homemade compact dome and by volume reduction of cell void space with simple addition of glass beads, thereby, enabling faster transfer of gases. The cell was also improved by adding a second thermocouple to directly monitor the temperature of the catalyst bed via the outlet port without requiring any additional machining. This modified design and the use of an optical fiber DR probe in conjunction with a miniature concave-CCD combination based spectrometer allowed fast acquisition of in situ UV-vis spectra in the order of seconds and at temperatures up to about 500 degrees C. It is also shown that, unlike probes used in tubular reactors, expensive high temperature DR probes are not required in this design. The flow dynamics of the reaction setup were followed by an analysis of residence time distributions (RTD) via pulse experiments of Ar, O-2, H-2, CO, and CO2 as analyzed online by mass spectrometry (MS). These tests enabled a rigorous analysis of the fluid dynamics of the modified cell showing average gas residence times (after correcting for transfer lines and MS contributions) of similar to 13 s at gas flow rates of 45 cm(3)/min (or similar to 4 s at gas flow rates of 120 cm(3)/min) and a fluid behavior that could be approximately described by a CSTR reactor model. The RTD method is of general application and can be easily implemented to other reaction cells to rigorously determine gas mean residence times and distribution, regardless of setup and transfer lines design, provided that a reaction cell bypass line is added to the system. A thermal analysis indicated that significant heat losses due to radiation, conduction, and convection contribute to the observed sample bed vs heater temperature differences. Additionally, an example is presented to show the utility of the modified cell to monitor quickly (every 2 s) and continuously UV-vis spectra over an extended period of time during the in situ dynamic response of gold surface plasmon resonance (Au-SPR) peak shifts on a Au(1 wt%)/ZrO2 catalyst as it is exposed to controlled and cycling oxidizing and reducing environments. The results showed that the Au-SPR peak responded rapidly and shifted reversibly at the studied cyclic oxidizing and reducing conditions. The reported modifications of the reaction cell setup were shown to enable in situ spectroscopic characterization of heterogeneous catalysts. It proved useful for monitoring adsorption and desorption of gas species near gold nanoparticles via Au-SPR and for potentially tracking rapid changes (within seconds) on catalysts with characteristic finger prints in the UV-vis region.
Keywords:In situ/operando spectroscopy;Reaction cell design;Diffuse reflectance;UV-vis;Plasmon resonance