Ontwikkeling en optimalisatie van fotochemische reactoren voor productiviteit en efficiëntie

Continuous photochemical reactors with conventional homogeneous illumination suffer from a photon efficiency problem, which is inherent to their design: dark zones arise near the reagent-rich inlet whereas the reagent depleted outlet is over-illuminated. Any attempt to mitigate dark zones by decreasing the initial concentration of reagent at the inlet of the reactor will only increase photon losses further downstream. This study reports the principles and model equations for co- and counter-current illumination in photochemical reactors, along with an optimization study to determine the most efficient and productive operating point. We prove that the use of co- and counter-current illuminated continuous reactors using photochemical reactions increases the energy efficiency while easing scalability by implementing larger path lengths, without altering the reactor’s geometry. We report a simple model to determine the conversion obtained by such novel (co- and counter-current) illumination techniques and compare it to the current state of the art (cross-current) illumination method. Two non-dimensional groups where derived that describe all possible reactor configurations, these are the initial absorbance of reagent and the quantum photon balance ($\rho\phi$), the ratio of photons and reagent influx. Variation of both parameters leads for non-competitive photochemical reactions to an optimal point for the current state of the art as well as the novel co-axial illumination. Ultimately, we recommend the use of an absorbance value of at least 1, and a photon quantum balance equal to 1 to introduce sufficient photons and enable near complete absorption of photons. The derived mathematical model for co- and counter-current illumination was validated by experimental data. We report the use of co- and counter-current illumination to significantly boost the photochemical space time yield (PSTY), a measure for photochemical reactor productivity, by a factor of 5 - 6 for same input concentration of reagent, solely by illuminating the reactor via axial co- and counter-current modi instead of cross-current modus. Furthermore, simply increasing the reagent concentration, thus absorbance, has a large influence on productivity, thus PSTY, of the reactor. By assessing the reactor’s performance parameters, derived by modelling, which are the initial absorbance value and the quantum photon balance, the reactor’s operating point can be determined and adapted to operate in a more productive and efficient regime. The use of non-collimated LEDs for co- and counter-current illuminated reactors was studied with the focus on scale-up. A ray tracing model was implemented in COMSOL and was validated using experimental data for this purpose. Via these experimental data, the regime of no kinetic limitations was observed as conversion is not hampered by increasing photon flux. Via the model results, it was determined that the light source for optimal photon absorption by the reagent was the most collimated LED possible, in this case with a total viewing angle of 10°. The optimal reactor set-up uses the most reflective material, preferably aluminium or silver, to recuperate diverging light rays for the used wavelength. Furthermore, the wall thickness of the glass reactor must not be excessively thick, with an optimum of 1.5 mm wall thickness for this case. The reactor glass itself is best made of a material which a refractive index sufficiently different from the reaction medium for optimal performance. In practice, glasses with high refraction index are best suited. Regarding reagents and absorbance, it is best to use a higher concentration and reduce reactor length as this increases reactor performance under the condition that quantum yield is stable. Via the use of non-collimated light, it was determined that the entrance efficiency can be increased compared to a fully collimated light source, at the cost of reflection losses that increase with path length.

State-of-the-art two-phase flow cross-current illuminated photochemical reactors are run in micro- to milli flow geometries with high gas hold-ups, and lacking energy efficiency required for industrial applicability. As an alternative, co-currently illuminated bubbly flow photochemical reactors are proposed. These have the advantage of running at higher absorbance values, thereby wasting less photons, improving cost efficiency, and decreasing gas hold-up by separating gas and liquid residence time in the reactor. A mathematical model was used in the study in combination with photosensitized reactions, which yielded three coupled differential equations and four non-dimensional groups governing the reactor performance. A parameter study yielded an optimal operating point at a quantum photon balance of 1, a sensitizer absorbance of at least 1, a chemical quenching parameter of less than 0.1, and a sensitizer stability ratio of less than 1. Optimal absorbance was determined via the use of a ray tracing study, yielding a theoretical optimal absorbance of 1.75 using 7 µM of rose bengal. The modelling results are experimentally validated using an in-house developed co-axially illuminated two-phase bubbly flow reactor. The model was shown to be suitable across a wide range of operating conditions. Optimal absorbance was experimentally evaluated to match the modelling results, higher absorbance values lead to more local absorbance, throttling reactor performance. The reactor was shown to be mass transfer limited at high photon fluxes, governed by the liquid side. Guidelines are presented to design this novel type of reactor and ways to improve the reactor performance in scaled up versions. The reactor was compared against the current state-of-the-art reactors, yielding an improvement of PSTY by 4.5 times and an increased reagent/sensitizer ratio by 100 times, both drastically decreasing operational cost of the reactor.