Catalytic sulphur trioxide splitting (STS) reaction in a pressurized experimental setup for a sulphur-based thermochemical cycle for long-term solar energy storage
Loading...
Date
2026-07-07
Authors
Καρύδης, Αριστείδης
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
The present diploma thesis deals with the modification, commissioning and operation of a pres-surized experimental set-up for the catalytic splitting of sulphur trioxide, SO₃ ⇌ SO₂ + ½ O₂. The STS reaction is one of the high-temperature steps of the SULPHURREAL sulphur-based thermochemical cycle, which is investigated as a possible route for long-term solar energy storage. The work focused on two main parts: first, the adaptation of an existing atmospheric-pressure unit into a pressurized SO₃ formation and SO₃ splitting set-up, and second, the catalytic evaluation of Sample 1, a short cy-lindrical honeycomb-type Fe₂O₃–clay structured catalyst body with a nominal diameter of approxi-mately 16 mm, a height of 4 mm and a total mass of 0.59 g. Additional catalyst geometries were also prepared and documented as part of the broader experimental campaign, but the detailed cata-lytic analysis presented in this thesis refers to Sample 1.
The final experimental configuration included the gas supply and flow-control section, the SO₃ formation reactor, the STS reactor, the heated transfer lines, the cold trapping system, the pressure-control unit and the online UV–Vis analysis section used for SO₂ quantification. Special attention was given to the position of the back-pressure regulator, the cold traps and the downstream filters, since the system had to operate with highly reactive, condensable and corrosive SO₃-containing streams. The final arrangement allowed operation of the reactor section up to 3 bara, while keeping the UV–Vis cell close to atmospheric pressure and protected from SO₃ breakthrough.
The catalytic experiments were carried out under dry SO₃ splitting conditions, mainly at 850 °C, while varying the pressure, WHSV and inlet SO₃ concentration. At 1 bara and 10.0% inlet SO₃, Sam-ple 1 reached an SO₃ conversion of 81.3%, close to the model prediction of approximately 86%. In-creasing the pressure reduced the conversion, in agreement with the thermodynamic limitation of SO₃ decomposition. For example, at approximately 10% inlet SO₃ and similar WHSV, the conversion decreased from 81.3% at 1 bara to 79.8% at 2 bara and to about 71.1–71.5% at 3 bara. The effect became stronger at higher SO₃ concentration, where the conversion decreased from about 80–81% at 1 bara to 65.4–66.2% at 3 bara. The model described the main trends, but the experimental val-ues were not always reproduced quantitatively, especially under the less favorable high-pressure and high-SO₃ conditions.
The kinetic analysis showed that the observed SO₃ decomposition rate had an almost first-order dependence on SO₃ partial pressure, with an apparent reaction order of 1.0193 and R² = 0.9795. The GHSV experiments confirmed the importance of residence time, although the plug-flow representa-tion was not equally valid under all operating conditions. At 850 °C and 10.0% inlet SO₃, the linear fit gave R² = 0.7907, showing that this case is located in a transition region between thermodynamically influenced behavior and kinetic control. This means that part of the conversion response is affected by the approach to the SO₃ equilibrium limit, while the remaining trend follows the expected kinetic dependence on residence time. At 850 °C and 15.0% inlet SO₃, the fit improved to R² = 0.9906, indi-cating that the system behaved closer to the kinetic region. At 780 °C and 15.0% inlet SO₃, the fit became even stronger, with R² = 0.9965, showing that the conversion followed the expected kinetic plug-flow trend more clearly at the lower temperature.
The Arrhenius analysis over the extended temperature range of 740–902 °C revealed three ap-parent temperature regions. The low-temperature region, between 740 and 810 °C, gave an appar-ent activation energy of 112 kJ mol⁻¹. In the intermediate region, between 821 and 860 °C, the ap-parent activation energy decreased to 48 kJ mol⁻¹, while in the high-temperature region, between 860 and 902 °C, it decreased further to 36 kJ mol⁻¹. This gradual decrease indicates a transition from a mainly kinetic-controlled regime towards a mixed kinetic–diffusion regime at higher temperatures. Therefore, the experiments around 850 °C should not be interpreted as purely kinetic, but rather as lying close to a transition range where transport and active-site accessibility begin to influence the measured conversion.
Overall, Sample 1 showed 7catalytic activity for SO₃ splitting and followed the expected trends with pressure, WHSV, inlet SO₃ concentration and temperature. At the same time, the results showed that the performance of the Fe₂O₃–clay catalyst is strongly affected by its structured form. Future improvement of this catalyst type should therefore focus not only on chemical composition, but also on catalyst architecture, wall thickness, internal porosity and exposure of the active Fe₂O₃ phase, in order to reduce diffusion limitations and improve the effective use of the active material.
Description
Keywords
Sulphur trioxide splitting, Sulphuric acid splitting, Sulphur-based themochemical cycle, Long-term solar energy storage, Fe2O3 catalyst