Membrane-Assisted Chemical Switching Reforming for pure hydrogen production with integrated CO2 capture
MetadataVis full innførsel
Concerns regarding the anthropogenic greenhouse gas emissions on climate change have drawn substantial research efforts to develop new technologies that can mitigate or, if possible, eliminate anthropogenic emissions of especially carbon dioxide into the atmosphere. Among the different mitigation strategies, the long-term proposed solution is to substitute technologies based on fossil fuels by technologies mainly based on renewable energy sources. As a short and mid-term solution, thereby buying us time to implement this major transition, Carbon Capture and Storage (CCS) technology is regarded as a promising route, which can contribute up to 20% to the reduction in total emissions. However, the primary challenge for large-scale exploitation of CCS is the large energy penalty involved, mainly related to CO2 capture, in currently available capture technologies. In recent years, many new technological approaches for CCS with reduced primary energy requirements have been proposed. Among the possible technological strategies, hydrogen production, as alternative carbon free energy carrier, integrated with CO2 capture in a single process is regarded as a promising technology in terms of process efficiency. Different systems have been investigated in this regard, with Chemical Looping Reforming (CLR) being considered as one of the most promising technologies. With CLR, natural gas is transformed into syngas through steam methane reforming, and the outlet stream is sent to Water Gas Shift (WGS) reactors and a Pressure Swing Adsorption (PSA) unit for pure H2 production. As an alternative, membrane reactors have recently emerged as a very promising technology for pure hydrogen production, as these reactors integrate the catalytic reactions, mostly reforming and water-gas shift reactions for hydrogen production and separation and purification through membranes in a single unit. This combination of process units brings a high degree of process intensification with additional benefits in terms of increased process efficiencies. In this PhD thesis, a novel hybrid reactor concept is proposed and developed, and it combines the advantages of chemical looping reforming technology and membrane reactor technology for ultra-pure hydrogen production with integrated CO2 capture from steam methane reforming. The so-called Membrane-Assisted Chemical Switching Reforming reactor (MA-CSR) employs a Ni-based oxygen carrier, which acts both as catalyst and heat/oxygen carrier to the endothermic reforming reaction, and is periodically exposed to fuel/steam and air streams. When air is fed to the reactor, the oxygen carrier is heated by the exothermic solids oxidation reaction. This heat is then utilized in the fuel stage, where the endothermic solids reduction and catalytic reactions produce syngas and regenerate the oxygen carrier. This process can be improved by using hydrogen perm-selective Pd-based membranes to directly recover pure hydrogen produced during steam methane reforming while simultaneously shifting steam reforming and water-gas shift reaction equilibria towards complete conversion at lower temperatures. The MA-CSR concept offers large benefits in design simplification, scale up and ease of operation at elevated pressures. The main aim of this study is to investigate the reactor behaviour and demonstrate the technical and economic feasibility of the novel concept of MA-CSR through dedicated experimental and modelling studies. The development of the proposed concept requires detailed knowledge of different aspects, including a fundamental study on the presence of and permeation through the membranes on the fluidization behaviour, membrane permeability, permselectivity and stability, reactivity of the selected oxygen carrier, thermodynamics and techno-economic analysis of the entire MA-CSR process. By combining all knowledge from these different fields, the behaviour and performance of the novel hybrid system (MA-CSR) can be fully understood, designed and demonstrated to provide an experimental proof-of-concept. All these topics are addressed in this thesis. One of the objectives of this thesis is to gain more fundamental understanding of fluidization with the presence of membranes. The presence of (and extraction of gas through) the membranes influences the hydrodynamics of the fluidized bed by altering the bubble behaviour and the extent of gas back-mixing. Therefore, a hydrodynamic study has been carried out on the membrane-assisted fluidized bed and this is presented in Chapter 2. An advanced non-invasive experimental technique coupling Particle Image Velocimetry (PIV) with Digital Image Analysis (DIA), combined with numerical simulations using a Two Fluid Model (TFM) approach closed by the kinetic theory of granular flow have been used to investigate the hydrodynamics of the fluidized bed membrane reactor. A pseudo 2D experimental setup with a multi-chamber porous plate (membranes) mounted at the back plate was used to simulate vertically immersed membranes. This setup allowed for gas extraction at specific locations from the back of the column, thus facilitating studies on the effects of gas extraction, and their rate and location on the bubble properties; the study included the characterization of bubble properties by quantifying the average bubble size, number, velocity and shape over a range of particle sizes, fluidization velocities at different gas extraction rates and extraction locations. In addition, by combining pressure fluctuation measurements and PIV measurements it was possible to validate the use of pressure measurements as a tool to early detect the formation of densified zones (induced by high gas extraction) in the vicinity of the membranes at different gas extraction rates and extraction locations. Chapter 3 presents the full mechanism of the formation of densified zones and quantitatively characterize the extent of their formation. Results show that the peak in the standard deviation of the pressure fluctuations, normally employed to indicate the transition from freely bubbling to turbulent fluidization, in fluidized beds with gas extraction through flat vertical membranes indicates the onset of the formation of densified zones rather than the transition towards turbulent fluidization. The CSR reactor without membranes using a Ni-based oxygen carrier has been successfully demonstrated at lab-scale and this is presented in Chapter 4. The catalytic performance of the oxygen carrier was first confirmed by showing that equilibrium concentrations for steam methane reforming were achieved even with relatively short gas residence times. The CSR reactor behaviour was subsequently investigated by varying different operating conditions, such as temperature, oxygen carrier utilization and feed rate. The demonstration also included a proof-of-concept for hydrogen production operated under autothermal conditions. All the knowledge gained from the previous investigations has been combined in Chapter 5, where the MA-CSR concept was experimentally demonstrated in a labscale reactor. Experiments in the MA-CSR reactor have led to the production of pure hydrogen recovered from the membranes (with more than 26% recovery) and methane conversions above 50% at relatively low temperatures (~500 °C), while the solid oxygen carrier was dynamically switched between oxidation and reduction/reforming atmospheres. The analysis also showed that these results could be further improved by operating at higher pressures or by integrating more membranes. Even though the concept has been successfully demonstrated, further research is required to develop suitable membranes, as post-experiment membrane characterization has revealed defects in the membrane selective layers as a result of the frequent exposure to thermal cycles with dynamic oxidative/reducing atmospheres and exposure to higher oxygen concentrations. Finally, the potential of the MA-CSR reactor concept for industrial applications by means of a techno-economic evaluation and subsequent comparison of the system with conventional reforming processes has been carried out and this detailed analysis is presented in Chapter 6. Process simulation results show that the MA-CSR process can in principle achieve a similar equivalent H2 production efficiency as the conventional steam methane reforming process without CO2 capture (81%). Moreover, the MA-GSR concept can realize a higher H2 production efficiency than conventional plants when integrated with CO2 capture (20% higher), when the stability of the Pd-based membranes is further improved.