• Energy Research

For the design, scale-up, and optimization of pressurized packed bed reactors for chemical-looping combustion (CLC), understanding of the effect of the pressure on the reactivity of the oxygen carriers is very important. In this work, the redox reactivity of CuO/Al2O3 and NiO/CaAl2O4 particles at elevated pressures have been measured in a pressurized high-temperature magnetic suspension balance. The experiments have demonstrated that the pressure has a negative influence on the reactivity and that this effect is kinetically controlled. The negative effect of the pressure might be caused by the decrease in the number of oxygen vacancies at higher pressures. Moreover, the reactant gas fraction has been demonstrated as an important parameter, probably related to competition between different species for adsorption on the oxygen carrier surface. These effects have been included in the kinetic model leading to a good description of the experimental results. The impact of these findings on packed bed CLC applications with larger oxygen carrier particles has been investigated with a particle model that considers diffusion limitations and kinetics. It has been shown that the impact of diffusion limitations decreases with increasing pressure, due to the decrease in reaction rates and the increase in diffusion fluxes caused by Knudsen diffusion. The results have been validated by experiments with 1.7 mm NiO/CaAl2O4 particles. These results corroborate that the selection of larger particles because of pressure drop considerations does not lead to a large decrease in effective reaction rates, which is beneficial for packed bed CLC applications.

Abstract Nowadays, nearly 50% of the hydrogen produced worldwide comes from Steam Methane Reforming (SMR) at an environmental burden of 10.5 tCO2,eq/tH2, accelerating the consequences of global warming. One way to produce clean hydrogen is via methane pyrolysis using melts of metals and salts. Compared to SMR, significant less CO2 is produced due to conversion of methane into hydrogen and carbon, making this route more sustainable to generate hydrogen. Hydrogen is produced with high purity, and solid carbon is segregated and deposited on the molten bath. Carbon may be sold as valuable co-product, making industrial scale promising. In this work, methane pyrolysis was performed in a quartz bubble column using molten gallium as heat transfer agent and catalyst. A maximum conversion of 91% was achieved at 1119 °C and ambient pressure, with a residence time of the bubbles in the liquid of 0.5 s. Based on in-depth analysis of the carbon, it can be characterized as carbon black. Techno-economic and sensitivity analyses of the industrial concept were done for different scenarios. The results showed that, if co-product carbon is saleable and a CO2 tax of 50 euro per tonne is imposed to the processes, the molten metal technology can be competitive with SMR.

A quantitative comparison of the performance of the most common reactor configurations proposed for the oxidative coupling of methane (OCM) is made on the basis of numerical calculations with phenomenological reactor models. The configurations that are analyzed can be divided into two main categories, namely, packed‐bed reactors (including conventional packed beds with external cooling, packed‐bed membrane reactors, and adiabatic packed beds with post cracking) and fluidized bed reactors (bubbling fluidized bed reactor, circulating fluidized bed reactor, and fluidized bed membrane reactor). The challenges in both configuration types, mainly the heat management in the case of the packed‐bed reactors and the low C2+ yields obtained in fluidized bed reactors, are evaluated and quantified. To ensure a fair comparison, La2O3/CaO is chosen as the OCM catalyst for all the considered cases, mainly in view of the availability of a comprehensive kinetics model. The results show that, with conventional configurations, it is not possible to achieve high C2+ yields that are needed to make the process economically viable. However, the results also indicate that the C2+ yield can be significantly improved by feeding the oxygen distributively along the reactor axial length.

Membrane reactor processes are being increasingly proposed as an attractive solution for pure hydrogen production due to the possibility to integrate production and separation inside a single reactor vessel. High hydrogen purity can be obtained through dense metallic membranes, especially palladium and its alloys, which are highly selective to hydrogen. The use of thin membranes seems to be a good industrial solution in order to increase the hydrogen flux while reducing the cost of materials. Typically, the diffusion through the membrane layer is the rate-limiting step and the hydrogen permeation through the membrane can be described by the Sieverts’ law but, when the membrane becomes thinner, the diffusion through the membrane bulk becomes less determinant and other mass transfer limitations might limit the permeation rate. Another way to increase the hydrogen flux at a given feed pressure, is to increase the driving force of the process by feeding a sweep gas in the permeate side. This effect can however be significantly reduced if mass transfer limitations in the permeate side exist. The aim of this work is to study the mass transfer limitation that occurs in the permeate side in presence of sweep gas. A complete model for the hydrogen permeation through Pd–Ag membranes has been developed, adding the effects of concentration polarization in retentate and permeate side and the presence of the porous support using the dusty gas model equation, which combines Knudsen diffusion, viscous flow and binary diffusion. By studying the influence of the sweep gas it has been observed that the reduction of the driving force is due to the stagnant sweep gas in the support pores while the concentration polarization in the permeate is negligible.

AbstractIn this work, the performance of an innovative plant for efficient hydrogen production using solar energy for the process heat duty requirements has been evaluated via a detailed 2D model. The steam‐reforming reactor consists of a bundle of coaxial double tubes assembled in a shell. The annular section of each tube is the reaction zone in which Ni‐based catalyst pellets are packed, whereas the inner tube is a dense Pd‐based selective membrane that is able to remove hydrogen from the reaction zone. By coupling reaction and hydrogen separation, equilibrium constrains inside the reactor are circumvented and high methane conversions at relatively low temperatures are achieved. The heat needed for the steam‐reforming reaction at this low operating temperature can be supplied by using a molten salt stream, heated up to 550 °C by a parabolic mirror solar plant, as heating fluid. The effects on membrane reactor performance of some operating conditions, as gas mixture residence time, reaction pressure and steam‐to‐carbon ratio, are assessed together with the enhancement of methane conversion with respect to the traditional process, evaluated in the range 40.5–130.9% at the same operating conditions. Moreover, owing to the use of a solar source for chemical process heat duty requirements, the greenhouse gases (GHG) reduction is estimated to be in the range 33–67%. Copyright © 2009 Curtin University of Technology and John Wiley & Sons, Ltd.

The DemoCLOCK (Demonstration of a cost effective medium size Chemical Looping Combustion through packed beds using solid hydrocarbons as fuel for power production with CO2 capture) project is co-financed by the European Commission and is being developed by a consortium of 11 partners from 7 European countries. The objec- tive of the project is to demonstrate the technical, economic and environmental feasibility for implementing packed bed based high temperature and high pressure chemical looping combustion in large-scale power plants. The aim of this paper is to present the preliminary results of a process simulation study of Integrated Gasification Chemical Looping Combustion (IG-CLC) plants, based on packed bed CLC reactors. The performance of complete IG-CLC power plants, including the coal gasification system, the CLC process and the power island, are predicted. Two configurations are addressed. In the first one, a gas-steam combined cycle is used, where the high temperature O2-depleted stream produced by a pressurized air reactor is expanded in a gas turbine. In the second option, power is generated by an advanced super critical steam cycle. The results obtained are reported in the paper and compared to competitive technologies considered by the EBTF (European Benchmarking Task Force). The results of the simulations indicate a high potential for this technology, with electric efficiencies 2.5 percentage points higher than the competitive IGCC plant with CO2 capture by physical absorption and more than 97% of CO2 avoided

The objective of this work is to study the performance of the oxygen carrier in a packed bed with periodic switching between oxidizing and reducing conditions. In this paper the performance of CuO/Al2O3 as the oxygen carrier in a packed bed reactor with syngas as the fuel are investigated, while also studying the (possible) carbon deposition and the effect of sulphur impurities on the stability of the carrier. Both experiments and simulations are used in this work. Cyclic experiments (oxidation with air and reduction with syngas) have been carried out in a lab scale packed bed reactor with 13 wt% CuO/Al2O3. The experimental results were well described by a 1D reactor model, provided that critical attention was given to the reaction rate for the complete reduction reaction, including a dramatic decrease in reaction rate at high solid conversions. Feeding syngas (pH2 = pCO = 0.1 bar) resulted in 1.1% carbon deposition of the feed. Steam was proven to be more effective in reducing carbon deposition than CO2. Moreover, it has been found that CuO/Al2O3 catalyzed the water gas shift reaction and the reaction rate was not permanently affected by exposure to H2S, two key factors for CLC operation. The results of this work imply that CuO/Al2O3 is an effective oxygen carrier as the first packed bed reactor in a TSCLC process and that the developed model is able to describe the performance at larger scales accurately.

In this paper, we report the performance of supported Pd–Ru membranes for possible applications to hydrogen purification and/or production. For this purpose, we fabricated three ultra-thin α-alumina-supported membranes by combined plating techniques: a Pd–Ag membrane (3 μm-thick ca.) and two Pd–Ru (1.8 μm-thick ca.). The former is set as a benchmark for comparison. The membranes were characterised using different methodologies: permeation tests, thermal treatment and SEM analysis. Preliminary leakage tests performed with nitrogen has revealed that the two Pd–Ru membranes, namely PdRu#1 and PdRu#2, show a non-ideal (non-infinite) selectivity, which is relatively low for the former (around 830 at 400 °C) and sufficiently high for the latter (2645 at 400 °C). This indicates a relevant presence of defects in the PdRu#2 membrane, differently from what observed for the Pd–Ag and PdRu#1 ones. The permeation tests show that the hydrogen permeating flux is stable up to around 550 °C, with an apparently unusual behaviour at higher temperatures (600 °C), where we observe a slightly decrease of hydrogen flux with an increase of the nitrogen one. Moreover, a peculiar bubble-shaped structure is observed in the metal layer of all membranes after usage by means of SEM image analysis. This is explained by considering the effect of the Pd-alloy grain surface energy, which tends to minimise the exposed surface area of the grain interface by creating sphere-like bubble in the lattice, similar to what occurs for soap bubbles in water. The above-mentioned decrease in hydrogen flux at 600 °C is explained to be caused by the bubble formation, which pushes the alloy deeper in the support pores.