Carbon dioxide methanation for synthetic natural gas (SNG) production

This work deals with the coupling between high temperature steam electrolysis and carbon dioxide methanation (hydrogenation) to produce a synthetic gas directly injectable in the natural gas distribution grid. This system concept is one of the existing possibilities related to a pathway named Power-to-gas. According to this strategy, low-priced surplus electric energy coming from renewable energy sources (RES) or nuclear plants can be converted into chemical energy of a fuel in order to store it for a longer time. Another purpose is to compensate the unbalances of the electric grid due to the daily and seasonally fluctuations of electricity production from RES (especially wind and solar). The integration between solid oxide electrolysis cell (SOEC) technology and methanation seems to be promising due to the possible thermal integration between exothermal hydrogenation and thermal energy required within the water splitting unit. Hydrogen generated through steam electrolysis can react with carbon dioxide producing methane and water (4 H2 + CO2 ↔ CH4 + 2 H2O). A preliminary screening of nickel-based catalysts has been carried out at atmospheric pressure in order to identify the catalyst(s) with higher activity. During this experimental activity seven samples have been tested: a standard NiO on γ-Al2O3 catalyst (Ni/A); three Ni-based samples (Ni/C5, Ni/C10 and Ni/C15) with a composite support containing a mixture of γ-Al2O3 and other promoters (CeO2, ZrO2 and TiO2); two nickel-aluminum hydrotalcites (Ni-Al 8.7 and Ni-Al 12, prepared at a pH of 8.7 and 12, respectively) and a commercial NiO/γ-Al2O3 catalyst (CRG-F). Hydrotalcites and commercial catalyst present better performance than oxides-supported nickel samples and are more active especially at low temperature (below 300 °C). The different activity far from equilibrium can be put in relation with some physic and chemical properties. Hydrotalcites and commercial catalyst presented higher nickel content than the other samples. Moreover, also the metal dispersion seems to play a role in order to enhance the catalytic performance. Concerning oxide-supported Ni-based catalysts, the addition of promoters (CeO2, TiO2 and ZrO2) to γ-Al2O3 within the support showed a beneficial effect on the activity due to the increased catalyst reducibility. A new test rig has been designed and set up in order to perform an experimental activity at high pressure (up to 30 bar). Samples at both small (0.25-0.5 mm) and pellet (3 mm) size have been tested at different pressure, temperature and inlet gas composition. As a preliminary activity, a commercial catalyst at pellet size was tested at 300 °C and different residence times with two different inlet mixtures (in order to reproduce a series of two reactors with an inter-condensation of the produced water). This experimental activity enabled the production of synthetic gas with hydrogen content lower than 5%, which is the maximum acceptable H2 amount considered in this work for the direct injection of SNG into the natural gas distribution grid. Thus, the feasibility of the process in terms of overall conversion into methane was verified. Then, a test campaign has been carried out by varying several operating parameters in order to describe the methanation kinetics for a commercial catalyst (NiO/γ-Al2O3) at small particles size (250-500 μm). Total and reactants partial pressures and temperature were varied during the experimental activity. The obtained experimental points have been used into an ideal PFR model for the kinetic parameters estimation. Both power law and Langmuir-Hinshelwood (LHHW) rate equations were considered. Concerning LHHW-type equations, the one leading to the best fitting of experimental data is based on the dissociative chemisorption of both hydrogen and carbon dioxide as reaction mechanism. The obtained kinetics was used as a basis for a 1D plug flow reactor model applied to a series of two cooled multi-tube fixed bed reactors for methane synthesis: the main goal is to estimate temperature and conversion profiles along the axial coordinate. Evaporating water at 240 °C (i.e. at ≈ 33 bar) has been considered as a coolant: this strategy ensures a high heat transfer coefficient on the shell side. Both micro and macro kinetics equations are solved. Thus, transport phenomena between gas and solid catalyst have been taken into account. In addition, the evaluation of the effectiveness factor for isothermal particles enabled the estimation of the mass transfer inside the porous catalyst. In order to moderate the temperature increase (i.e. to prevent the hot spot risk) especially within the first reactor, part of the reacting CO2 is conveyed directly to the second reactor by-passing the first one. A carbon dioxide split ratio of 0.7 (meaning that the 30% of the total CO2 flow by-passes the first reactor) ensures that the maximum temperature reached within the solid catalyst is lower than 600 °C (this value has been fixed at the maximum acceptable temperature). The length of the second reactor was adjusted in order to ensure a methane fraction in the outlet gas equal to 95% (on dry basis), enabling the production of a synthetic gas with a H2 content lower than 5% (i.e. injectable in the NG pipeline). Inlet pressure has been set equal to 15 bar. The obtained results from the 1D model have been used for the design of the methanation unit consisting in a series of two cooled reactors with steam inter-condensation. Then, the process modeling of a plant coupling high temperature electrolysis and methanation is presented: the main goal of this analysis is the calculation of an overall plant efficiency (in terms of electricity-to-SNG chemical energy). The plant size has been set considering a 10 MWel SOEC-based electrolysis unit. It has been assumed that the heat produced from the exothermal methanation is entirely used for the water evaporation; the as generated steam is the key reactant of the electrolysis unit. Through the pinch analysis, a further thermal integration between hot and cold streams was performed. The external heat requirement obtained through the minimization of thermal needing was equal to 121 kW (≈1% of the electrolyser duty). However, such integration requires a too high number of heat exchangers, resulting in increasing costs and higher system complexity. Thus, the heat exchangers network has been re-designed in order to reduce the number of components. Hot and cold streams of electrolysis section have been coupled; a similar procedure has been applied to the methanation unit. Thus, the only integration between the two sections is represented by the reaction heat used for the water evaporation. The reduced complexity results in a higher external heat requirement (272 kW): this new value has been considered acceptable (≈3% of the electrolysis power). Efficiency was calculated as the ratio between the SNG chemical power and the overall electric input (including electrolysis power, compression duties and external heating). The SOEC-based power-to-gas system presented an HHV-based efficiency equal to ≈ 86 % (≈ 77 % on LHV basis).