- Email: [email protected]
Forest residues gasification integrated with electrolysis for production of SNG – modelling and assessment
Mario R. Eden, Marianthi Ierapetritou and Gavin P. Towler (Editors) Proceedings of the 13th International Symposium on Process Systems Engineering – PSE 2018 July 1-5, 2018, San Diego, California, USA © 2018 Elsevier B.V. All rights reserved. https://doi.org/10.1016/B978-0-444-64241-7.50013-6
Forest residues gasification integrated with electrolysis for production of SNG – modelling and assessment Johan M. Ahlströma*, Simon Harveya, Stavros Papadokonstantakisa a
Chalmers University of Technology, Division of Energy Technology, Maskingränd 7b, 412 58, Gothenburg, Sweden [email protected]
Abstract This study investigates opportunities for integrating an electrolysis unit with a biomass gasifier for production of synthetic natural gas (SNG). Gasification is a key technology for production of biofuels and chemicals from lignocellulosic biomass, for which an increased demand is expected in the future. H2 produced through an electrolyser can be used to increase the output of a gasifier by reaction with CO2 to form CH4. Four integrated flowsheet configurations are evaluated with respect to system energy efficiency and process operating revenue. The system energy efficiencies are in the range of 0.55 – 0.8, and the maximum value of operating revenues is 0.245 $/kWhdry biomass. The results show that feeding the Sabatier reactor with the full product gas flow coming from the methanation unit, and separating the unreacted CO2 afterwards, is the most attractive configuration with respect to operating revenue. Keywords: Power-to-gas, Gasification, Biomass, SNG
1. Introduction Large scale indirect gasification of biomass has been demonstrated at the GoBiGas plant in Gothenburg (Alamia, 2016) and in Güssing (Bolhàr-Nordenkampf et al., 2002). Direct fluidized bed gasification, which is the subject of this work, has been investigated by Hannula and Kurkela (2012) and Gassner and Maréchal (2012). However, no direct blown gasifier with a full scale downstream gas up-grading section have been constructed. When the raw gas from the gasifier is used for methane production, it needs to be upgraded to increase the methane content. The CO2 remaining after the upgrading step has to be removed, which is done in expensive and energy-intensive sequences of separation steps. Hence, converting the remaining CO2 into valuable products could significantly improve the economic performance of the process. For this purpose, a hydrogen source is required. The share of intermittent electricity generation in the power mix is expected to continue to increase significantly. This will induce more volatility in power generation and thereby also in the electricity price (Woo et al., 2011), thereby creating opportunities for variable load applications with a short response time that can help balance the grid. Power-to-gas concepts which produce H2 and O2 when electricity prices are low, can contribute to this balancing. H2 produced by electrolysis can react with CO2 to produce methane through the Sabatier reaction, whereas O2 can be used as a gasifying agent in the direct gasifier unit. In this way, the product output of the process is enhanced while the energy loads of
J.M. Ahlström et al.
110
the air separation unit and the CO2 separation sequence are decreased. To evaluate this concept, it is essential to estimate how the process performance is affected by the amount of H2 fed to the process. This paper investigates four different power-to-gas configurations integrated with a direct, oxygen blown, biomass gasifier. The gasifier is assumed to be fed with forestry logging residues. The configurations are compared with respect to system energy efficiency and operating revenues.
2. Process configuration Forest biomass generally has a moisture content of 40-60%wt, and the fuel must therefore be dried before it is fed to the gasifier to increase efficiency. This study considers the belt dryer concept proposed by Alamia et al. (2015). The moisture remaining in the fuel is vaporized in the gasifier before the fuel volatile components are released. Thereafter, the char gasification starts. At the bottom of the gasifier, combustion of char occurs, which releases the energy that sustains the endothermic reactions in the gasifier. The raw gas leaving the gasifier contains H2, CO2, CO, CH4, H2O, inorganic impurities (e.g. H2S) and organic compounds such as tars. The ash and traces of char in the raw gas are removed in a cyclone, thereafter H2S is removed. In a pre-methanation step the ratio of H2/CO is adjusted through the water-gas-shift reaction. The methanation reaction occurs in a series of three adiabatic, fixed bed, reactors. Figure 2 provides an overview of the basic process flowsheet.
Figure 1. Process flowsheet
After the methanation section, the gas (containing only CH4 and CO2) is fed into the final CO2 removing sequence. Two specification values are considered for the Wobbe index of the SNG product: 44.7-46.7 MJ/Nm3 and 43.9-47.3 MJ/Nm3, corresponding to the A, respectively B standards of the Swedish national gas grid. The Wobbe index essentially limits the concentrations of both CO2 and H2 in the gas product. A-type SNG is produced if possible, since it can be sold at a higher price, otherwise type B SNG is produced instead. Four possible configurations for the final CO2 removal sequence are investigated (see Figure 2): i. The gas from the methanation section is mixed with H2 from the electrolyser and then compressed before being fed to the Sabatier reactor where H2 reacts with CO2 to increase the share of CH4 in the gas. The gas stream is then dried before
Forest residues gasification integrated with electrolysis for production of SNG
ii.
iii.
iv.
111
removing the remaining CO2, in a sequence of two amine-based CO2 separators, each with a removal efficiency of 90%. The yield of the Sabatier reactor entails that there will be H2 left in the gas after the reactor if all CO2 is to be converted. Since configuration (i) does not include a H2 separation sequence and the gas standards limits the concentration of H2 in the gas product, there will always be CO2 in the gas after the Sabatier reactor. H2 is added to the gas mix in sufficient quantity to convert all remaining CO2, thus removing the need for a CO2 removal step. This configuration requires that the fraction of H2 in the gas has to be decreased, which is achieved with a H2 separation unit. As in configuration (i), the gas is dried before the final separation sequence. The separated H2 is recirculated back to the mixing step before the Sabatier reactor. CO2 is separated from the product gas and mixed with H2 in the Sabatier reactor. The produced gas, containing mainly CH4 but also some remaining CO2, is dried and recirculated to the inlet gas stream before the CO2 removal step. Similar to Configuration (iii), with the difference that a H2 separation step is added to the process after the drying step. This provides a process in which all the CO2 in the raw gas can potentially be reacted to methane, since the excess H2 can be removed. This results in a more flexible process configuration.
Figure 2. The four process configurations.
Candidate electrolyser technologies include alkaline and Polymer Electrolyte Membrane (PEM) electrolysers. PEM technology is characterized by a shorter start up time, but a lower efficiency. Alkaline electrolyser technology has reached a higher development level, and was therefore selected for this work. The main difference between the process configurations investigated is the degree of operational flexibility. Configuration (i) is limited by the fraction of CO2 that can be reacted, since there will be H2 in the produced gas if all CO2 is reacted to methane. Configuration (ii) is limited by the absence of CO2 separation units, meaning that the Sabatier reactor must always be fed with enough H2 to fully react the CO2. Thus configuration (i) is more flexible than configuration (ii). In configurations (iii) and (iv), the CO2 is separated before it is reacted with the H2. Here configuration (iv) provides the more flexible option; enough H2 to react all CO2 can be fed to the process, since a H2 separation sequence is included.
J.M. Ahlström et al.
112
3. Methodology All process modelling is performed in Aspen Plus Version 8.8. The biomass feedstock drying model assumes steady-state isothermal operation and negligible pressure drop. The assumed energy required for biomass drying is 0.51 MJ/kg (Alamia et al. (2015)). All simulation results are based on a feed of a 100 kgdry biomass/h, which gives a flowrate of 2.55 kmol CO2/h and 1.52 kmol CH4/h from the methanation unit. The gasifier is modelled to mimic the results published by Hannula and Kurkela (2012), within an error margin of 10% in terms of gas composition. It is assumed that the tar components are decomposed to CO and H2. Pressure drop in the gasifier is assumed to be negligible. The methanation section is modelled assuming that the reaction reaches chemical equilibrium. The model accounts for the significant pressure drop in the methanation section (see Alamia (2016)). The Sabatier reactor is modelled as a plug flow reactor with Langmuir-Hinshelwood-Hougen-Watsonis kinetics, as described by Schlereth (2015). The CO2 and H2 separation sequences are not rigorously modelled; instead the major component recoveries are inferred by literature data (Heyne and Harvey (2014) and Mivechian and Pakizeh (2013)). Similarly, based on Brynolf et al. (2018), a H2 yield of 23.7 kgH2/MWh electricity is assumed for the Alkaline electrolyser. The system energy efficiency suggested by Heyne and Harvey (2013) is used, including the process by-products and commodities, such as electricity and excess heat
ߟ௦௬௦௧ ൌ
σ ݉ሶ ܸܪܮ ܳି ିܧ σ ݉ሶ ܸܪܮ ܳା ܧା
(1)
ṁ indicates the mass flow of either product, p, or of fuel, f. Q is heat and E electricity, and +, refer to energy flows leaving or entering the process. Process stream heating and cooling requirements are used to perform heat targeting using pinch analysis and evaluate the possible heat recovery for the process. The economic performance metric is the operating revenue, obtained by subtracting the cost of biomass feedstock and all utilities from the revenues of selling SNG, process excess heat and the oxygen produced in the electrolyser. Indicative prices for Swedish conditions are assumed; all input data and results are presented in US$. The reference electricity price is set to 0.024 $/kWh and since it is assumed that the electrolyser is run at periods when the electricity price is low, the electricity used for H2 production is assumed to have a 50% lower cost. Revenue from sales of the SNG product is valued at 1.98 $/kg for A-grade quality and 1.5 $/kg for B-grade quality, the biomass price is 23 $/MWh. All other price data, e.g. catalysts, can be found in the report by Gambardella and Yahya (2017). Sensitivity analysis varying the feed of H2, together with the recirculation of CO2 is performed for all configurations except (ii), in which the H2 flow is constant since the process is designed to react all CO2 available in the gas.
4. Results For configuration (ii) the H2 flow is constant at 10 kmol/h, which is the flowrate required to convert all CO2 in the gas mix. The system energy efficiency of configuration (ii) is 0.801 and the operating revenues are 0.245 $/kWhdry biomass. Configuration (ii) results in
Forest residues gasification integrated with electrolysis for production of SNG
113
the highest revenues and a high system energy efficiency, which is due to the fact that all CO2 in the raw gas is converted to CH4. Sensitivity analysis, varying the feed of H2, is performed for configurations (i), (iii) and (iv). The impact on the operating revenues and system energy efficiency is shown in Figure 3.
Figure 3. Total system energy efficiency and operating revenues as a function of CO2 feed. The nuances correspondent to different configurations and the line types correspond to the amounts of H2 feed.
ηsystem, decreases with increased H2 feed and CO2 recirculation rate. This is due to the conversion losses in the electrolyser. The ranges of each curve indicate the cases where the produced gas fulfills the standards of the Wobbe index. The process operating revenues increase for all configurations with H2 feed and CO2 recirculation. This indicates that the additional SNG produced by the increased addition of H2 outweighs the cost of generating the H2. The increase in revenues is essentially linear, except for some rapid increases and decreases in revenues. These rapid changes indicate the thresholds for the types of SNG produced, namely when the model has to change from type A to type B. Only configuration (i) has an increase in revenue that can be achieved without CO2 recirculation, which is due to the fact that there is already CO2 present in the incoming gas flow. For configurations (iii) and (iv), a certain amount of CO2 has to be recirculated to provide the second reactant to the Sabatier reactor.
5. Conclusions The results show that the operating revenue increases with increased addition of hydrogen. This indicates that there is an incentive for integration of an electrolyser unit with the process, provided that the investment costs of the processes does not outweigh the increase of operating revenue. This also suggests that the input feed of H2 should be maximized if this type of power-to-gas concept is implemented. From an economic perspective, the best performing configuration is (ii), which reaches a maximum operating revenue of 0.245 $/kWhdry biomass followed by (i). Even if configuration (ii) outperforms configuration (i) in terms of revenues, it is not necessarily the better alternative; configuration (i) is more flexible, since it allows for different flows of H2, which could be a major advantage if electricity prices fluctuate a lot. Both configurations (iii) and (iv), in which CO2 is separated before it is mixed with the H2, generates lower revenues than the first two configurations. This is a result that highlights that it is more
114
J.M. Ahlström et al.
beneficial to mix the H2 with the raw gas, rather than to separate the CO2 before the reactor.
6. Outlook and further work An essential parameter for future evaluation is the impact of electricity price on the choice of process configuration and design. The model presented in this work constitutes a starting point for such evaluation. To investigate the impact of the electricity price on the process design, a rolling horizon, planning and scheduling optimization algorithm will be developed. By considering the possibility to store hydrogen, and co-running the optimization model with a model of the European electricity system, it becomes possible to determine the optimal process configuration design, accounting for the fluctuations in electricity price. To enable this sort of analysis, it is a necessity to first estimate the investment costs of the different process configurations
7. References Alamia, A., 2016. Large-Scale Production and Use of Biomethane, Chalmers university of Technology. Alamia, A., Ström, H., Thunman, H,. 2015. Design of an integrated dryer and conveyor belt for woody biofuels, Biomass and Bioenergy, 77, 92-109. Bolhàr-Nordenkampf, M., Bosch, K., Rauch, R., Kaiser, S., Tremmel, H., Aichernig, C. & Hofbauer, H., 2002. Scale-up of a 100kWth pilot FICFB-gasifier to a 8 MWth FICFBgasifier demonstration plant in Güssing (Austria), Proc. 1st International Ukrainian Conference on Biomass For Energy, Kyiv, Ukraine. Brynolf, S., Taljegård, M., Grahn, M., Hansson, J., 2018. Electrofuels for the transport sector: A review of production costs, Renewable and Sustainable Energy Reviews, 81 (Part 2), 1887-905. Gambardella, A. and Yasin, Y., 2017 Power-to-Gas concepts integrated with syngas production through gasification of forest residues, Chalmers university of Technology. Gassner, M., and Maréchal, F., 2012. Thermo-economic optimisation of the polygeneration of synthetic natural gas (SNG), power and heat from lignocellulosic biomass by gasification and methanation, Energy and Environmental Science, 5 (2), 5768-89. Hannula, I. and Kurkela, E., 2012. A parametric modelling study for pressurised steam/O2-blown fluidised-bed gasification of wood with catalytic reforming, Biomass and Bioenergy, 38 (Supplement C), 58-67. Heyne, S. and Harvey, S., 2014. Impact of choice of CO2 separation technology on thermoeconomic performance of Bio-SNG production processes, International Journal of Energy Research, 38 (3), 299-318. Heyne, S. and Harvey, S., 2013. Assessment of the energy and economic performance of second generation biofuel production processes using energy market scenarios, Applied Energy, 101, 203-12. Mivechian, A. and Pakizeh, M., 2013. Hydrogen recovery from Tehran refinery off-gas using pressure swing adsorption, gas absorption and membrane separation technologies: Simulation and economic evaluation, Korean Journal of Chemical Engineering, 30 (4), 937-48. Schlereth, D., 2015. Kinetic and Reactor Modeling for the Methanation of Carbon Dioxide, (Dissertation, Technische Universität München, München. I). Woo C.K., Horowitz, I., Moore, J., Pachecho, A,. 2011. The impact of wind generation on the electricity spot-market price level and variance: The Texas experience, Energy Policy, 39 (7), 3939-44.
Forest residues gasification integrated with electrolysis for production of SNG – modelling and assessment Johan M. Ahlströma*, Simon Harveya, Stavros Papadokonstantakisa a
Chalmers University of Technology, Division of Energy Technology, Maskingränd 7b, 412 58, Gothenburg, Sweden [email protected]
Abstract This study investigates opportunities for integrating an electrolysis unit with a biomass gasifier for production of synthetic natural gas (SNG). Gasification is a key technology for production of biofuels and chemicals from lignocellulosic biomass, for which an increased demand is expected in the future. H2 produced through an electrolyser can be used to increase the output of a gasifier by reaction with CO2 to form CH4. Four integrated flowsheet configurations are evaluated with respect to system energy efficiency and process operating revenue. The system energy efficiencies are in the range of 0.55 – 0.8, and the maximum value of operating revenues is 0.245 $/kWhdry biomass. The results show that feeding the Sabatier reactor with the full product gas flow coming from the methanation unit, and separating the unreacted CO2 afterwards, is the most attractive configuration with respect to operating revenue. Keywords: Power-to-gas, Gasification, Biomass, SNG
1. Introduction Large scale indirect gasification of biomass has been demonstrated at the GoBiGas plant in Gothenburg (Alamia, 2016) and in Güssing (Bolhàr-Nordenkampf et al., 2002). Direct fluidized bed gasification, which is the subject of this work, has been investigated by Hannula and Kurkela (2012) and Gassner and Maréchal (2012). However, no direct blown gasifier with a full scale downstream gas up-grading section have been constructed. When the raw gas from the gasifier is used for methane production, it needs to be upgraded to increase the methane content. The CO2 remaining after the upgrading step has to be removed, which is done in expensive and energy-intensive sequences of separation steps. Hence, converting the remaining CO2 into valuable products could significantly improve the economic performance of the process. For this purpose, a hydrogen source is required. The share of intermittent electricity generation in the power mix is expected to continue to increase significantly. This will induce more volatility in power generation and thereby also in the electricity price (Woo et al., 2011), thereby creating opportunities for variable load applications with a short response time that can help balance the grid. Power-to-gas concepts which produce H2 and O2 when electricity prices are low, can contribute to this balancing. H2 produced by electrolysis can react with CO2 to produce methane through the Sabatier reaction, whereas O2 can be used as a gasifying agent in the direct gasifier unit. In this way, the product output of the process is enhanced while the energy loads of
J.M. Ahlström et al.
110
the air separation unit and the CO2 separation sequence are decreased. To evaluate this concept, it is essential to estimate how the process performance is affected by the amount of H2 fed to the process. This paper investigates four different power-to-gas configurations integrated with a direct, oxygen blown, biomass gasifier. The gasifier is assumed to be fed with forestry logging residues. The configurations are compared with respect to system energy efficiency and operating revenues.
2. Process configuration Forest biomass generally has a moisture content of 40-60%wt, and the fuel must therefore be dried before it is fed to the gasifier to increase efficiency. This study considers the belt dryer concept proposed by Alamia et al. (2015). The moisture remaining in the fuel is vaporized in the gasifier before the fuel volatile components are released. Thereafter, the char gasification starts. At the bottom of the gasifier, combustion of char occurs, which releases the energy that sustains the endothermic reactions in the gasifier. The raw gas leaving the gasifier contains H2, CO2, CO, CH4, H2O, inorganic impurities (e.g. H2S) and organic compounds such as tars. The ash and traces of char in the raw gas are removed in a cyclone, thereafter H2S is removed. In a pre-methanation step the ratio of H2/CO is adjusted through the water-gas-shift reaction. The methanation reaction occurs in a series of three adiabatic, fixed bed, reactors. Figure 2 provides an overview of the basic process flowsheet.
Figure 1. Process flowsheet
After the methanation section, the gas (containing only CH4 and CO2) is fed into the final CO2 removing sequence. Two specification values are considered for the Wobbe index of the SNG product: 44.7-46.7 MJ/Nm3 and 43.9-47.3 MJ/Nm3, corresponding to the A, respectively B standards of the Swedish national gas grid. The Wobbe index essentially limits the concentrations of both CO2 and H2 in the gas product. A-type SNG is produced if possible, since it can be sold at a higher price, otherwise type B SNG is produced instead. Four possible configurations for the final CO2 removal sequence are investigated (see Figure 2): i. The gas from the methanation section is mixed with H2 from the electrolyser and then compressed before being fed to the Sabatier reactor where H2 reacts with CO2 to increase the share of CH4 in the gas. The gas stream is then dried before
Forest residues gasification integrated with electrolysis for production of SNG
ii.
iii.
iv.
111
removing the remaining CO2, in a sequence of two amine-based CO2 separators, each with a removal efficiency of 90%. The yield of the Sabatier reactor entails that there will be H2 left in the gas after the reactor if all CO2 is to be converted. Since configuration (i) does not include a H2 separation sequence and the gas standards limits the concentration of H2 in the gas product, there will always be CO2 in the gas after the Sabatier reactor. H2 is added to the gas mix in sufficient quantity to convert all remaining CO2, thus removing the need for a CO2 removal step. This configuration requires that the fraction of H2 in the gas has to be decreased, which is achieved with a H2 separation unit. As in configuration (i), the gas is dried before the final separation sequence. The separated H2 is recirculated back to the mixing step before the Sabatier reactor. CO2 is separated from the product gas and mixed with H2 in the Sabatier reactor. The produced gas, containing mainly CH4 but also some remaining CO2, is dried and recirculated to the inlet gas stream before the CO2 removal step. Similar to Configuration (iii), with the difference that a H2 separation step is added to the process after the drying step. This provides a process in which all the CO2 in the raw gas can potentially be reacted to methane, since the excess H2 can be removed. This results in a more flexible process configuration.
Figure 2. The four process configurations.
Candidate electrolyser technologies include alkaline and Polymer Electrolyte Membrane (PEM) electrolysers. PEM technology is characterized by a shorter start up time, but a lower efficiency. Alkaline electrolyser technology has reached a higher development level, and was therefore selected for this work. The main difference between the process configurations investigated is the degree of operational flexibility. Configuration (i) is limited by the fraction of CO2 that can be reacted, since there will be H2 in the produced gas if all CO2 is reacted to methane. Configuration (ii) is limited by the absence of CO2 separation units, meaning that the Sabatier reactor must always be fed with enough H2 to fully react the CO2. Thus configuration (i) is more flexible than configuration (ii). In configurations (iii) and (iv), the CO2 is separated before it is reacted with the H2. Here configuration (iv) provides the more flexible option; enough H2 to react all CO2 can be fed to the process, since a H2 separation sequence is included.
J.M. Ahlström et al.
112
3. Methodology All process modelling is performed in Aspen Plus Version 8.8. The biomass feedstock drying model assumes steady-state isothermal operation and negligible pressure drop. The assumed energy required for biomass drying is 0.51 MJ/kg (Alamia et al. (2015)). All simulation results are based on a feed of a 100 kgdry biomass/h, which gives a flowrate of 2.55 kmol CO2/h and 1.52 kmol CH4/h from the methanation unit. The gasifier is modelled to mimic the results published by Hannula and Kurkela (2012), within an error margin of 10% in terms of gas composition. It is assumed that the tar components are decomposed to CO and H2. Pressure drop in the gasifier is assumed to be negligible. The methanation section is modelled assuming that the reaction reaches chemical equilibrium. The model accounts for the significant pressure drop in the methanation section (see Alamia (2016)). The Sabatier reactor is modelled as a plug flow reactor with Langmuir-Hinshelwood-Hougen-Watsonis kinetics, as described by Schlereth (2015). The CO2 and H2 separation sequences are not rigorously modelled; instead the major component recoveries are inferred by literature data (Heyne and Harvey (2014) and Mivechian and Pakizeh (2013)). Similarly, based on Brynolf et al. (2018), a H2 yield of 23.7 kgH2/MWh electricity is assumed for the Alkaline electrolyser. The system energy efficiency suggested by Heyne and Harvey (2013) is used, including the process by-products and commodities, such as electricity and excess heat
ߟ௦௬௦௧ ൌ
σ ݉ሶ ܸܪܮ ܳି ିܧ σ ݉ሶ ܸܪܮ ܳା ܧା
(1)
ṁ indicates the mass flow of either product, p, or of fuel, f. Q is heat and E electricity, and +, refer to energy flows leaving or entering the process. Process stream heating and cooling requirements are used to perform heat targeting using pinch analysis and evaluate the possible heat recovery for the process. The economic performance metric is the operating revenue, obtained by subtracting the cost of biomass feedstock and all utilities from the revenues of selling SNG, process excess heat and the oxygen produced in the electrolyser. Indicative prices for Swedish conditions are assumed; all input data and results are presented in US$. The reference electricity price is set to 0.024 $/kWh and since it is assumed that the electrolyser is run at periods when the electricity price is low, the electricity used for H2 production is assumed to have a 50% lower cost. Revenue from sales of the SNG product is valued at 1.98 $/kg for A-grade quality and 1.5 $/kg for B-grade quality, the biomass price is 23 $/MWh. All other price data, e.g. catalysts, can be found in the report by Gambardella and Yahya (2017). Sensitivity analysis varying the feed of H2, together with the recirculation of CO2 is performed for all configurations except (ii), in which the H2 flow is constant since the process is designed to react all CO2 available in the gas.
4. Results For configuration (ii) the H2 flow is constant at 10 kmol/h, which is the flowrate required to convert all CO2 in the gas mix. The system energy efficiency of configuration (ii) is 0.801 and the operating revenues are 0.245 $/kWhdry biomass. Configuration (ii) results in
Forest residues gasification integrated with electrolysis for production of SNG
113
the highest revenues and a high system energy efficiency, which is due to the fact that all CO2 in the raw gas is converted to CH4. Sensitivity analysis, varying the feed of H2, is performed for configurations (i), (iii) and (iv). The impact on the operating revenues and system energy efficiency is shown in Figure 3.
Figure 3. Total system energy efficiency and operating revenues as a function of CO2 feed. The nuances correspondent to different configurations and the line types correspond to the amounts of H2 feed.
ηsystem, decreases with increased H2 feed and CO2 recirculation rate. This is due to the conversion losses in the electrolyser. The ranges of each curve indicate the cases where the produced gas fulfills the standards of the Wobbe index. The process operating revenues increase for all configurations with H2 feed and CO2 recirculation. This indicates that the additional SNG produced by the increased addition of H2 outweighs the cost of generating the H2. The increase in revenues is essentially linear, except for some rapid increases and decreases in revenues. These rapid changes indicate the thresholds for the types of SNG produced, namely when the model has to change from type A to type B. Only configuration (i) has an increase in revenue that can be achieved without CO2 recirculation, which is due to the fact that there is already CO2 present in the incoming gas flow. For configurations (iii) and (iv), a certain amount of CO2 has to be recirculated to provide the second reactant to the Sabatier reactor.
5. Conclusions The results show that the operating revenue increases with increased addition of hydrogen. This indicates that there is an incentive for integration of an electrolyser unit with the process, provided that the investment costs of the processes does not outweigh the increase of operating revenue. This also suggests that the input feed of H2 should be maximized if this type of power-to-gas concept is implemented. From an economic perspective, the best performing configuration is (ii), which reaches a maximum operating revenue of 0.245 $/kWhdry biomass followed by (i). Even if configuration (ii) outperforms configuration (i) in terms of revenues, it is not necessarily the better alternative; configuration (i) is more flexible, since it allows for different flows of H2, which could be a major advantage if electricity prices fluctuate a lot. Both configurations (iii) and (iv), in which CO2 is separated before it is mixed with the H2, generates lower revenues than the first two configurations. This is a result that highlights that it is more
114
J.M. Ahlström et al.
beneficial to mix the H2 with the raw gas, rather than to separate the CO2 before the reactor.
6. Outlook and further work An essential parameter for future evaluation is the impact of electricity price on the choice of process configuration and design. The model presented in this work constitutes a starting point for such evaluation. To investigate the impact of the electricity price on the process design, a rolling horizon, planning and scheduling optimization algorithm will be developed. By considering the possibility to store hydrogen, and co-running the optimization model with a model of the European electricity system, it becomes possible to determine the optimal process configuration design, accounting for the fluctuations in electricity price. To enable this sort of analysis, it is a necessity to first estimate the investment costs of the different process configurations
7. References Alamia, A., 2016. Large-Scale Production and Use of Biomethane, Chalmers university of Technology. Alamia, A., Ström, H., Thunman, H,. 2015. Design of an integrated dryer and conveyor belt for woody biofuels, Biomass and Bioenergy, 77, 92-109. Bolhàr-Nordenkampf, M., Bosch, K., Rauch, R., Kaiser, S., Tremmel, H., Aichernig, C. & Hofbauer, H., 2002. Scale-up of a 100kWth pilot FICFB-gasifier to a 8 MWth FICFBgasifier demonstration plant in Güssing (Austria), Proc. 1st International Ukrainian Conference on Biomass For Energy, Kyiv, Ukraine. Brynolf, S., Taljegård, M., Grahn, M., Hansson, J., 2018. Electrofuels for the transport sector: A review of production costs, Renewable and Sustainable Energy Reviews, 81 (Part 2), 1887-905. Gambardella, A. and Yasin, Y., 2017 Power-to-Gas concepts integrated with syngas production through gasification of forest residues, Chalmers university of Technology. Gassner, M., and Maréchal, F., 2012. Thermo-economic optimisation of the polygeneration of synthetic natural gas (SNG), power and heat from lignocellulosic biomass by gasification and methanation, Energy and Environmental Science, 5 (2), 5768-89. Hannula, I. and Kurkela, E., 2012. A parametric modelling study for pressurised steam/O2-blown fluidised-bed gasification of wood with catalytic reforming, Biomass and Bioenergy, 38 (Supplement C), 58-67. Heyne, S. and Harvey, S., 2014. Impact of choice of CO2 separation technology on thermoeconomic performance of Bio-SNG production processes, International Journal of Energy Research, 38 (3), 299-318. Heyne, S. and Harvey, S., 2013. Assessment of the energy and economic performance of second generation biofuel production processes using energy market scenarios, Applied Energy, 101, 203-12. Mivechian, A. and Pakizeh, M., 2013. Hydrogen recovery from Tehran refinery off-gas using pressure swing adsorption, gas absorption and membrane separation technologies: Simulation and economic evaluation, Korean Journal of Chemical Engineering, 30 (4), 937-48. Schlereth, D., 2015. Kinetic and Reactor Modeling for the Methanation of Carbon Dioxide, (Dissertation, Technische Universität München, München. I). Woo C.K., Horowitz, I., Moore, J., Pachecho, A,. 2011. The impact of wind generation on the electricity spot-market price level and variance: The Texas experience, Energy Policy, 39 (7), 3939-44.