Autore FURTADO AMARAL, ANDRE
Relatore MANENTI, FLAVIO
Coordinatore FRASSOLDATI, ALESSIO
Tutor MELE, ANDREA
Data 28-ott-2019
Titolo della tesi Multiscale design, integration & optimization of biorefineries for the production of liquid fuels
Abstract in italiano Biorefineries have been conceptualized to substitute the traditional oil refineries, producing heat, electricity and chemicals (among which, liquid fuels) from biomass. In this work, these processes were studied from a multiscale perspective using computer simulations. Four different software were used: commercial process simulator Aspen HYSYS; energy integration software Aspen Energy Analyzer; GasDS, a gasification / pyrolysis simulator and the MATLAB programming environment. Lignocellulosic biomass gasification was described with the aid of a detailed phenomenological model. In this model, biomass is considered a mixture of cellulose, hemicellulose and three surrogate compounds that account for the most abundant monomers in lignin. Biomass composition was determined from an innovative data fitting method based on Lagrange multipliers. The calculated biomass composition produces lower heating values (LHVs) that are consistent with experimental observations. The relative LHV error was not bigger than 10% for any of the biomasses studied. The developed method represents an improvement from the previous 'triangle model', especially because it uses experimental information in a more systematic approach to quantify biomass composition. An entrained flow gasifier was simulated using a detailed, phenomenological model, implemented in the GasDS program. The model considered a kinetic mechanism based on the above mentioned results on biomass composition to successfully predict biomass conversion and syngas yield for a given oxygen consumption. Almond shells and olive pits were the two biomasses with the biggest syngas yield per oxygen input, with a value of 314 mol syngas / mol Oxygen. Biomass conversion values compared well with experimental values and were close to chemical equilibrium. The simulator displayed numerical instabilities during the unsteady state operation, due to the strategy used to increase the step size. This effect is not present during steady state operation and, therefore, does not influence these results. The coproduction of heat, electricity and chemicals from second-generation biomass was assessed. Two different scale sizes were considered, with biomass lower heating value inputs of 1 and 100 MW, respectively. These scales are representative of decentralized and centralized production concepts, each of which with its own characteristic transformation pathways. For the centralized concept, biomass gasification was considered. Two final uses for syngas were considered: production of methanol and production of Fischer-Tropsch (FT) fuels. The FT product distribution model considered oleffin readsorption and it was solved using an innovative power series solution. Methanol production is the superior process, both in economic and in terms of final conversion to liquid fuels. The economics of the Fischer-Tropsch process suffers due to the low energetic yield of the reaction in terms of high valued liquid products. It remains to be confirmed (1) whether if the correlations used are adequate to represent the FT reaction system and (2) if further income could be expected if the other reaction products could be sold as high value products. Both processes are economically unfeasible, with product costs that range (approximately) from 60 to 90 /MWh (MeOH) and 80 to 210 €/MWh (FT). Even so, methanol production is an interesting alternative to current biogas concepts. The minimum subsidy cost of this process ranges from half to one third of current biogas subsidy costs. The decentralized utilization concept considers the anaerobic digestion of biomass for the production of biogas. Three biogas processes were assessed: HPC (biogas to methanol), BioCH4 (biogas to biomethane) and CHP (biogas to heat & electricity). The last two processes are already used commercially with the aid of subsidy policies. The economic analysis indicates that, without these policies, none of these attain self-sustainability due to high overall manufacturing costs; the estimated minimum support cost (MSCs) were 108, 62 and 110 €/MWh for the HPC, BioCH4 and CHP processes, respectively. The model could explain currently practised government subsidies in Italy and Germany. It was seen that the newly proposed HPC process is economically comparable to the traditional CHP process. Therefore, the HPC process is a possible alternative to biogas usage. A subsidy policy was proposed: 50, 66, 158 and 148 €/MWh for available heat, methane, electricity and methanol (respectively). The proposed policy results in a 10% OpEx rate of return for any of the processes, thus avoiding a disparity in the production of different products.
Abstract in inglese Biorefineries have been conceptualized to substitute the traditional oil refineries, producing heat, electricity and chemicals (among which, liquid fuels) from biomass. In this work, these processes were studied from a multiscale perspective using computer simulations. Four different software were used: commercial process simulator Aspen HYSYS; energy integration software Aspen Energy Analyzer; GasDS, a gasification / pyrolysis simulator and the MATLAB programming environment. Lignocellulosic biomass gasification was described with the aid of a detailed phenomenological model. In this model, biomass is considered a mixture of cellulose, hemicellulose and three surrogate compounds that account for the most abundant monomers in lignin. Biomass composition was determined from an innovative data fitting method based on Lagrange multipliers. The calculated biomass composition produces lower heating values (LHVs) that are consistent with experimental observations. The relative LHV error was not bigger than 10% for any of the biomasses studied. The developed method represents an improvement from the previous 'triangle model', especially because it uses experimental information in a more systematic approach to quantify biomass composition. An entrained flow gasifier was simulated using a detailed, phenomenological model, implemented in the GasDS program. The model considered a kinetic mechanism based on the above mentioned results on biomass composition to successfully predict biomass conversion and syngas yield for a given oxygen consumption. Almond shells and olive pits were the two biomasses with the biggest syngas yield per oxygen input, with a value of 314 mol syngas / mol Oxygen. Biomass conversion values compared well with experimental values and were close to chemical equilibrium. The simulator displayed numerical instabilities during the unsteady state operation, due to the strategy used to increase the step size. This effect is not present during steady state operation and, therefore, does not influence these results. The coproduction of heat, electricity and chemicals from second-generation biomass was assessed. Two different scale sizes were considered, with biomass lower heating value inputs of 1 and 100 MW, respectively. These scales are representative of decentralized and centralized production concepts, each of which with its own characteristic transformation pathways. For the centralized concept, biomass gasification was considered. Two final uses for syngas were considered: production of methanol and production of Fischer-Tropsch (FT) fuels. The FT product distribution model considered oleffin readsorption and it was solved using an innovative power series solution. Methanol production is the superior process, both in economic and in terms of final conversion to liquid fuels. The economics of the Fischer-Tropsch process suffers due to the low energetic yield of the reaction in terms of high valued liquid products. It remains to be confirmed (1) whether if the correlations used are adequate to represent the FT reaction system and (2) if further income could be expected if the other reaction products could be sold as high value products. Both processes are economically unfeasible, with product costs that range (approximately) from 60 to 90 /MWh (MeOH) and 80 to 210 €/MWh (FT). Even so, methanol production is an interesting alternative to current biogas concepts. The minimum subsidy cost of this process ranges from half to one third of current biogas subsidy costs. The decentralized utilization concept considers the anaerobic digestion of biomass for the production of biogas. Three biogas processes were assessed: HPC (biogas to methanol), BioCH4 (biogas to biomethane) and CHP (biogas to heat & electricity). The last two processes are already used commercially with the aid of subsidy policies. The economic analysis indicates that, without these policies, none of these attain self-sustainability due to high overall manufacturing costs; the estimated minimum support cost (MSCs) were 108, 62 and 110 €/MWh for the HPC, BioCH4 and CHP processes, respectively. The model could explain currently practised government subsidies in Italy and Germany. It was seen that the newly proposed HPC process is economically comparable to the traditional CHP process. Therefore, the HPC process is a possible alternative to biogas usage. A subsidy policy was proposed: 50, 66, 158 and 148 €/MWh for available heat, methane, electricity and methanol (respectively). The proposed policy results in a 10% OpEx rate of return for any of the processes, thus avoiding a disparity in the production of different products.
Tipo di documento Tesi di dottorato
Appare nelle tipologie: Tesi di Dottorato
File allegati
File Dimensione Formato  
Thesis PhD AFA.pdf

accesso aperto

5.51 MB Adobe PDF Visualizza/Apri

I documenti in POLITesi sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: http://hdl.handle.net/10589/149532