• Energy Research

The auto-ignition process of propane/oxygen mixtures was experimentally and numerically studied over a range of temperatures (850-1250 K) and mixture compositions (from fuel-ultra-lean to fuel-rich conditions) under MILD operative conditions. The mixtures were diluted in CO2 or H2O from 90 up to 97%. The experimental tests were realized in a Tubular Flow Reactor (TFR) at atmospheric pressure. Several combustion regimes were identified as a function of the mixture composition and inlet temper- ature. The experimental results showed that CO2 and H2O significantly alter the ignition process. In par- ticular, a significant slowing of the system reactivity was observed with respect to the mixtures that were diluted in nitrogen. Numerical simulations were performed by commercial codes and detailed kinetic mechanisms. Com- parisons between experimental and numerical results pointed out that kinetic models are not able to cor- rectly reproduce system behaviors in all the experimental conditions. For CO2-diluted mixtures a good agreement between experimental and numerical analysis was obtained for fuel lean mixtures, whereas for stoichiometric and fuel-rich mixtures conditions the consis- tency of predicted data was less satisfactory. In the case of steam-diluted systems, the discrepancy between the experimental data and the predic- tions is about one order of magnitude for any mixture composition, but the model can reproduce the slight dependence of the ignition data on the mixture compositions. Further numerical analyses were performed to identify the reactions controlling the ignition process under MILD operative conditions in presence of CO2 and H2O. Results suggested that steam and carbon dioxide drastically alter the main branching mechanisms as third molecular species in termolecular reactions and/or by decomposition reactions.

The auto-ignition process of propane/oxygen mixtures was experimentally and numerically studied over a range of temperatures (850-1250 K) and mixture compositions (from fuel-ultra-lean to fuel-rich conditions) under MILD operative conditions. The mixtures were diluted in CO2 or H2O from 90 up to 97%. The experimental tests were realized in a Tubular Flow Reactor (TFR) at atmospheric pressure. Several combustion regimes were identified as a function of the mixture composition and inlet temper- ature. The experimental results showed that CO2 and H2O significantly alter the ignition process. In par- ticular, a significant slowing of the system reactivity was observed with respect to the mixtures that were diluted in nitrogen. Numerical simulations were performed by commercial codes and detailed kinetic mechanisms. Com- parisons between experimental and numerical results pointed out that kinetic models are not able to cor- rectly reproduce system behaviors in all the experimental conditions. For CO2-diluted mixtures a good agreement between experimental and numerical analysis was obtained for fuel lean mixtures, whereas for stoichiometric and fuel-rich mixtures conditions the consis- tency of predicted data was less satisfactory. In the case of steam-diluted systems, the discrepancy between the experimental data and the predic- tions is about one order of magnitude for any mixture composition, but the model can reproduce the slight dependence of the ignition data on the mixture compositions. Further numerical analyses were performed to identify the reactions controlling the ignition process under MILD operative conditions in presence of CO2 and H2O. Results suggested that steam and carbon dioxide drastically alter the main branching mechanisms as third molecular species in termolecular reactions and/or by decomposition reactions.

Biomass pyrolysis was extensively studied for its peculiar flexibility with respect to the desired products - bio-oil, bio- gas and char - that can be used as fuels [1, 2], chemicals sources [2], and low cost materials for a wide range of applications [1]. Given the high variability in biomass composition and the presence of inherent metal ions in the biomass, demineralization and metal ions doping procedures were applied to different biomasses and cellulose to study the influence of inorganics on products yield, gas and liquid composition, and structural properties of char [3, 4]. In this framework, less work was done on hemicellulose and lignin. It was found that alkali and earth alkali metals (AAEMs) in the biomass undergoing pyrolysis diverted the decomposition mechanisms of cellulose from depolymerisation reactions to monomers fragmentation, thus altering products distribution in favour of gaseous and condensable lower molecular weight compounds [5]. Considering its polysaccharide nature, hemicellulose was expected to react similarly to cellulose. However, only few works reported about the effect of AAEMs on extracted hemicelluloses or proxy macromolecules [6, 7] despite of their high content in both annual and perennial plants. The chemical variety of hemicelluloses and the difficulty to apply the standard washing or impregnation procedures usually adopted for biomasses and cellulose make this kind of approach difficult to handle. Moreover, despite of the many efforts directed to the numerical modelling of the kinetics of biomass pyrolysis, little attention was given to the impact of AAEMs on the pyrolysis process in terms of reactions involved. Based on the work of Leng et al [8], Ferreiro et al. [9] proposed a modified kinetic mechanism for the pyrolysis of cellulose based on the Bio-PoliMI cellulose sub-mechanism. The modified kinetic mechanism was capable of reproducing the increase of char and gas yield induced by the addition of KCl. However, gas speciation was not well captured. Recently, Ranzi et al. [10] proposed an adjustment in their previous kinetic mechanism [11] to capture the catalytic effect of the ash during the pyrolysis process of biomass. The effect of ash was included in both cellulose and hemicellulose mechanisms. Similarly to cellulose, the presence of ashes in hemicellulose pyrolysis increased the formation of char and dehydration products while the formation of higher molecular weight species was inhibited. In the present work, an experimental campaign on the effect of Na+ on hemicellulose pyrolysis was conducted with the aim of producing a database for the further validation of the updated version of Bio-PoliMI mechanism, more specifically the hemicellulose sub-mechanism. To carry out the experimental work xylan was used as proxy compound for hardwood hemicellulose. Xylan was demineralized through a cation exchange resin in order to reduce the presence of inherent inorganics. Then, the demineralized sample was doped with three different Na+ concentrations, namely 0.1, 0.3 and 0.6 wt. %. Thermogravimetric analyses (TGA) of the samples were performed under inert atmosphere at 7 K/min up to 973 K to obtain TG and DTG curves. Then, pyrolysis tests of the same samples were performed under the same operating conditions as TGA to obtain products yields, gas composition and yields of the main condensable species [9]. The results show that doped and demineralized xylan samples exhibited different pyrolytic behaviours. The gas production was favoured at the expense of pyrolysis liquids and the evolution of the releasing rate of permanent gases (mainly CO2 and CO) along the temperature was greatly altered by the presence of Na+.

Biomass pyrolysis was extensively studied for its peculiar flexibility with respect to the desired products - bio-oil, bio- gas and char - that can be used as fuels [1, 2], chemicals sources [2], and low cost materials for a wide range of applications [1]. Given the high variability in biomass composition and the presence of inherent metal ions in the biomass, demineralization and metal ions doping procedures were applied to different biomasses and cellulose to study the influence of inorganics on products yield, gas and liquid composition, and structural properties of char [3, 4]. In this framework, less work was done on hemicellulose and lignin. It was found that alkali and earth alkali metals (AAEMs) in the biomass undergoing pyrolysis diverted the decomposition mechanisms of cellulose from depolymerisation reactions to monomers fragmentation, thus altering products distribution in favour of gaseous and condensable lower molecular weight compounds [5]. Considering its polysaccharide nature, hemicellulose was expected to react similarly to cellulose. However, only few works reported about the effect of AAEMs on extracted hemicelluloses or proxy macromolecules [6, 7] despite of their high content in both annual and perennial plants. The chemical variety of hemicelluloses and the difficulty to apply the standard washing or impregnation procedures usually adopted for biomasses and cellulose make this kind of approach difficult to handle. Moreover, despite of the many efforts directed to the numerical modelling of the kinetics of biomass pyrolysis, little attention was given to the impact of AAEMs on the pyrolysis process in terms of reactions involved. Based on the work of Leng et al [8], Ferreiro et al. [9] proposed a modified kinetic mechanism for the pyrolysis of cellulose based on the Bio-PoliMI cellulose sub-mechanism. The modified kinetic mechanism was capable of reproducing the increase of char and gas yield induced by the addition of KCl. However, gas speciation was not well captured. Recently, Ranzi et al. [10] proposed an adjustment in their previous kinetic mechanism [11] to capture the catalytic effect of the ash during the pyrolysis process of biomass. The effect of ash was included in both cellulose and hemicellulose mechanisms. Similarly to cellulose, the presence of ashes in hemicellulose pyrolysis increased the formation of char and dehydration products while the formation of higher molecular weight species was inhibited. In the present work, an experimental campaign on the effect of Na+ on hemicellulose pyrolysis was conducted with the aim of producing a database for the further validation of the updated version of Bio-PoliMI mechanism, more specifically the hemicellulose sub-mechanism. To carry out the experimental work xylan was used as proxy compound for hardwood hemicellulose. Xylan was demineralized through a cation exchange resin in order to reduce the presence of inherent inorganics. Then, the demineralized sample was doped with three different Na+ concentrations, namely 0.1, 0.3 and 0.6 wt. %. Thermogravimetric analyses (TGA) of the samples were performed under inert atmosphere at 7 K/min up to 973 K to obtain TG and DTG curves. Then, pyrolysis tests of the same samples were performed under the same operating conditions as TGA to obtain products yields, gas composition and yields of the main condensable species [9]. The results show that doped and demineralized xylan samples exhibited different pyrolytic behaviours. The gas production was favoured at the expense of pyrolysis liquids and the evolution of the releasing rate of permanent gases (mainly CO2 and CO) along the temperature was greatly altered by the presence of Na+.

[object ObjectMILD combustion is a very attractive technology because of its intrinsic features for energy production from diluted gas deriving from bio- or thermochemical degradation of biomass. An effective use of such a technology for diluted fuel requires a thorough analysis of ignition and oxidation behavior to highlight the potential effects of the different fuel components on the basis of temperature and diluent/oxygen/fuel mixture composition. In this work, ignition and oxidation of a model gas surrogate for the gaseous fraction of biomass pyrolysis products containing C1-C2 species, CO and CO2 were experimentally and numerically studied over a wide range of temperature and overall composition in the presence of large amounts of CO2 or H2O. Experimental results showed that such species significantly alter the evolution of the ignition process in dependence on temperature range and mixture composition. Several kinetic models were tested to simulate experimental results. Significant discrepancies occur, especially in the case of steam dilution. Numerical analyses suggested that such diluents acted mainly as third body species at low temperatures, conditioning both radical production pathways and the relative weight of C1 oxidation/recombination routes, while strongly interacting with the H2/O2 high temperature branching mechanisms at high temperatures. Further analyses are mandatory to improve the predictability of the models and extend the applicability of the chemical schemes to non-standard conditions.

[object ObjectMILD combustion is a very attractive technology because of its intrinsic features for energy production from diluted gas deriving from bio- or thermochemical degradation of biomass. An effective use of such a technology for diluted fuel requires a thorough analysis of ignition and oxidation behavior to highlight the potential effects of the different fuel components on the basis of temperature and diluent/oxygen/fuel mixture composition. In this work, ignition and oxidation of a model gas surrogate for the gaseous fraction of biomass pyrolysis products containing C1-C2 species, CO and CO2 were experimentally and numerically studied over a wide range of temperature and overall composition in the presence of large amounts of CO2 or H2O. Experimental results showed that such species significantly alter the evolution of the ignition process in dependence on temperature range and mixture composition. Several kinetic models were tested to simulate experimental results. Significant discrepancies occur, especially in the case of steam dilution. Numerical analyses suggested that such diluents acted mainly as third body species at low temperatures, conditioning both radical production pathways and the relative weight of C1 oxidation/recombination routes, while strongly interacting with the H2/O2 high temperature branching mechanisms at high temperatures. Further analyses are mandatory to improve the predictability of the models and extend the applicability of the chemical schemes to non-standard conditions.