Carbon-free energy storage is a key enabler of renewable resources, given its role in overcoming the structural intermittency of such resources. Grid-scale storage can be flexibly provided by chemical energy via hydrogen or alternative carbon-neutral vectors, ammonia being a promising candidate. Ammonia possesses some crucial advantages over hydrogen, namely a lower cost per unit of stored energy, higher volumetric energy density, and a widespread production and distribution capacity [1]. Ammonia can indeed be utilized via direct combustion in gas turbines and internal combustion engines although it poses challenges due to its low reactivity and high NOx emissions. This calls for an intense research effort in ammonia combustion both at the experimental and numerical levels. Computational fluid dynamics research on ammonia combustion is still in its infancy and new numerical tools are therefore needed. In this study, we utilize an in-house, high order, low-Mach number reactive flow solver, especially suited for DNS, based on the massively parallel spectral element code nek5000. We utilize the CFD infrastructure on problems specifically targeting ammonia combustion. Given the high hydrogen content and the resulting low effective Lewis numbers of typical reactive ammonia mixtures, the focus is directed towards the analysis of thermodiffusively unstable ammonia flames, exhibiting a characteristic cellular conformation. Such flames are known to greatly affect the global consumption speed due to their wrinkling tendency which may also affect NOx emissions. Twodimensional simulations are performed to analyse the main features of thermodiffusively unstable laminar ammonia flames. We also recently proposed novel data-driven models [2,3] for the sub-grid modelling of such intrinsic instabilities which can prove of great value in LES codes. In this context, we seek DNS datasets that are minimal in size and yet still fully representative of the morphological and propagative features of larger ammonia flames. [1] Valera-Medina, A., et al., Prog. In En. And Comb. Sci. (2018), 69, pp.63-102. [2] Lapenna, P.E., et al., Comb. Th. And Modelling (2021), 25 (6), pp.1064-1085. [3] Lapenna, P.E., et al., Proc. Comb. Inst. (2021), 38(2), pp.2001-2011.
44th Meeting of the Italian section of the combustion institute. Combustion for sustainability / D'Alessio, Francesco; Lapenna, P. E.; Romano, C.; Cerutti, M.; Creta, F.. - 1:44(2022), pp. 78-78. (Intervento presentato al convegno 44th Meeting of the Italian section of the combustion institute tenutosi a Neaples).
44th Meeting of the Italian section of the combustion institute. Combustion for sustainability
Francesco D’Alessio
;P. E. Lapenna;F. Creta
2022
Abstract
Carbon-free energy storage is a key enabler of renewable resources, given its role in overcoming the structural intermittency of such resources. Grid-scale storage can be flexibly provided by chemical energy via hydrogen or alternative carbon-neutral vectors, ammonia being a promising candidate. Ammonia possesses some crucial advantages over hydrogen, namely a lower cost per unit of stored energy, higher volumetric energy density, and a widespread production and distribution capacity [1]. Ammonia can indeed be utilized via direct combustion in gas turbines and internal combustion engines although it poses challenges due to its low reactivity and high NOx emissions. This calls for an intense research effort in ammonia combustion both at the experimental and numerical levels. Computational fluid dynamics research on ammonia combustion is still in its infancy and new numerical tools are therefore needed. In this study, we utilize an in-house, high order, low-Mach number reactive flow solver, especially suited for DNS, based on the massively parallel spectral element code nek5000. We utilize the CFD infrastructure on problems specifically targeting ammonia combustion. Given the high hydrogen content and the resulting low effective Lewis numbers of typical reactive ammonia mixtures, the focus is directed towards the analysis of thermodiffusively unstable ammonia flames, exhibiting a characteristic cellular conformation. Such flames are known to greatly affect the global consumption speed due to their wrinkling tendency which may also affect NOx emissions. Twodimensional simulations are performed to analyse the main features of thermodiffusively unstable laminar ammonia flames. We also recently proposed novel data-driven models [2,3] for the sub-grid modelling of such intrinsic instabilities which can prove of great value in LES codes. In this context, we seek DNS datasets that are minimal in size and yet still fully representative of the morphological and propagative features of larger ammonia flames. [1] Valera-Medina, A., et al., Prog. In En. And Comb. Sci. (2018), 69, pp.63-102. [2] Lapenna, P.E., et al., Comb. Th. And Modelling (2021), 25 (6), pp.1064-1085. [3] Lapenna, P.E., et al., Proc. Comb. Inst. (2021), 38(2), pp.2001-2011.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.