Theoretical and numerical analysis of isolated ethanol droplets: evaporation and combustion
- Autores: Alejandro Millán Merino
- Directores de la Tesis: Eduardo Fernández Tarrazo (dir. tes.), Mario Sánchez Sanz (codir. tes.)
- Lectura: En la Universidad Carlos III de Madrid ( España ) en 2020
- Idioma: español
- Tribunal Calificador de la Tesis: Javier Manuel Ballester Castañer (presid.), Pierre Boivin (secret.), Francesco Saverio Marra (voc.)
- Programa de doctorado: Programa de Doctorado en Mecánica de Fluidos por la Universidad Carlos III de Madrid; la Universidad de Jaén; la Universidad de Zaragoza; la Universidad Nacional de Educación a Distancia; la Universidad Politécnica de Madrid y la Universidad Rovira i Virgili
- Materias:
- Enlaces
- Tesis en acceso abierto en: e-Archivo
- Resumen
On Thursday 28 November 2019, the European Union (EU) Parliament declared a global climate and environmental emergency, with 429 out of 654 votes in favors of immediately taking action against global warming by implementing ambitious actions to limit the average temperature rise to 1.5°C. The commitment of the EU implies a reduction of 55% in the CO2 emissions by 2030 to become climate neutral by 2050. As indicated by the BP Energy Outlook 2019, the world final energy demand by end users is expected to increase in the next decades, with around 80% of this energy coming from carbon-containing fuels: Oil, Natural Gas and Coal. The data published by the EU revealed that the share of renewable energy in 2017 was 17.5%, with the total energy consumption growing for third year in a row to reach 1561 Mtoe, in line with a worldwide trend that augmented CO2 emissions in a 2.7% in 2018 with respect to the values of 2017.
Even if the plan of the EU is achieved, by 2030 only 27% of all the energy consumed will come from Renewable Energy Sources (RES). The remaining energy will necessarily come from the increasingly large range of greener fuels (hydrogen, ammonia, biofuels, synthetic natural gas and other synthetic fuels as DME), with a much better carbon footprint than heavy hydrocarbons (gasoline, diesel and coal), and they should be used as efficiently as possible to achieve the target of reducing by 55% the emissions of greenhouse gases by 2030.
In this thesis we focus in ethanol C2H5OH, byproduct of plant fermentation that is been extensively used in engines to lower Green House Gas (GHG) emissions. In most practical applications, the fuel or mix of fuels are injected as a spray inside the combustion chamber, forming a cloud of small droplets that evaporates, first, mixing with the surrounding ambient air before the flame is formed either by autoignition or with the help of an external source of heat. In this thesis we will study in detail the transient one-dimensional evaporation and combustion of an isolated ethanol droplet immersed in a hot and humid atmosphere.
In Chapter 2 of the thesis we formulate the problem, writing the system of equations that describes the whole evaporation and combustion process, using first principles only and, intentionally, avoiding the utilization of semi-empirical correlations or ad-hoc fitting parameters often used in the literature to improve the matching between numerical simulations and experimental measurements.
Chapter 3 deals with pure evaporation of spherically-symmetric multicomponent ethanol droplets in an inert ambient. After performing an exhaustive validation of both the model and the numerical code used to integrate the system of equations, we theoretically determined the different characteristic times of the evaporation process setting the limits of validation of the classical d2-law, according to which the square of the droplet diameter decays linearly with time. This theory is valid in the limit in which the heat conductive time in both the liquid and gas phases is much shorter than the droplet lifetime. Even though this limit is valid in small droplets, we demonstrate in this chapter, using both theoretical and numerical arguments, that this hypothesis is only valid for sufficiently small droplets or ambient temperatures close to the boiling temperature of the liquid phase, when the radiation from the gas surrounding the droplet can be neglected. As the droplet radius and/or the ambient temperature are increased, radiative heating gains relevance, deviating the droplet evaporation rate from the predictions of the d2-law. Moreover, the characteristic radiation time, defined as the time needed to evaporate the droplet using only the heat radiated from the ambient, becomes comparable to the characteristic heat conductive time of the liquid phase, suggesting that the heating of the liquid fuel is not quasi-steady. This chapter also analyses the effect of water content and ambient moisture in the evaporation rate of ethanol droplets. Their effect is opposed regarding the vaporization rate of ethanol, with moisture reducing the time needed to vaporize the liquid fuel. On the other hand, both the presence of ambient humidity and water in the droplet content would increase the droplet lifetime, mainly as a consequence of a lower vaporization rate induced by the large concentration of water in the droplet during the last stages of the vaporization process.
In Chapters 4 and 5 we turned our attention to droplet autoignition and combustion. Droplet combustion is a complex unsteady phenomena that involves autoignition and rich and lean unsteady premixed and diffusion flames. The description of such complicate phenomena requires the utilization of detailed combustion chemical kinetics, hundreds of elementary reaction steps and radical species to make the calculations accurate but, also, computationally very expensive. To cut the cost of the calculations, it is normal to use skeletal or reduced chemical schemes. Nevertheless, they are usually developed to describe some specific combustion configurations that are not capable of describing the complex dynamics undergone by the flame after the sudden autoignition event. For that reason, Chapter 4 is dedicated to develop a multipurpose reduced chemical scheme, that involves only 14 overall steps among 16 reactive chemical species, capable of describing autoignition, premixed and diffusion flames and extinction. The new mechanism is derived by introducing chemical-kinetic steady-state approximations for the relevant intermediate species, using, as starting point, a skeletal mechanism of 66 steps and 31 chemical species. The reduced chemistry description here described achieved computation time savings of 93 % when compared with the computational time of the detailed mechanism.
Finally, in Chapter 5 the reduced mechanism is used to study the conditions that would lead to droplet autoignition. During the simulations it was found that the reduced mechanisms developed in Chapter 4 predicted accurately autoiginition times but, after ignition, during the transition to the quasi-steady droplet combustion, the simulations with the reduced mechanism stalled, giving nonphysical values of species concentrations. Detailed analysis {of the flame structure} revealed that the quasi-steady hypothesis fails for the intermediate radical α-hydroxyethyl. After taking this radical out of the steady state, a modified and more robust reduced mechanism of 15 steps and 17 reactive chemical species was capable of completely describing the unsteady combustion of ethanol droplets.
Our numerical results depicted a map that defines the critical ambient temperature T_∞^c below which autoignition can not take place. In ambient temperatures below T_∞^c, the droplet evaporates completely before ignition. For sufficiently large ambient temperatures T_∞ > T_∞^c, the autoignition event suddenly raised the temperature to form a lean premixed flame that propagates towards the droplet surface consuming all the available oxygen before bouncing back as a diffusion flame, that evolves to reach a quasi-steady state.
Additionally, we analyzed the effect of the droplet water content and of the ambient moisture and their effect on the autoignition time, to discover that ambient humidity accelerates droplet autoignition while the initial droplet water content delays the beginning of combustion. The analysis extends to describe the structure of the quasi-steady flame structures that emerge during the last stage of the combustion process. The flame is located at a distance that ranges between three and seven times the initial droplet radius, comparable to the inter-droplet distance in sprays. Surprisingly, the reaction region turns out to be much thicker than expected, with a chemically active region that spans distances of the order of several droplet diameters. This result contrasts with previous results and questions a great number of theoretical and asymptotic studies that assumed infinitely thin reaction regions.
