Devolatilization is a heat-driven thermochemical process in which a liquid or a solid fuel releases mass in the form of volatile compounds after drying, as a result of the combination of endothermic and exothermic reactions. It differs from pyrolysis in that it does not require an inert atmosphere and that the reactant must be either solid or liquid. Devolatilization is present in every industrial process involving high temperature thermochemical conversion of solid fuels, such as combustion or gasification.

Biomass devolatilization is a complex process which entangles dynamic changes in internal heat transfer with phase change and chemical kinetics. The higher the heating rate and the temperature at which devolatilization takes place, the more uncertain the outcome of these processes and the more challenging to measure. Furthermore, since devolatilization involves heat and mass transfer to its surroundings, this can also have an effect on the external conditions, for example by altering the surrounding gas temperature or composition, or by transferring momentum to the particle or the surrounding gas. The inherent uncertainty in biomass thermochemical properties and composition difficult the fundamental understanding even further.

Suspension firing of pulverised fuels is a technique applied to high temperature thermochemical conversion processes, in which a particle-laden gas flow is injected into a hot atmosphere. Due to the extreme heat transfer conditions, the particles exhibit a very fast heating rate and achieve quickly a very high temperature during the stage at which they devolatilize. Measurements of mass loss and composition under these conditions are difficult to achieve using laboratory equipment, such as thermogravimetric analysers.

The aim of this PhD thesis is to investigate the devolatilization of biomass particles under high heating rate conditions, by measuring their morphology and velocity dynamics while reacting in a hot laminar gas flow. The measurement techniques applied, allow a very fast sampling rate and, while they prevent a fundamental understanding of the underlying mechanisms of high temperature devolatilization. They provide valuable and practical knowledge that can be applied to realistic conditions and allow the introduction of uncertainties in biomass size, shape and composition. 

The work carried for this thesis provides experimental measurements of particle size, shape and velocity for a devolatilizing stream of biomass particles, using high-speed imaging diagnostics. An interesting phenomenon related to the interaction between devolatilization reactions and particle momentum is investigated in detail. Additional modeling and simulation work is provided, to assess the model’s performance and the importance of this phenomenon. Particle dispersion caused by this phenomenon is compared to the one achievable by active flow manipulation techniques, such as swirling and vortex generation.

This work provides important information regarding the complex fluid-solid interactions caused by the dynamics of biomass devolatilization from a stochastic and modelling and simulation point of view. Although these results are not directly applicable to industrially-realistic conditions, the methodology for this investigation can be applied to more complex flows and further work must be conducted to understand the mechanisms behind the phenomenon observed and the consequences for devolatilization. 

The use of by-products from forestry and agricultural sectors can increase the bioenergy share for heat/power production and industrial processes. Moreover, the integration with carbon capture technologies has a significant potential for CO₂ reduction with BECCS (bioenergy with carbon capture and storage) technologies. Entrained flow reactors (EFRs) are commonly applied in the direct combustion and gasification of pulverized fuels. In both technologies, particle-laden flow characteristics can significantly influence the reactor operation, with an impact on performance and emissions. This thesis investigates a broad range of particle flow parameters in EFRs, with an experimental analysis combining high-speed imaging methods with sampling techniques. A comprehensive analysis was carried out using different biomass feedstocks (sawdust, pine bark, and rice husk), operating conditions (non-reacting, air and oxygen-enriched combustion, and gasification), and flow manipulation techniques (swirling flow and acoustic forcing).

The latter technique, acoustic forcing, resulted in a high potential for soot reduction in previous experiments when applied to biomass injection in small lab-scale reactors under laminar conditions. Soot emissions represent important environmental concerns and a major technical problem due to the required downstream cleaning processes. For this reason, acoustic forcing was further studied in this work using a larger pulverized swirl burner. Post-processed shadowgraph images from cold-flow experiments provided insights into the near-field particle distribution and quantified particle dispersion in a broad range of operating conditions. Particle dispersion increased near-linearly with the pressure amplitude of the acoustic forcing, which presented the strongest effect followed by the swirl intensity of the secondary air. Both techniques applied simultaneously had a synergetic effect, especially for small particle size (e.g. dispersion angle increased from 0.9 to 9.1° for particles in the size range of 63-112 μm).

High particle dispersion significantly reduced the flame liftoff distance (ignition characteristic) during combustion, which was identified by the high-speed imaging technique. The reduction in liftoff distance, caused by the acoustic forcing in combustion conditions, varied from 6 to 28%. Higher reduction was identified for high oxygen level enrichment and small particles. Acoustic forcing applied at conditions with low secondary air momentum flux resulted in lower CO emissions and higher combustion efficiency, with higher NO emissions. Under gasification, the ignition occurred at earlier stages than in combustion as demonstrated by the changes in liftoff distance, which was strongly affected by the producer gas recirculation (containing CO and H2). The acoustic forcing presented a sharper effect on liftoff in such conditions, decreasing by 42% at low equivalence ratios (λ of 0.4). Moreover, acoustic forcing increased cold-gas efficiency by 12%, by increasing the yields of CO and H₂.

Particle emissions were characterized by particulate matter (PM) isokinetic sampling and coarse particle collection with further thermogravimetric, elemental, and particle size distribution analysis. The coarse particles presented a small reduction of carbon content for combustion conditions under acoustic forcing. In gasification conditions, acoustically forced cases presented up to 25% lower PM emissions, while coarse particle emissions increased substantially. Ultimate and thermogravimetric analysis suggests that soot was an important component of the PM emissions. Coarse particles during gasification mainly consisted of fragmented char, which yield increased with acoustic forcing, apparently due to high velocities imposed on the particles around the flow centerline, which gave them a shorter residence time at high temperatures.

Experiments in a larger scale reactor, with 100 kW thermal capacity, were used for studies focused on the particle emissions and deposition from high-temperature oxygenenriched combustion of rice husks. A completely different ash morphology was identified in such experiments, which mainly presented coarse ash fraction deposit build-ups with high Si content and minor ash-forming elements. These characteristics can be beneficial both for bioenergy applications and ash valorization processes. The current work brings new experimental results of EFRs under different particle-laden flow characteristics. The implications in particle dispersion, flame morphology, and emissions could be addressed in further investigations, from fundamental aspects to optimization of burners of EFRs.