Woody biomass is being considered a potential co-firing fuel to reduce coal consumption in iron-ore pelletizing rotary kilns. An important consideration is the slagging inside the kiln caused by ash deposition that can lead to process disturbances or shutdowns. In terms of ash chemistry, co-firing woody biomass implies the addition of mainly Ca and K to the Si- and Al-dominated coal-ash (characteristic of high-rank coals) and Fe from the iron-ore that are both inherent to the process. An alkali-laden gaseous atmosphere is also present due to the accumulation of alkali via the recirculation of flue gas in the system. The slagging propensity of blending woody biomass with coal in the grate-kiln process was studied based on the viscosity of the molten phases predicted by global thermochemical equilibrium modeling. This was carried out for variations in temperature, gaseous KOH atmosphere, and fuel blending levels. Results were evaluated and compared using a qualitative slagging indicator previously proposed by the authors where an inverse relationship between deposition tendency and the viscosity of the molten fraction of the ash was established. The results were also compared with a set of co-firing experiments performed in a pilot-scale (0.4 MW) experimental combustion furnace. In general, the co-firing of woody biomass would likely increase the slagging tendency via the increased formation of low-viscosity melts. The fluxing behavior of biomass-ash potentially reduces the viscosity of the Fe-rich aluminosilicate melt and intensifies deposition. However, the results also revealed that there are certain conditions where deposition tendency may decrease via the formation of high-melting-point alkali-containing solid phases (e.g., leucite). 

Biomass fast pyrolysis vapors contain a significant quantity of persistent aerosols, which can impact downstream processing by e.g. fouling of surfaces and deposition on downstream catalysts. In this study, aerosol concentrations and size distributions were measured by an impactor in two pyrolysis systems, a bench-scale fluidized-bed pyrolyzer and a pilot-scale cyclone pyrolyzer. In both units, the mass-based mode aerosol diameter was approximately 1 μm before aerosol collection devices in cooled vapors of 300–370 K but the number-based median was < 0.1 μm. Aerosols < 1 μm were formed and aerosols > 1 μm deposited during cooling of pyrolysis vapors from 620 to 370 K in the fluidized-bed pyrolysis system. The oil fraction collected from the aerosols constituted approximately 40 wt% of the total oils collected in both systems. Compared to the total collected oil, the oil fraction from the aerosols was enriched in lignin-derived components and anhydrosugars and had lower concentrations of low molecular weight cellulose derived oxygenates, such as hydroxyketones. 

Induration (or sintering) of iron-ore pellets requires high temperature (∼1300 °C), which today is generated by burning fuel oil in the firing zone of the straight-grate plant (SG) or coal in the rotary kiln of grate-kiln (GK) plants. In this study, ∼150 kWth combustion experiments were used to investigate the opportunity of totally replacing fuel oil with H2 or pyrolysis oil and replacing coal with wood pellets or black pellets powder. For SG plants, the fuel oil can probably be replaced with either H2 or pyrolysis oil without any major concerns, except for slightly or much higher NOx emissions in the case of pyrolysis oil and H2, respectively. For GK induration machines, it is probably challenging to replace coal entirely with biomass since the temperature profile will be different, and there is a risk for increased ash related operational problems. For both SG and GK plants, the slightly lower O2 concentration in the flue gas observed during biomass combustion (pyrolysis oil, wood pellets, and black pellets) may, however, be negative for the induration process, and this needs to be clarified in future research.

Iron-ore pelletizing plants use high-rank coals to supply the heat necessary to process ores. Ash material from coal, in combination with iron-ore dust originating from the disintegration of the pellets, can cause deposition/slagging which often leads to severe production losses and damage. Deposition/slagging is most prominent in the hot areas of the grate-kiln setup and is more severe at the inlet of the rotary-kiln, i.e., the transfer-chute. Following on from our previous work, high-rank bituminous coals with potential for use in the pelletizing process were combusted in a pilot-scale (0.4 MW) pulverized-coal fired experimental combustion furnace (ECF). The fly-ash particles and short-term deposits were characterized to shed light on the observed difference in slagging/deposition tendencies of the coals. Global thermodynamic equilibrium modeling, in combination with viscosity estimates, was used to interpret the experimental findings and investigate the effect of the coal-ash composition upon deposition/slagging. This approach was carried out with and without the presence of Fe2O3-rich pellet-dust under oxidizing conditions within the temperature range at the transfer-chute of iron-ore pelletizing rotary-kilns. Based on the findings, a Qualitative Slagging Indicator (QSI) was proposed that can help pre-screen new solid fuels for potential slagging issues. The proposed QSI highlights the following: (1) an inverse relationship between viscosity and slagging/deposition tendency of the coals was observed (2) as viscosity decreases (either with increasing temperature or due to the change in the coal-ash composition), stronger deposits will form that will complicate the mechanical removal of the deposited layer. It was therefore inferred that low viscosity molten phases facilitate deposition/slagging, which is exacerbated by the presence of fluxing agents (e.g., CaO, MgO, K2O, Na2O, and Fe2O3) in the deposits. The low viscosity coal-ash-induced molten phases are also more likely to interact with the Fe2O3-rich pellet-dust that results in further decreases in viscosity, thereby intensifying depositions. The results from this work complement the on-going research by our group to elucidate and alleviate ash-related problems in industrial grate kilns.

The objective of this work was to move towards developing a comprehensible Computational Fluid Dynamics (CFD) model to facilitate the predictive modeling of Fast Pyrolysis Oil (FPO) spray combustion. A CFD model was implemented from the literature and results were compared to 2D data from non-intrusive optical diagnostics involving Planar Laser Induced Fluorescence of the OH radical, Mie scattering imaging and two-color pyrometry using a laboratory-scale, CH 4 /air flat-flame with an air-assist atomizer. Furthermore, flame radiation and contributions from graybody sources, chemiluminescence and soot were studied experimentally using emission spectroscopy and Laser Induced Incandescence (LII). Reasonable qualitative agreement was found between experimental and model results in terms of flame structure and temperature. Emission spectroscopy and LII results revealed and confirmed earlier observations regarding the low soot concentration of FPO spray flames; furthermore, it was shown that a significant portion of flame radiation originated from graybody char radiation and chemiluminescence from the Na-content of the FPO. These suggest that the treatment of soot formation might not be important in future computational models; however, the description of char formation and Na chemiluminescence will be important for accurately predicting temperature and radiation profiles, important from the point of e.g., large-scale power applications. Confirmed low soot concentrations are promising from an environmental point of view.

The oxidation of soot is a complex process due to the heterogeneous structure of the material. Several mechanisms have been hypothesized based on ex situ studies, but need confirmation from in situ observation; furthermore, deeper insight is needed to develop and validate structure-dependent reaction mechanisms. In this work, soot oxidation was for the first time observed at atomic scale in situ, in real-time, using a spherical aberration-corrected Environmental Transmission Electron Microscope. The transformation of individual soot particles was followed through from initiation to complete conversion. Observations clearly showed the existence of different burning modes and particle fragmentation previously hypothesized in the literature. Furthermore, transitioning between the modes—affected by temperature and O2 pressure—was unambiguously observed, explaining previous observations regarding structure-dependent and time-varying oxidation rates. A new mode of burning in which oxidation happens rapidly in the bulk phase with the disruption of long-range lamellar order was observed and is suspected to be dominant at practically relevant conditions. The ability to unambiguously relate different burning modes in terms of nanostructure will be of importance for optimizing both soot emission abatement and properties of nanoparticulate carbon products.

In this work, we are the first to report a detailed comparison between the predictions of a current Computational Fluid Dynamics (CFD) model for describing Fast Pyrolysis Oil (FPO) spray combustion and results from a laboratory-scale experiment. The objectives were to assess the predictive power of the CFD model, evaluate its usefulness in a numerical optimization scenario and characterize the spray flame. The spray flame was produced by using an air-assist atomizer piloted by a CH4/air flat-flame. Pyrolysis oil from a cyclone fast pyrolysis plant was combusted. The flame was characterized by using two-color pyrometry, Tunable Diode Laser Absorption Spectroscopy and high-magnification shadowgraphy. Overall, the assessed model correctly predicted flame structure and seemed appropriate for engineering applications, but lacked predictive power in estimating droplet size distributions. Numerical results were the most sensitive to variations in the initial droplet size distribution; however, seemed robust to changes in the multicomponent fuel formulation. Several conclusions were drawn regarding FPO spray combustion itself; e.g., the amount of produced soot in the flames was very low and droplets exhibited microexplosion behavior in a characteristic size-shape regime. 

Combustion facilities run on pulverized biomass often exhibit fluctuations in fuel feeding and, thus, equivalence ratio and would benefit from fast process control based on optical λ sensors installed in the reactor core. The conversion of softwood powder is investigated in an atmospheric entrained-flow reactor (EFR) operated close to stoichiometry using two different burners (swirl and jet) and three oxygen concentrations (21, 30, and 40%). Tunable diode laser absorption spectroscopy (TDLAS) is used to conduct time-resolved (0.1-1 s) in situ measurements of the gas temperature, carbon monoxide (CO), water vapor (H2O), gaseous atomic potassium [K(g)], and soot volume fraction in the lower part of the reactor core and in the exhaust of the EFR. At both locations, the measurement parameters show significant, correlating fluctuations. The local equivalence ratio is derived from a comparison of measured CO and H2O concentrations (for fuel-rich and fuel-lean conditions, respectively) to thermodynamic equilibrium calculations (TEC) and found to vary in a wide range (0.8-1.3). Soot production decreases with an increasing local equivalence ratio and oxygen enrichment and is lower for the swirl compared to the jet burner. The measured K(g) concentrations follow the general behavior predicted by TEC around stoichiometry. In the relevant temperature range (1100-1700 K), K(g) is 2-4 orders of magnitude higher under fuel-rich than fuel-lean conditions, with a sharp transition at stoichiometry. While K(g) concentrations are lower than TEC in the reactor core and under fuel-rich conditions, excellent agreement is found at the exhaust after complete fuel conversion. Precise, wide dynamic range detection of K(g) using TDLAS enables discrimination between fuel-rich and fuel-lean conditions and has the potential for lambda sensing close to the hot reaction zone of combustion plants.

Pyrolysis transforms bulky and heterogeneous lignocellulosic biomass into more easily-handled oils that can be upgraded into bio-based transportation fuels. Existing systems for monitoring pyrolysis processes and characterizing their products rely on slow and time-consuming wet chemical analyses. On-line near-infrared (NIR) spectroscopy could potentially replace such analyses, providing real-time data and reducing costs. To test the usefulness of NIR methods in characterizing pyrolysis oils and processes, biomass from conifers, Salix, and reed canary grass was milled and pyrolyzed at 675, 750, and 775 °C. Two separate pyrolytic fractions (aerosol and condensed) were produced in each experiment, and NIR spectra were collected for each fraction. Multivariate modelling of the resulting data clearly showed that the samples’ NIR spectra could be used to accurately predict important properties of the pyrolysis oils such as their energy values, main organic element (C, H and O) contents, and water content. The spectra also contained predictive information on the samples’ origins, fraction, and temperature treatment, demonstrating the potential of on-line NIR techniques for monitoring pyrolytic production processes and characterizing important properties of pyrolytic oils from lignocellulosic biomass.

A combination of image processing techniques and regression models was evaluated for predicting equivalence ratio and major species concentration (H2, CO, CO2 and CH4) based on real-time image data from the luminous reaction zone in conditions and reactors relevant to biomass gasification. Two simple image pre-processing routines were tested: reduction to statistical moments and pixel binning (subsampling). Image features obtained by using these two pre-processing methods were then used as inputs for two regression algorithms: Gaussian Process Regression and Artificial Neural Networks. The methods were evaluated by using a laboratory-scale flat-flame burner and a pilot-scale entrained flow biomass gasifier. For the flat-flame burner, the root mean square error (RMSE) were on the order of the uncertainty of the experimental measurements. For the gasifier, the RMSE was approximately three times higher than the experimental uncertainty – however, the main source of the error was the quantization of the training dataset. The accuracy of the predictions was found to be sufficient for process monitoring purposes. As a feature extraction step, reduction to statistical moments proved to be superior compared to pixel binning.