Biomass gasification usually takes place in the temperature range of 600 – 900 ֯C and in the presence of a gasifying agent including steam, air/oxygen, carbon dioxide or their combination. In the absence of an oxidizing agent, biomass particles undergo pyrolysis, which involves their decomposition into light gases, char, tar and other contaminant.
The thermal degradation of biomass in the atmosphere of nitrogen can be measured using the thermogravimetric analysis (TGA). During the pyrolysis, the fuel particles first go through the heating up and drying process when the temperature < 125 ֯C. Within 125 – 500 ֯C, an active pyrolysis takes place, where most of the volatiles are released.
Above 500 ֯C, the passive stage (secondary pyrolysis) sets in, leading to cracking of tar molecules into light gases and inert tar component. The composition and product yields from the biomass pyrolysis depend on the heating rate, biomass composition and the degree of nitrogen flux [33 - 36].
There are several studies on the biomass gasification in fluidized bed based on different heating methods, which include direct and indirect means. In an air/oxygen-blown gasifier, the required heat is generated internally due to partial oxidation of fuel species by the available oxygen, giving rise to an auto-thermal process. For the gasification with pure steam, an allothermal process is used where the required heat is supplied from an external source. The heat supplied for a steam-biomass gasification can be provided by a discontinuous intermittent operation of a single fluidized bed [37, 38], a circulation of particles between two interconnected fluidized bed [39 - 41] and an indirectly heated fluidized bed [42, 43]. The review of Karl and Pröll [17] provides a summary of the state of the art with respect to the layout and dimensioning of indirect heating processes for biomass gasification.
___
13
The gasification efficiency, gas composition, product yield and quality depend on a number of factors such as biomass type, amount of oxidizer relative to the biomass supply, gasification temperature and bed material used. In addition to Karl and Pröll [17], Kumar et al. [10] also presents a summary of effects of these parameters on the gasification behaviour as observed in different studies. The amount of gasifying agent influences the superficial gas velocity in the bed. The gas velocity in a bubbling bed gasifier is often within 5 – 10 times the minimum fluidization velocity [17]. Due to the increasing gas volume during gasification, which thus increases the superficial gas velocity, most designs introduce a gradual increase in the bed diameter along the reactor axis [17].
The composition of biomass influences the gasification performance. Hanaoka et al. [44]
showed that for cellulose, xylan and lignin based biomasses, the carbon conversion efficiencies at 900 ֯C are 97.7, 92.2 and 52.8%, respectively. The compositions of the product gas are similar in the last two-biomass types. While the CO and CH4 mole fractions are higher in the cellulose material, the amounts of CO2 and H2 are lower than in the xylan and lignin materials. The use of biomass with a high moisture content (>
10%) increases the energy requirement but reduces the amount of steam required in the gasifier [45]. The energy input for gasification also increases when using biomass with low carbon content due to the low char generation and high tar yield [46, 47].
Decreasing the biomass particle size increases the energy efficiencies and yield of CO [48 - 50]. In addition, Lv et al. [48] observed increasing amounts of CH4 and C2H4, and decreasing amounts of CO2 and H2 as the particle size is decreased. However, Rapagna and Latif [49] observed a decreasing trend in the yield of CO2 while Luo et al. [50]
observed an increasing trend for H2 yield. Decreasing the fuel particle size increases the specific surface area, which enhances the heat transfer, and thus the process efficiency.
The biomass flow rate also affects the gasification performance. Over feeding of biomass leads to plugging of the bed and a reduced conversion efficiency while under feeding results in lower gas yields. The optimum biomass flowrate depends on the gasifier design and the amount of the gasifying agent applied. For gasification with air, the equivalence ratio (ER) is used to relate the fuel supply with the amount of air applied.
The total gas yield and lower heating value increase with increasing value of ER, although different trends of the gas composition have been reported in different studies [51 - 54].
With an increase in the equivalence ratio within 0 - 0.45, the amounts of CO, H2, CH4 and tar decrease [51], H2 yield varies slightly until the optimum ER = 0.23 value [52], and CO and H2 yields increase [54]. For gasification with steam, a high steam flowrate decreases the cold gas efficiency and tar content of the product gas. A high steam flowrate also promotes char conversion and prevents the downstream soot and coke formations when the temperature is above 700 ֯C [17, 55]. Naraez et al. [51] showed that by
___
14
increasing H/C ratio (where hydrogen are derived from the moisture content, the external steam supply and the biomass composition) from 1.6 to 2.2, the hydrogen yield increases while the lower heating value increases from 4 to 6 MJ/Nm3 and tar content decreases from 18 to 2 g/Nm3. By varying the mass of steam to biomass ratio above 2.7, Lv et al. [52] observed that the gas composition does not vary significantly, but with an increase in the steam-biomass ratio from 1.35 to 2.7, the CO and CH4 yields decrease whereas the CO2 and H2 yields increase.
The bed material size and properties influence the reactor dimension and gas composition [17]. Due to slow gasification rate of char particles, high reactor volume is required to increase the residence time for effective conversions. The catalytic nature of some bed materials can also enhance the tar decomposition and CO shift for a higher H2 production [56, 57]. The most commonly used bed materials are olivine, silica sand and calcites due to their high specific heat capacity and ability to withstand high temperature [58].
Moreover, different models have been proposed for simulation of biomass gasification behaviour. As briefly highlighted in Article [A10], the models can be based on the thermodynamic equilibrium, reaction kinetics and a combination of the two. The procedures for modelling a gasifier are recently summarized by Mazaheri et al. [59]. For detailed analysis, models based on the computational fluid dynamics and computational fluid-particle dynamics are applied. Due to complexities of the multi-dimensional computational models, several one-dimensional models are available for prediction of the gas composition and studying of the effects of different operating parameters on the gasifier performance. Most of the existing one-dimensional models are based on the two-phase theory, which assumes that the gas flow through a fluidized bed exists in two separate phases (bubble and emulsion). In addition to the one-dimensional hydrodynamic model based on the computational fluid dynamics presented in the article [A2] for predicting the behaviour across different regimes in a non-reacting fluidized bed, a detailed 1D model based on the conservation of mass, momentum and energy for biomass gasification in a bubbling bed is also proposed in [A10]. As illustrated in the article [A10], the model can be used to study the effect of gasifier design choices and operating conditions.
___
15
3 Experimental Setups
There are two different experimental setups used in this study. The first is operated in the cold flow conditions to investigate the behaviour of fluidized beds at different gas velocities, particle sizes and bed heights. This setup is easy to control, and due to the cold operating environment, advanced measurement techniques such as ECT (electrical capacitance tomography) could be applied. The second setup is used to study the bed behaviour in hot flow conditions, and gasification of biomass under the atmospheric pressure condition. This chapter presents the detailed descriptions of the different setups and the methods employed in the measurement of the bed dynamic properties, which include the bubble dimeter, bubble velocity, bubble frequency, mixing and segregation pattern, and the biomass residence time over the conversion period.