The world’s current energy requirements are largely met by fossil fuels, such as oil, coal, and natural gas, which are estimated to account for 80% of the world’s energy consumption (Fernando, Adhikari, Chandrapal, & Murali, 2006; Xiao, Meng, Le, & Takarada, 2011). The increase in fuel costs, limited fuel sources, and environmental problems, such as global warming and acid rain, are all problems caused by the usage of fossil fuels. These crises have prompted mankind to look for renewable energy in order to meet the increasing energy demand. Among renewable energy resources, biomass is the only one that can produce not only heat and electricity, but also fuels.

Biomass-based energy accounted for roughly 10% of world’s total primary energy supply in 2009. Most of this biomass energy is consumed in developing countries for cooking and heating via very inefficient open fires or simple cook stoves with considerable impact on health (smoke pollution) and the environment (deforestation). Modern bioenergy supply, on the other hand, is comparably small, but has been growing steadily in the last decade. A total of 280 TWh of bioenergy electricity, or 1.5% of the world’s electricity generation, was produced globally in 2010, and 8 EJ of bioenergy for heat were used in the industrial sector (Xiao et al., 2011).

11.2.2 Origins of biomass resources

Biomass is plant material derived through photosynthesis, a set of reactions in through which CO₂ in the air, water, and sunlight produce the carbohydrates that form the building blocks of biomass.

The solar energy that drives photosynthesis is stored in the chemical bonds of the structural components of biomass (Peter, 2002), and biomass is the term used to describe all biologically produced matter. Biomass resources include wood and wood wastes, agricultural crops and their waste by-products, municipal solid waste (MSW), animal wastes, waste from food processing, and aquatic plants and algae. On average, the majority of biomass energy is produced from wood and wood waste (64%), followed by MSW (24%), agricultural waste (5%), and landfill gases (5%) (Demirbas, 2000).

11.2.3 Properties of biomass materials

Biomass can be used for fuels, power production, and products that would otherwise be made from fossil fuels, and it can provide an array of benefits. The use of biomass energy has the potential to greatly reduce greenhouse gas emissions. Biomass releases carbon dioxide that is largely balanced by the carbon dioxide captured in during its formation. Burning biomass produces 90% less sulfur than burning coal, and the use of biomass can reduce dependence on foreign oil because biofuels are the only renewable fuel that can be transported as a liquid (Demirbas, 2001). However, when compared to coal, biomass also has several shortcomings.

Table 11.1 shows the comparison of fixed carbon content, volatile matter, moisture content, heating value, and bulk density between coal and biomass. As can be seen, coal has a higher bulk density, higher heating value, and lower moisture content. Compared to pulverized coal fuels, biomass is generally more volatile, has more moisture, and has a lower heating value than coal. In general, biomass energy densities are approximately one-tenth of fossil fuels, such as petroleum or high-quality coal. In coal, the ratio of volatiles to fixed carbon content is low, always less than one. In biomass, however, this ratio is as high as four. This volatile-to-fixed-carbon ratio describes how easily a fuel is volatilized and affects the subsequent system and products. The high volatility is considered an advantage for biomass and allows the fuel to burn at a high power output (Demirbas, 2004).

The difference between the O/C and H/C ratios of solid fuels can be illustrated using a Van Krevelen diagram, as shown in Figure 11.1. The composition of the ash-free organic components of biomass is relatively uniform. The major components are carbon, oxygen, and hydrogen. Most biomass also contains a small proportion of nitrogen. A comparison of biomass with coal shows clearly that there are higher proportions of oxygen and hydrogen in biomass. Consequently, the higher oxygen and hydrogen content reduce the energy value of biomass as a fuel, due to the lower energy contained in carbon–oxygen and carbon–hydrogen bonds, than in carbon–carbon bonds.

Research shows that the mineral composition of coal and biomass has a strong impact on processing, application, and environmental and technological concerns related to these fuels. For biomass, variability in mineral content among plants can be considerable, as it depends on genetic and environmental factors, as well as physiological and morphological differences between crops. Table 11.2 presents the ash compositions of typical biomass and coal samples. It clearly shows that the ash compositions of biomass and coal are different. Biomass ash is mainly composed of K, Na, Mg, Al, Ca, and P, in the form of oxides, silicates, and chlorides, while coal ash consists mainly of Al and Si in the form of oxides. The ash from biomass also contains high alkali metal content, which can lead to corrosion in the gasifier instruments and downstream setup.

11.4.2 Gasification parameters

There is a series of parameters that are crucial for the efficiency of gasification, and thus the optimum values of these factors should be maintained to ensure a constant quality with high process performance.

11.4.2.1 Gasification temperature

As it controls the cracking and conversion of biomass, temperature plays a vital role in gasification. Figure 11.6 provides the gasification property under different temperatures during sawdust gasification (Chen, Li, Yang, Yang, & Zhang, 2008). A higher operating temperature (> 800 °C) is always favorable for higher hydrogen and lower tar content yield (decrease from 13.2 to 6.5 g/m³) in the product gas. Meanwhile, temperature not only influences the amount of tar, but also the tar composition by changing the chemical reactions during gasification (Devi, Ptasinski, & Janssen, 2003; Meng, De Jong, Fu, & Verkooijen, 2011; Mayerhofer et al., 2012). The effects of temperature on tar are shown in detail in the section on tar. With the rise of temperature, the carbon conversion rate increases, as does the thermal efficiency, but the rising temperature may cause more severe fouling and slagging problems when the biomass has higher contents of K, Cl, and other inorganic materials. Therefore, the optimal temperature is based on the conversion and application condition.

11.4.2.3 Gasification pressure

Pressure also influences gasification behavior. As shown in Figure 11.7, Mayerhofer investigated the effects of pressure on gas composition and tar content (Mayerhofer et al., 2012). The enhancement of WGS reactions under pressurized conditions makes the gas composition shift to higher CH₄ and CO₂ content, while CO decreases. Figure 11.7b indicates that an increase in total tar content was observed when pressure increased from 0.1 to 0.25 MPa. But an opposite trend in tar content has also been reported (Wolfesberger, Aigner, & Hofbauer, 2009). Furthermore, the syngas produced at high pressures is favorable for downstream high-pressure units, such as turbines and FT synthesis. However, high-pressure gasification seems uneconomical when extra equipment is needed to ensure the stability of the over-pressurized gasification conditions. Other operational conditions, such as the bed material in fluidized beds (Meng et al., 2011) and the biomass feeding rate (Lv et al., 2007), influence gas distribution and tar formation as well.

11.6.1 Formation mechanism of tar

The characteristics of tar mainly depend on the composition of the tar, particularly the tar’s heavy compound content. The components of tar are very complex, and more than 200 kinds can be detected in a single sample. This diversity of components is due to the fact that tar is formed from volatiles during biomass pyrolysis, and the composition of volatiles is dependent on temperature. The main composition of liquid tar obtained from variable temperatures is shown in Figure 11.9. When the temperature is below 550 °C, the volatiles are formed from the direct degradation of cellulose and hemicellulose. Therefore, most of the resulting tar is formed by low-molecular-weight and oxygen-containing compounds, such as acids, esters, ketones, furans, cyclopentene, guaiacols, and phenols. This type of tar easily undergoes further reforming. With the temperature increasing up to 650 °C, complex phenols replace these low-molecular-weight and oxygen-containing compounds (Hernández, Ballesteros, & Aranda, 2013; Zheng, Zhu, Guo, & Zhu, 2006). However, during the secondary reaction of the volatiles, the elimination reactions of the oxygen-containing functional groups produce some aromatic compounds containing benzenes, naphthalenes, biphenyls, benzofurans, and benzaldehyde. Above 650 °C, the molecular weight and the number of aromatic rings within the tar components increase significantly because of the generation of a large number of PAHs. Although complex phenols are observed in the tar obtained at 750 °C, the weight content of complex phenols is rather lower, and the branched structures become uncomplicated. As the temperature rises to 950 °C, the dehydrogenation condensation reaction generates some PAHs with more than three aromatic rings, such as acenaphthylenes, benzopyrenes, fluoranthen, and pyrene, which are the precursors of the particulate matter called “soot” (Chen, Yang, Wang, Zhang, & Chen, 2012; Qin, Feng, & Li, 2010), The thermal stability of such tar is very high, so it is difficult to crack and remove.

11.6.2 Tar cracking

Tar removal is seen as one of the greatest technical challenges to overcome for the successful development of advanced, commercially viable gasification technologies (Minlne & Evans, 1998; Li & Suzuki, 2009). Tar cracking technology can be divided into two basic methods: thermal cracking and catalytic cracking. However, the former method is not considered to be a feasible option, as it requires temperatures higher than 1100 °C to achieve high cleaning efficiency, and it also produces soot (Aznar, Corella, Delgado, & Lahoz, 1993).

In comparison with high temperature thermal cracking, catalytic cracking is highly efficient. Catalysts used in biomass conversion can be divided into three distinct groups. They are dolomite catalysts, alkali metal and alkali earth metal catalysts, and nickel-based catalysts.

Low carbon (char) conversion and high tar content in the gas are the two main problems that have blocked the development of biomass gasification (Gómez-Barea et al., 2013). Therefore, tar cracking and char gasification are two critical issues for syngas quality upgrading. Higher temperature is favorable for tar removal and char gasification (Hasler & Nussbaumer, 1999; Sutton, Kelleher, & Ross, 2001). But a new challenge is arising, as biomass facilities may be at risk of agglomeration at high temperatures, due to the release of a high proportion of alkali metals from certain feedstocks such as agricultural straws (Nilsson, Gomez-Barea, Fuentes Cano, & Ollero, et al., 2012). Staged gasification is a suitable way to reach high char conversion, while yielding a gas with a low concentration of heavy tar (Nilsson et al., 2012; Gómez-Barea, Ollero, et al., 2013). Gas produced by stage gasification is ideal for direct thermal applications, including gas engines, boilers, and fuel cells.

Staged gasification creates at least two different temperature zones and various thermal levels in the gasification bed by staging the oxidant. The biomass gets devolatilized at relatively low temperatures and then gasified at elevated temperatures with the remaining oxidant. As is shown in Figure 11.11, the process is divided into two parts: the pyrolysis stage at temperatures of 350 ~ 600 °C and the gasification stage at temperatures of 800-1000 °C. This process is convenient for the optimization of simultaneous char conversion and tar cracking.

Early in the 1994, Bui, Loof, and Bhattacharya (1994) found that tar yield in multi-stage reactors for thermal gasification was a factor of 40 times lower than the tar yield produced by one-stage gasification. Henriksen, et al. (Brandt, Larsen, & Henriksen, 2000; Henriksen et al., 2006) designed a 75-kw two-stage fixed-bed gasifier that has been operated for more than 2000 h, and this gasifier has produced product with a tar content < 15 mg/m³. Šulc et al. (2012) found that two-stage gasification systems can significantly decrease aromatic compounds with two or more benzene rings. Stage gasification of biomass not only sharply decreases the tar yield, but it also can increase the LHV of the resulting syngas. Hamel, Hasselbach, Weil, and Krumm (2007) produced high calorific value gas in a bubbling fluidized bed-fixed bed stage gasification system, and when using this process, the LHV of gas from MSW steam gasification increased to 14 MJ/Nm³. As for the release of AAEM and the reaction rate of gasification at about 800 °C lower than traditional conditions, Sharma, Saito, and Takanohashi (2008) studied the effects of K₂CO₃ catalytic gasification characteristics on ash-washed coal, and they pointed out that the low temperature of coal gasification (< 650 °C) is feasible. Sueyasu et al. (2012) proposed a two-stage conversion of biomass into gas, during which pyrolysis occurs at 500-600 °C and steam reforming/gasification occurs at 600-700 °C, with the K₂CO₃ catalyst being recycled. The addition of K can obviously increase the gasification characteristics of pine sawdust at low temperature. As a reslt of adding potassium, the concentration of hydrogen in the product gas exceeded 50 vol.%, and the tar yield was as low as 20 mg/Nm³.

Furthermore, the produced char can be gasified in the second stage and as the heat-carrier or catalyst for the further conversion of tar. Gómez-Barea (Gómez-Barea, Leckner, et al., 2013; Gómez-Barea, Ollero, et al., 2013) has proposed the three-stage concept shown in Figure 11.12, which includes fluidized bed devolatilization (first stage), the non-catalytic air/steam reforming of the gas from the devolatilizer (second stage), and the chemical filtering of gas in a moving bed supplied with the char generated in the devolatilizer (third stage). Air and steam can be injected at various points, such as the devolatilizer, steam reformer, and seal, with different proportions of the two reactants. The fuel is fed near the bed’s surface and has to circulate down to the bottom before leaving the bed. The devolatilizer, where a high yield of fresh tar is generated, is operated at relatively low temperatures (700-750 °C). The fresh tar compounds are drastically reduced in the reformer, where a temperature of up to 1200 °C is created. The injection of steam into the reformer avoids coking and polymerization of the tar. The gas is then filtered through a moving bed made of char which comes from the loop seal. The loop seal can be operated as an oxidizer fed with enriched air or as a light reformer fed with H₂O, depending on the fuel’s reactivity and ash properties. The char filter also cools down the gas through an endothermic char gasification reaction with steam, while the char also acts as a catalytic filter promoting tar decomposition reactions with steam.

In the new three-stage system, the char conversion is up to 98%, leading to higher process efficiency. By optimizing the operating condition of air and steam in the three-stage system, the cold gas efficiency and gas HHV can increase to 0.81 and 6.9 MJ/Nm³ dry gas respectively. When using 40 vol.% oxygen-enriched air instead of typical air, the cold gas efficiency increases to almost 0.85, and the resulting gas can have an HHV of 10.8 MJ/Nm³dry gas. Given that the heavy tar content is lower than 0.01 g/Nm³, being virtually converted in the system, the obtained low dew point gas can be burned in a gas engine, and as a result, this gas is ideal for power production.

Nowadays, staged gasification tends to involve large-scale applications and continuous operation. The University of Canterbury has built a 100 kW fast internal circulating fluidized bed gasification system that incorporates two closely coupled fluidized bed stages, a bubbling bed for gasification and a fast circulating bed for combustion (Brown, Dobbs, Devenish, & Gilmour, 2006). This duality provides a medium calorific value producer gas suitable for use as a fuel in a gas engine or gas turbine. Steam gasification is used in the bubbling fluid bed, at ~ 800 °C, to form a product gas that is rich in hydrogen. Residual char is transferred with bed material to the circulating fluid bed, where it is combusted, along with LPG, to heat the bed material. The hot bed material is then circulated back to the gasification stage, providing heat for the endothermic gasification reactions. On the other hand, char, char-supported catalysts, and ilmenite are investigated for the steam reforming of biomass tar, as opposed to the use of a precious metal catalyst. Min et al. (Min, Asadullah, et al., 2011; Min, Yimsiri, Asadullah, Zhang, & Li, 2011) clearly indicated that the chars from the pyrolysis and gasification of biomass could support a new class of cheap industrial catalysts with superior performance. Char would not only disperse the catalysts, but it would also interact with the catalysts to enhance their involvement in the steam reforming of tar. The physical and chemical property of support could play important roles for the activities of the catalysts and the reaction pathways of the catalysts. The char-supported iron/nickel catalysts exhibited much higher activity for the reforming of tar than char itself. So the utilization of pyrolysis char as catalyst support is another research point.

11.8.2 Sorption-enhanced steam gasification of biomass for H₂ production

Together with electricity, hydrogen is considered to be one of the two main terminal energies produced by gasification in the twenty-first century. Hydrogen can also be utilized for ammonia production and, with suitable CO concentrations, for methanol and FT synthesis. Through the use of a calcium oxide (CaO) sorbent for carbon dioxide (CO₂) capture, the steam gasification of biomass offers a potential means for the renewable and sustainable hydrogen (H₂) production.

Sorption-enhanced steam gasification of biomass is a novel one-step conversion technology developed for high-concentration H₂ production. In this process, CO₂ sorbents are introduced into the process of biomass steam gasification for the continuously in situ removal of CO₂ as soon as it was formed in the gasification process. As a result, the chemical equilibrium of the gasification reactions changes, and the process produces more H₂ (Hanaoka et al., 2005; Harrison, 2009). The gasification unit, the WGS unit, and the CO₂ separation unit are integrated into a single-stage reactor in this process, in order to achieve in situ CO₂ removal, energy integration of the endothermic gasification and exothermic WGS and CO₂ sorption processes, facility and process simplification, and steam usage reduction. As a result, the overall efficiency and economic feasibility of biomass gasification are improved.

Considering the spent CO₂ sorbent regeneration, the whole H₂ production process is a looping system, and the circulation between sorption and desorption ceaselessly transforms each cycle. With CaO as an example, the principle of this process could be described as shown in Figure 11.13 (Koppatz et al., 2009), and the main reactions that take place in the process could be summarized as follows:

$\mathrm{Biomass}\mathrm{steam}\mathrm{gasification}:{\mathrm{C}}_{\mathit{x}}{\mathrm{H}}_{\mathit{y}}{\mathrm{O}}_{\mathit{z}}\left(\mathrm{Biomass}\right)+\left(1-\mathit{y}\right){\mathrm{H}}_2\mathrm{O}\to \mathrm{CO}+\left(0.5\mathit{x}+1-\mathit{y}\right){\mathrm{H}}_2$

$\mathrm{WGS}\mathrm{reaction}:\mathrm{CO}+{\mathrm{H}}_2\mathrm{O}\to {\mathrm{CO}}_2+{\mathrm{H}}_2,\mathrm{\Delta H}=-41.2\mathrm{kJ}/\mathrm{mol}$

${\mathrm{CO}}_2\mathrm{absorption}:\mathrm{CaO}+{\mathrm{CO}}_2\to {\mathrm{CaCO}}_3,\mathrm{\Delta H}=-178.3\mathrm{kJ}/\mathrm{mol}$

Li et al. (2011) have reported that a H₂ concentration of 94.92% could be obtained at the gasification temperature of 600-700 °C, with the atmospheric pressure being appropriate for chemical equilibrium. Marquard-Möllenstedt et al. (2004) also reported that the H₂ concentration could increase from 40% to about 75%, with the addition of CaO, while the CO₂ concentration decreased significantly.

CaO-enhanced steam gasification of biomass still has several advantages: (1) the temperature of the CaO carbonation reaction coincides with the temperatures required for biomass gasification; (2) CaO also has a catalytic effect on the process of biomass gasification and tar cracking and reforming. Considering the appropriate conditions of the CaO carbonation reaction, the optimized carbonation temperature is between 600 and 700 °C at atmospheric pressure. In this temperature range, biomass steam gasification enhanced by CaO sorption could produce a high H₂ concentration.

However, due to the lower temperature (600-700 °C) of this process compared to the conventional gasification process (800-900 °C), and the reaction activity of H₂O is relatively low compared to air, and although the H₂ concentration is very high, the conversion rate of biomass and total carbon is still low. This low conversation rate is due to the fact that a portion of the carbon remains in the solid char, and other parts go into the liquid tars, thus resulting in the lower H₂ yield, as shown in Figure 11.14. To improve the biomass/carbon conversion rate and the H₂ yield, two methods could be used:

(1) Increase the gasification temperature to enhance the reaction rate and accelerate the gasification process, thereby improving the biomass conversion rate and the H₂ yield. Considering that the temperature necessary for CO₂ absorption would enter the high range under pressurized conditions, a feasible solution might be high-temperature pressurized gasification, such as the HyPr-RING process (Lin, Harada, Suzuki, & Hatano 2002; Lin, Suzuki, Hatano, & Harada, 2001) or pressurized fluidized bed gasification (Han et al., 2011).

(2) Introduce the catalysts into the gasification process to enhance the gasification reaction rate, as well as the selectivity and conversion rate of the reforming reaction of tars and carbonaceous gases. Introducing catalysts would realize the high conversion rate of biomass and the high H₂ selectivity in the product gas at relatively low temperatures. Such catalysts include Fe/CaO (calcined dolomite) and Ni/CaO (calcined dolomite) catalysts (Di Felice et al., 2009; Di Felice, Courson, Foscolo, & Kiennemann, 2011), Ni-Mg-Al-CaO catalysts (Nahil et al., 2013), and Pd-Co-Ni + dolomite catalysts (Fermoso, Rubiera, & Chen, 2012), etc. For example, Nahil et al. (2013) found that over 80% H₂ was produced with a Ni-Mg-Al-CaO bed during biomass gasification.

11.9.1 Thermodynamic equilibrium models

The “equilibrium model” refers to thermodynamic equilibrium models, such as models of chemical reaction equilibrium, which are based on the minimization of the Gibbs free energy of the system. Gibbs free energy is minimized when the species of a reaction system reach equilibrium and will no longer experience any change over time.

Mansaray, Ghaly, Al-Tawell, Ugursal, and Hamdullahpur (2000) simulate rice husk gasification based on material balance, energy balance, and chemical equilibrium relations. Their model attempted to predict the core, annulus, and exit temperatures; the mole fractions of the combustible components of the gas; the higher heating value of the gas; and the overall carbon conversion under various operating conditions, including bed height, fluidization velocity, equivalence ratio, and the moisture content of rice husk.

However, the definition of chemical equilibrium implies that the residence time is long enough to allow the chemical reactions to reach stasis, which is rarely reached in real gasifier. Hence, in some research, the models employed the restricted equilibrium method (Doherty, Reynolds, & Kennedy, 2009), and the temperature approach for the gasification reactions was specified. Li et al. (2004) introduced a phenomenological model that incorporated experimental results regarding unconverted carbon and methane to account for non-equilibrium factors, and this model predicted product gas compositions, heating value, and cold gas efficiency in good agreement with the experimental data.

Thermodynamic equilibrium models, which only take into account thermodynamic limitations, inherently disregard specific reaction mechanisms that are independent of the gasifier design (Jand, Brandani, & Foscolo, 2006). Also, the results may not be achieved at low temperatures, so the calculations may not be representative of the real situation, when kinetic constraints become the major factor (Ju et al., 2009). Hence, thermodynamic equilibrium models cannot produce accurate results, but these models are efficient for the process of optimizing and prediction (Ju et al., 2009; Mahishi & Goswami, 2007).

11.9.2 Kinetics models

Kinetic models differ from thermodynamic equilibrium models, as they describe the char reduction process using kinetic rate expressions obtained from experiments, thereby permitting better simulation of the experimental data when the residence time of gas and biomass is relatively short.

Kaushal, Abedi, and Mahinpey (2010) developed a one-dimensional steady state model with two phases (bubble and emulsion) and two zones (bottom dense bed and upper freeboard). This model was based on global reaction, kinetic, mass, and energy balances of biomass gasification in bubbling fluidized gasifiers, and it is capable of predicting temperature, solid hold up, and gas concentration along the reactor’s major axis.

Wu, Zhang, Yang, and Blasiak (2013) built a two-dimensional computational fluid dynamics (CFD) model to study the gasification process in a downdraft configuration, considering drying, pyrolysis, combustion, and gasification reactions. The gas and solid phases were resolved using an Euler-Euler multiphase approach, with exchange terms for the momentum, mass, and energy.

Miao et al. (2013) developed a new mathematical model that combined hydrodynamics with chemical reaction kinetics to predict the overall performance of a biomass gasification process in fluidized beds. The fluidized bed gasifier was divided into two distinct sections: a dense region at the bottom and a dilute region at the top. Each section was divided into a number of small cells, over which mass and energy balances were applied. The model is capable of predicting the bed temperature distribution along the gasifier, the concentration and distribution of each species in the vertical direction of the bed, the composition and heating value of produced gas, the gasification efficiency, the overall carbon conversion, and the produced gas production rate well.

Kinetic models provide essential information on kinetic mechanisms to describe the conversion during biomass gasification, which is crucial in designing, evaluating, and improving gasifiers. These kinetic models are accurate and detailed, but they are also computationally intensive. In addition, the basic experiment data is necessary when constructing and running the models. Sometimes, thermodynamic equilibrium models have been combined with kinetic models. Lee, Yang, Yan, and Liang (2007) extracted substitutable gas phase compositions from thermodynamic calculations, before entering the gas phase compositions into the Sandia PSR code to consider the potential kinetic constraints involved in the pyrolysis. Lee and colleagues also hoped to obtain the distributions of gas products. The result, in this case, is much closer to the realistic situation (Lee et al., 2007).

11.9.3 Neural networks model

Artificial neural networks are normally based on mathematical regression to correlate input and output streams to and from process units. Such models principally rely on a large number of experimental data. Guo, Li, Cheng, Lu, and Shen (2001) created a hybrid neural network model for the purpose of predicting biomass gasification profiles at atmospheric pressure with steam. Artificial neural networks differ from traditional regression because they have more potential for finding the unseen structure (Guo et al., 2001).

Each type of model has its own strengths and limitations, as mentioned above. In order to achieve efficient calculation and design, Brown, Fuchino, and Mar Chal (2006) combined the three models together, given that fuels and chars are defined as pseudospecies with properties derived from their ultimate analyses, and tars are defined as a subset of known molecular species with their distribution determined by equilibrium calculations. While the reforming of gas, tar, and char formation was explained by applying reaction temperature differences to a complete set of stoichiometric equations, the changes in temperature related to fuel composition and operational variables were determined with nonlinear regression via an artificial neural network. This method improves the accuracy of equilibrium calculations and reduces the amount of required data by preventing the neural network from learning atomic and heat balances.