Computational Study on Parametric Variation with Solar Heat Induction of an Entrained Flow Gasifier

Abstract

Gasification has played an important role in the sustainable use of waste biomass, providing useful combustible gases in the process. Gasification has an important role in waste management and promotes energy independence for many oil-deficit countries. The gasification process has been studied by various researchers, and improvements have been achieved in its sub-processes such as devolatilization, feed input methods, and so on. We examined the influence of gasifier operation parameters, such as oxidizer content, moisture content in the feedstock, and solar flux input inside the gasifier, on the temperature distribution, velocity distribution, and product gas yields of the gasifier. The results indicate that inducing solar energy at different stages of the gasifier leads to different yields of product gas composition (CO and H₂).

1. Introduction

Biomass waste is one of the problems in today’s waste product management. Yearly biomass waste in India reaches up to 500 million metric tonnes per year, as reported by MNRE, India. The thermochemical breakdown of these biomass wastes and extracting useful product gases can be a viable solution. Energy problems in remote areas can be resolved by this method. Gasification has proved to be an acceptable solution for biomass energy recovery. It provides a secondary solution for the growing energy demand all around the world. This process is carried out in a chemical reactor called a gasifier, in which there are four primary zones based on different temperature limits. The input biomass undergoes drying, devolatilization, and reduction before it is converted into useful combustible gases. These combustible gases can be used for further applications such as cooking, furnace fuel, or even electricity production. Gasification is dependent on various crucial factors such as the composition and supply rate of input compounds, temperature variations in the reactor body, etc. Power production in gasification is executed with the help of integrated cycle plants in which the gasifier plays an important role as the source of fuel gases. The study of gasification inside the reactor body and the effects of various input parameters are of foremost importance. In our research, we have tried to study those effects with the help of computational fluid dynamics. Solar gasification is a branch of gasification that further increases the sustainable quotient in the gasification process. It provides a gateway for higher calorific value fuels with less investment in the biomass feedstock of gasifier. Various experimental observations have been made by researchers regarding these factors. Pinto et al. (2003) [1] studied the effect of temperature and different oxidizer composition on the product gases. An increase in temperature reduced the tar content and promoted hydrogen production. An increase in steam content in the oxidizer also promoted the reduction reactions and further increased the hydrogen content. Walawender et al. (1988) [2] conducted experiments regarding varying biomass species and quality with the same operating conditions. Different biomass species had a minute effect on the dry gas heating value of the product gas, but when deteriorated wood was supplied, a reduction in energy output was noticed. In the current investigation, we focused on an industrial-based gasifier that has been used for electricity generation for a long time. The project “Wabash River Coal Gasification Repowering Project” (Foster (1994) [3]) has been computationally studied by various authors for the constant development of the physical models involved in the gasification process. Shi et al. (2006) [4] used a Lagrangian approach to model the coal slurry flow. Slezak et al. (2010) [5] also investigated the same geometry and studied the effects of varying coal particle size and density. Kaundal et al. [6] numerically investigated the effect of air supply in domestic cookstoves with conventional fuel and produced experimental results. They also gave a detailed review of various alternative fuels based on cookstoves and gasification in another study (Kaundal et al. [7]). Ma et al. (2012) [8] studied this project in ANSYS Fluent software and developed a user-defined devolatilization method. Chyou et al. (2013) [9] investigated the effects of changing the second stage inlet in the same gasifier geometry and concluded that the tangential entry of feed in the second stage of gasifier improved the temperature distribution and the height of the gasifier can be shortened. In this study, we used ANSYS Fluent software to study the effects of changing input parameters on the product gas yield and composition. Studying these effects computationally can help us to determine the optimal inlet conditions for such industrially used gasifiers and many valuable resources can be saved. Solar-assisted gasification is a part of gasification in which a portion of required heat energy is employed from solar radiation. In our study, we excluded the use of radiation, but the effect of heat at certain areas of the present gasifier was examined and an effect on product gas yields was observed. Gregg et al. (1980) [10] investigated a packed bed solar reactor and reported an improved heating value in the product gas. Taylor et al. (1983) [11] reported that solar energy was stored in the fuel gas in the form of chemical energy while investigating fluidized bed and packed bed reactors combined with solar energy inputs. Bai et al. [12] used cotton stalk as feed in a solar-driven gasifier and obtained increased efficiency of the whole ethanol production system and improved the subsequent cost effectiveness. In our previous investigation, we discussed in detail various processes and design trends in solar-assisted gasification (Singh et al. (2019) [13]).

Gregg et al. (1980) [10] and Qader et al. (2001) [14] both showed that product gas yields increased to almost double when the process heat was supplied from the solar power. Zedtwitz et al. (2003) [15] reported that product gas obtained at the end of gasification is free of any unwanted compounds and is much cleaner than the conventional process yields. Wang et al. [16] reported a reduction in carbon emissions when solar energy was used in gasification. Solar energy reduces the need for feed (input) required for the combustion process and thus saves the biomass feed that will be converted to useful fuel gas. Abanades et al. [17] observed significant increases in yields of hydrogen and methane after using solar energy in their experimental investigations.

We learned from a review of the literature that many computational studies were conducted to discover and test new workings of this gasifier geometry [8], but its parametric behavior and response to solar energy induction were not reported to be investigated. As a result, we decided to use this study as our primary reference and test calculated variations in various factors involved in gasification. As the demand and supply of fossil fuels remain skewed, developing countries must invest more in alternative fuel technologies. Similar future studies to ours will be useful in establishing large-scale plants for such purposes. The step-by-step approach of our study is shown in Figure 1.

2. Computational Modeling Approach

Mass, momentum, and energy conservation equations are used to model the thermal and fluid phenomenon in the gasifier. The partial differential equations for the gas phase are discretized over a structured 3D mesh. We used ANSYS Fluent 2018 to implement our models. Most of the model equations mentioned in the paper are available in ANSYS software [18] The continuity equation represents the conservation of mass, as given in Equation (1).

$$\frac{\partial \rho }{\partial t}+ \nabla .\left(\rho \overrightarrow{u}\right)=S$$

(1)

where ρ is the density of the fluid, $\overrightarrow{u}$ represents velocity vector, and $S$ is a source term for any user-defined phase or fluid.

The momentum equation is modeled using the Reynolds Averaged Navier Stokes (RANS) model, which averages the turbulence fluctuations and gives a physically acceptable solution. The general equation of momentum is described in Equation (2).

$$\frac{\partial \left(\rho \overrightarrow{u}\right)}{\partial t}+\nabla .\left(\rho \overrightarrow{u}\overrightarrow{u}\right)=-\nabla p+ \nabla .\stackrel{=}{\tau } +\rho \overrightarrow{g}+\overrightarrow{f}$$

(2)

where p is static pressure, $\stackrel{=}{\tau }$ is the stress tensor in the fluid flow, $\rho \overrightarrow{g}$ is the gravitational effect, and $\overrightarrow{f}$ is the external body force. The stress tensor for a particular direction, e.g., x, is defined in Equation (3).

$$\stackrel{=}{\tau }{ }_{xx}=\lambda \left(\nabla .\overrightarrow{u}\right)+ 2\mu \frac{\partial \overrightarrow{u}}{\partial t}$$

(3)

where $\mu$ is the molecular viscosity coefficient and $\lambda$ = $-2\mu /3$. To accommodate the turbulence in the domain, the momentum equations are edited, and two terms are induced to the momentum equation, turbulent kinetic energy (k) and turbulent dissipation rate ($ϵ$). This further adds two additional transport equations for (k) and ($ϵ$), and the model used here is called the (k-$ϵ$) turbulent model. In this study, the inlet mixture enters from the sides and tends to mix at the center of the cylindrical body of the first stage of the gasifier. Thus, we needed a high Reynolds number formulation model, such as the standard k-$ϵ$ model. The damping functions in this model provide enough accuracy to account for the near-to-wall turbulent energy dissipation. Additionally, due to the absence of adverse pressure gradients in our case, the probability of accurate results from this model increases (Wilcox 1993 [19]). This model is widely used in industrial applications because it is a RANS model and considers the averaged values which are required for calculations and gives a trend of results based on experimental inputs. It is also a widely validated model in comparison to other two-equation turbulent models (Z. Zhai et al. (2007) [20]). The respective Equations are (4)–(6).

$$\frac{\partial \left(\rho k\right)}{\partial t}+\nabla .\left(\mathsf{\rho }\overrightarrow{u}\mathrm{k}\right)=\nabla .\left(\left(\mu +\frac{{\mu }_t}{{\sigma }_k}\right)\nabla \mathrm{k}\right) +P_k+P_b-\rho ϵ+S_k$$

(4)

The transient turbulence energy term and convection term are there on the left-hand side of the equation. The diffusion term is also included on the right-hand side with the source and sinks terms. ${\mu }_t$ represents the turbulent or eddy viscosity. $P_k$ and $P_b$ are the turbulent kinetic energy generation to capture the effects of velocity gradients and buoyancy. As $ϵ$stands for viscous dissipation, $\rho ϵ$ can be considered as a sink term. ${\sigma }_k$ is the turbulent Prandtl number for k.$S_k$ is the source term.

$$\frac{\partial \left(\rho ϵ\right)}{\partial t}+ \nabla .\left(\mathsf{\rho }\overrightarrow{u}ϵ\right) =\nabla .\left(\left(\mu +\frac{{\mu }_t}{{\sigma }_{ϵ}}\right)\nabla ϵ\right) +C_1\frac{ϵ}{k}\left(P_k+C_3{P}_b\right)-C_2\rho \frac{{ϵ}^2}{k}+S_{ϵ}$$

(5)

Viscous dissipation is described in Equation (5). We can observe the analogy between the above two equations, and the major difference is between the source and sink terms. ${\sigma }_k$ and ${\sigma }_{ϵ}$ are turbulent Prandtl numbers for k and $ϵ$, respectively. C₁, C₂, and C₃ are constants. $S_{ϵ}$is the source term. The energy equation models the conservation of energy in the computational domain and is responsible for defining the temperature distribution.

$$\frac{\partial \left(\rho E\right)}{\partial t}+ \nabla .\left(\overrightarrow{u}\left(\mathsf{\rho }\mathrm{E}+\mathrm{p}\right)\right)=\nabla .\left(k_{eff} \Delta T-{\displaystyle \sum }_j{h}_j{\overrightarrow{J}}_j+\stackrel{=}{\tau }.\overrightarrow{u} \right)+S_h$$

(6)

Physical phenomena of conduction diffusion and viscous dissipation are given in the first term on the right-hand side of Equation (6). $k_{eff}$ stands for effective thermal conductivity and includes the effect of turbulence on conductivity, as well. $\sum _j{h}_j{\overrightarrow{J}}_j$ stands for enthalpy diffusion for a particular species present in the domain. $\stackrel{=}{\tau }.\overrightarrow{u}$ stands for the viscous dissipation of heat.$S_h$ is the source term. The discrete ordinated (DO) radiation model is used to model the effect of radiation on the flow domain. This model uses a transport equation for radiation intensity in three-dimensional space coordinates [18].

2.1. Coal and Moisture Injection Models

The input feed, coal, is induced in the reactor body in the form of particles in the computational domain by using the discrete phase model (DPM). A Lagrangian approach is used in this model by tracking the particle trajectories. A force balance equation is the basis of DPM, as given in Equation (7). The coal particles undergo inert heating, vaporization, heat transfer, and devolatilization, and each phenomenon has some specific mathematical model. All these physical phenomena are demarcated by the temperature reading of the particle.

$$\frac{d{u}_p}{dp}=F_d\left(u-u_p\right)+ g_x \frac{\rho -{\rho }_p}{{\rho }_p}+F_x$$

(7)

In Equation (7), $F_d\left(u - u_p\right)$represents the drag force relating the particle, u is the phase velocity, and $u_p$ is the particle velocity. $\rho$ and ${\rho }_p$are fluid phase and particle densities, respectively. The term $F_x$represents all the additional forces that can be of significant magnitude in certain circumstances. Rotational forces, thermophoretic forces, forces due to the Magnus effect or Brownian effect, etc., are some examples included in these terms.

This study includes coal and water vapors in the form of discrete particles, and various mathematical models must be used to simulate various physical effects. Inert heating, vaporization, heat transfer, and devolatilization are the significant models which decide the course of this study. Inert heating is applied for the heating of the particle until the specified temperature given by the user is achieved. After reaching the vaporization temperature, the equation for vaporization is used for single discrete phases. This law is applicable until the boiling point is achieved. During the vaporization temperature, the mass transfer takes place in the form of diffusion, and Equation (8) describes the phenomenon.

$$N{ }_i=k_c\left(C_{i,s}-C_{i,\infty }\right)$$

(8)

where N_i represents the molar flux of vapor, k_c is the mass transfer coefficient, and $C_{i,s}$ and $C_{i,\infty }$are vapor concentrations at droplet surface and bulk gas, respectively. Similarly, the temperature of the discrete particle is updated by taking into account the sensible, convective, and latent heat transfers, which are included in the heat balance Equations (9) and (10).

$$m_p{c}_p\frac{d{T}_p}{dt}=h{A}_d\left(T_{\infty }-T_p\right)+ \frac{d{m}_p}{dt}{h}_{fg} +A_p{ϵ}_p\sigma \left({\theta }_R^4-T_p^4\right)$$

(9)

where $T_p$ and $T_{\infty }$are the particle and droplet temperatures, respectively; $\theta$is the radiation temperature; $c_p$ is the discrete particle heat capacity; h is the convective heat transfer coefficient; $h_{fg}$ is the latent heat; and ${ϵ}_p$and $\sigma$ are the particle emissivity and Stefan Boltzmann constant, respectively. The right-hand side of Equation (9) represents convection, evaporation, and radiation terms, consecutively. After vaporization, the boiling phenomena is modeled using the boiling equation, which includes the radiation and convection terms as given in Equation (10).

$$-\frac{d(d_p)}{dt}=2/{\rho }_p{h}_{fg}\frac{2k_{\infty }\left(1+0.23\sqrt{R{e}_d}\right)}{d_p}\left(T_{\infty }-T_p\right)+ \frac{d{m}_p}{dt}{h}_{fg} +ϵ\sigma \left({\theta }_R^4-T_p^4\right)$$

(10)

$d_p$ is the particle diameter. The boiling model is used in the case of a liquid droplet for modeling water vaporization. In the case of a combustive particle, we cannot use the boiling model; therefore, after inert heating, devolatilization is modeled using the model given by Kobayashi et al. [11]. Coal is used as a combustible particle. This model is used when the temperature of the particle reaches the vaporization temperature ($T_{vap}$). Boiling point and vaporization temperatures are the same for a combustible particle. This model states two rates that control the de-volatilization over different temperature ranges. The need for the two rates arises because the reaction rates are different for higher and lower temperatures. The reaction rates are given in Equation (11).

$$\begin{gathered} r_1=B_{1}exp\left(E_1/R{T}_p\right) \\ {r}_2=B_{2}exp\left(E_2/R{T}_p\right) \end{gathered}$$

(11)

where $r_1$ and $r_2$ are the two reaction rates. B and E are the frequency factor (or pre-exponential factor) and activation energies for the first and second reactions with their respective subscripts. $T_p$ is the particle temperature. The reaction rates are employed to calculate the weight loss of the particle $\Delta W$.

$$\Delta W={\int }_0^t ( {\alpha }_1{r}_1+{\alpha }_2 r_2 )(\exp {\int }_0^t(r_1+r_2)dt) dt$$

(12)

${\alpha }_1$ and ${\alpha }_2$are the volatile yield factors for the two competing reactions. The Equations (13) and (14) reactions include the yield factors.

$$Coal \stackrel{Heat}{\to } {\alpha }_1 Volatile+\left(1-{\alpha }_1\right)Char$$

(13)

$$Coal \stackrel{Heat}{\to } {\alpha }_2 Volatile+\left(1-{\alpha }_2\right)Char$$

(14)

The values of B and E, as mentioned in Equations (11) and (12), are taken as given by Ubhayakar et al. (1977) [21]. Numerical values of ${\alpha }_1$and ${\alpha }_2$were taken from the proximate analysis by Ma et al. (2012) [8]. This atmospheric value was then edited by using a correction factor because of the high operating pressure in this case. After devolatilization, the surface combustion model is enabled, in which a combustion reaction takes place with the particles, which is activated after the formation of volatiles. After the surface combustion, the inert heating model is again activated.

Another model which is important related to the size of input particles is the Rosin–Rammler model. In this model, the particles have a spectrum of diameter sizes, which is more realistic as compared to a uniform diameter approach, as given in Equation (15).

$$Y_d= e^{-{\left\{\frac{d}{\overline{d}}\right\}}^n}$$

(15)

where $\overline{d}$ is the size constant and n is the size distribution parameter. d is the particle diameter. Physically, this equation depicts that there is an exponential relation between droplet diameter (d) and the mass fraction of droplets having a diameter greater than d.

2.2. Reaction Modeling

The species transport model available in Fluent was used to model the combustion and gasification reactions. This model is based on the convection–diffusion equation, which can capture both diffusion and convection, as given in Equation (16).

$$\frac{\partial \left(\rho Y_i\right)}{\partial t}+ \nabla .\left(\rho \overrightarrow{v}{Y}_i\right) =-\nabla .\overrightarrow{J_i}+R_i+S_i$$

(16)

$\rho$represents the overall density or mixture density of species. Y_i is the mass fraction of a particular species i. $\overrightarrow{v}$stands for the velocity vector for all three directions.$\overrightarrow{J_i}$ represents the diffusion flux for species i, which means the influx of species due to the phenomena of diffusion. $R_i$ and $S_i$ are the rate of generation of species and the source term, respectively. A total of fourteen reactions were employed for simulating the combustion and gasification reactions. The first reaction is the devolatilization reaction, which decides the course of further reactions. After devolatilization, the coal is converted into char and char reactions take place. The char reactions include char combustion with O₂, and gasification of char by H₂O, CO₂, and H₂. After char reactions, the gasification reactions are followed by tar reactions in the end. All reactions mentioned above are volumetric, except the char reactions, which are modeled as particle surface. Practically, the reactions described above are dependent on the turbulent flow of reactants, so a finite rate model with the eddy dissipation concept is used, which enables the turbulence–chemistry interaction. In this option, both finite rate reaction rates and eddy dissipation rates are calculated and the minimum rate from both is selected for the reaction.

After the coal is combusted in the first stage of the gasifier, the heat released is used in various endothermic reactions, which constitute the whole gasification process. The coal is broken into char and volatiles due to pyrolysis. Volatiles further break into some compounds, and this process is called devolatilization. The products of devolatilization strictly depend on the type of coal we use, though commonly, it is seen that the volatile breakup leads to CO, H₂, CO₂, CH₄, etc. In this model, the considered gasification products are CO, CH₄, C₆H₆, N₂, H₂S, and COS. Char reactions follow the devolatilization reactions and are responsible for a significant portion of carbon monoxide, hydrogen, and methane production. Moreover, the volatile breakup was considered to be different for the first and second stages due to their proximity to the combustion zone. Vol1 and Vol2 are the two different reactions to signify the first and second stage volatile breakup. All the reactions are given in

Table 1. Important reactions involved in the study [18].

Reactions	Pre-Exponential Factor	Activation Energy (J/kg mol)
$Vol 1 \to Products$	2.119 × 1011	2.027 × 108
$Vol 2 \to Products$	2.119 × 1011	2.027 × 108
$C+O_2 \to C{O}_2$	87,100	1.494 × 108
$C+H_{2}O \to CO+H_2$	2470	1.75 × 108
$C+C{O}_2 \to 2CO$	2470	1.751 × 108
$C+2 H_2 \to C{H}_4$	1.2	1.49 × 108
$CO+0.5 O_2 \to C{O}_2$	2.239 × 1012	1.674 × 108
$C{H}_4+2O_2 \to C{O}_2+2H_{2}O$	2.119 × 1011	2.025 × 108
$H_2+0.5 O_2 \to H_{2}O$	9.87 × 108	3.1 × 107
$CO+H_{2}O \to C{O}_2+H_2$	2.34 × 1010	2.88 × 108
$C{O}_2+H_2 \to CO+H_{2}O$	2.2 × 107	1.9 × 108
$C{H}_4+H_{2}O \to CO+3 H_2$	8 × 107	2.51 × 108
$C_6{H}_6+7.5 O_2 \to 6 C{O}_2+3 H_{2}O$	1.125 × 109	1.256 × 108
$C_6{H}_6+7.5 O_2 \to 6 C{O}_2+3 H_{2}O$	8 × 108	2.51 × 108
$H_{2}O \to H_2+0.5 O_2$	2.5 × 1010	3.5 × 108

. Benzene present in the volatile breakup represents the tar content of the gasifier, and it is assumed that tar is either oxidized or reacts with water vapor to contribute to H₂ and CO production. The volatile reactions are given in Equations (17) and (18).

$$Vol1 \to 0.29 CO+0.5493 C{H}_4+0.1289 C_6{H}_6+0.037 N_2+0.0485 H_{2}S+0.0054 COS$$

(17)

$$Vol2 \to 0.333 CO+0.755 C{H}_4+0.0653 C_6{H}_6+0.0436 N_2+0.0557 H_{2}S+0.0062 COS$$

(18)

The char reactions are described by the multiple surface reaction model, where the fraction of char mass is depleted or added depending on the reactions. Equation (19) direct the particle surface interaction within the domain.

$$R= D_0\left(C_g-C_s\right) =R_c{\left(C_s\right)}^N$$

(19)

where R is the particle reaction rate, $D_0$ is the bulk diffusion coefficient, and C_g and C_s are the mean gas species concentrations in the bulk and at the particle surface, respectively. R_c is the chemical reaction rate coefficient and N is the apparent reaction order. In general, the reaction follows the scheme given in Equation (20).

$$particle species\left(p\right)+gas phase reactants \left(g\right) \to products$$

(20)

$$\overline{R}{ }_{p,g}=A_p {\eta }_r Y_p R_{p,g}$$

(21)

In Equation (21), $\overline{R}{ }_{p,g}$ is the particle surface depletion rate and $R_{p,g}$ is the rate of reaction for particle species. $A_p$ is the particle surface area, $Y_p$ is the mass fraction of surface species p in the particle, and ${\eta }_r$ is the effectiveness factor.

3. Gasifier Parameter Description and Validation

The geometry of a Phillips E-Gas gasifier industrial-scale gasifier (Ma et al. (2012)) [8] was used for the validation of the model and to conduct the parametric study. This gasifier geometry has two stages of the gasifier; the first stage inhibits the combustion process. The terminologies of the gasifier used are shown in Figure 2.

The oxidizer is induced in this stage only. The second stage is not provided with an oxidizer, and the majority of the gasification reactions take place in this stage. The whole gasifier is 8 m wide and 12 m long. Further geometrical details with mesh representation are given in Figure 3. The first stage has two inlets (Figure 3a), and the second stage has one inlet (Figure 3b). A computational grid of 123,368 elements was finalized after grid independence tests. Almost all elements in the mesh were of hexahedral nature, except the T-joint part, where tetrahedral mesh was used, as shown in Figure 3c. The first stage is fed with coal slurry, which is a mixture of coal powder and water. Computationally, both are given as discrete-phase particles from a single face. For a single inlet in the first stage, the oxidizer is divided between two inlets, and in between these two inlets is the coal slurry inlet. Oxidizer composition was fixed in the ratio O₂:N₂:Ar (95:1:4) for the baseline case during validation. Coal input details are given in

Table 2. Ultimate analysis of coal used in study.

S. No	Element	Ultimate Analysis (wt %)
1.	C	80.52
2.	H	5.68
3.	O	8.68
4.	N	1.57
5.	S	3.53

and

Table 3. Proximate analysis of coal used in study.

S. No	Compound	Proximate Analysis (wt %)
1.	C	34.99
2.	H	44.19
3.	O	9.7
4.	N	11.12

.

Table 4. Boundary conditions.

S. No	Boundary Condition	Values
1.	Coal mass flow rate (First stage)	10.837 (kg/s)
2.	Coal mass flow rate (Second stage)	6.11 (kg/s)
3	Water mass flow rate (First stage)	5.582 (kg/s)
4.	Water mass flow rate (Second Stage)	3.148 (kg/s)
5.	Oxidizer mass flow rate	21.2 (kg/s)
6.	Wall heat loss (First Stage)	30 (W/m2K)
7.	Wall heat loss (Second Stage)	150 (W/m2K)

contains all the boundary conditions required for solving the modeled case in ANSYS.

Heat loss through the gasifier is mainly from the slag rejected during the operation. Walls are insulated and exit gas does carry significant heat with it. To simulate the heat loss, the wall convection heat transfer coefficients are given because slag flow out of the gasifier was not simulated. Free stream temperature and internal emissivity were 300 K and 0.6, respectively. The convection coefficients are also given in Table 4. The input temperature of the oxidizer was kept at 390 K and the discrete particle (coal and water) temperature was 300 K. The velocity of discrete phase particles was kept at 33.1 m/s.

Various industrially used gasifiers have been researched so that validated models for various processes and sub-processes of gasification could be developed. As discussed, earlier gasifiers such as Phillips E Gas, GE gasifier, Siemens gasifier, and KBR Transport gasifier (Breault et al. (2010) [22]) have been used by researchers for different studies, and most of them have been utilized for the development of physical models regarding gasification. In this investigation, we studied the computational models used by Ma et al. (2012) [8]. This study is based on an industrial coal-fired gasifier used in the Wabash River Coal Gasification Repowering Project (Foster (1994) [3]). This gasifier has a significant number of working hours and the project has demonstrated the use of gasification in electricity generation. The Philips E-gas gasifier has been used extensively for energy purposes and as an experimental reference in many computational studies. Many researchers have developed various models that can be used in gasification modeling, whether it is the reaction modeling or the volatile break-up model (Shi et al. (2006) [4], Ma et al. (2012) [8]). Chyou et al. (2013) [9] varied the injector design of this gasifier and studied it computationally. Large-scale energy units, such as this gasifier, can be efficiently designed and desired gas compositions can be simulated. This exercise can be achieved by a computational study in which we can vary the ratios of oxidizer and the fuel used in the gasifiers. Various combinations can be studied, and ratios can be found for which we can obtain the best results for a particular plant. A comparison of the results with previous studies can be carried out to validate the model based on some key parameters, namely, temperature distribution, velocity distribution, and yield comparison at the outlet boundary of the computational model. The results from the previous studies have been compared, as shown in Figure 4, Figure 5 and Figure 6. This zone provides the required temperature for the whole gasification process. The second stage of the gasifier is fed with only coal slurry. The heat from the first stage is utilized in the second stage for various endothermic processes of gasification. The majority of the product yield gases are produced in the second stage.

The high-temperature presence in the first stage was observed, and the second stage showed low-temperature readings compared to the former, as described in Figure 4. The two needle-like projections from the inlets in the first stage can be deduced to be a flame structure interior of the gasifier. We were also able to capture this phenomenon in our study similar to the reference case results of Ma et al. (2012) [8]. The velocity distribution also showed good agreement with the reference case (Figure 5). There was a high-velocity region in the central axis along the horizontal line of the first stage of the gasifier in the reference case; in our case, this region was observed to be slightly smaller in size. CO, H₂, CO₂, and CH₄ were the species that were compared with both experimental and computational reference data. The yields of the present study were found to be in good agreement with the plant data as well as the previous computational study (Figure 6). The major deviation was found in CO yields, which may be due to the difference in the temperature distribution of the first stage, and the difference was close to 10% when compared to plant data, whereas the hydrogen came remarkably close to plant data. CH₄ was also well within the given plant data range and CO₂ was within 10% of the plant data.

4. Results and Discussions

Various parameters that affect the product gas composition were varied and the results were analyzed. We have discussed the effects of oxidizer compositions, the oxidizer to coal ratio, the water to coal ratio, and the solar heat input on the temperature distribution, velocity distribution, and product gas yields of the gasifier. The temperature and velocity distributions were observed on a central plane of the gasifier body. This parametric study can find applications in any industrial unit based on the gasification process and can be modified to predict the results and adjust the required conditions accordingly.

The temperature distribution was observed along the central axis plane of the gasifier covering all important regions. The first stage of the oxidizer seemed to have a higher temperature distribution than the second stage; this can be attributed to the presence of a combustion region in this stage. Higher oxygen content in the oxidizer (95% O₂) caused a rise in the overall temperature of the gasifier, while the oxidizer with the lowest oxygen (25% O₂) led to a decrease, as can be seen in Figure 7.

Similar effects were seen when the O/C (oxidizer to coal) ratio was varied, but in this case, the changes were prominent. The W/C (water to coal) ratio variation gave results that contrasted the oxidizer composition variation, as the temperature decreased (dominantly) in the first stage of the gasifier body as the water content was increased. The induction of solar heat was explored in the study and the effects of induction of such heat in amounts corresponding to 1% of total fuel input suggested a negligible effect on temperature distribution, whereas when this amount was increased to 10%, there was a significant temperature rise in the first stage, as shown in Figure 8.

In velocity distributions, we could easily have an idea of flow trends inside the two-stage gasifier. Velocity contours showed significant deviations in the case of oxidizer composition, as when nitrogen content was increased, the total mass flow rate of the oxidizer also increased. This can be observed in Figure 9. The velocity did not show a sizeable change both in the case of varying O/C and W/C ratios. When solar energy induction was simulated, the velocity values elevated at the throat area of the gasifier only in the case when 10% (of total fuel input) solar energy was induced in the first stage of the gasifier.

The product gas mole fractions are the indicators of the changes that are crucial to the performance of a gasifier when the key parameters are changed. There are many components of product gases, but CO, CO₂, H₂, and CH₄ are used for comparison in this study.

Higher content of CO, CH₄, or H₂ is preferred in a product gas, as it is primarily used for combustion purposes. CO content was seen to increase when the oxygen percentage was increased in the input oxidizer. The same gas showed uneven behavior when the O/C ratio was varied. As CO is a product of partial oxidation of carbon, the reduction in the O/C ratio leads to an increase in CO yield, and the yield reduced when the ratio was increased. Similar observations were reported by Wang et al. [16] and Kinoshita et al. [23]. The CO yield can also be seen as sensitive to the interior body temperature of the gasifier, as in cases of varying oxidizer composition, O/C ratio, and W/C ratio, the increased temperatures of the body resulted in higher CO production, strengthening the theoretical fact that most of the CO-producing reactions are endothermic. Figure 10 also shows CO yield variation with the oxidizer composition. As H₂ has water as its main contributing reactant, the H₂ yield increased in the case of increasing W/C ratios. It reacted unevenly to the O/C ratio, as it increased only for an optimal value of 0.76 and decreased for the remaining values of 0.66 and 0.86. When the oxidizer composition was adjusted (from 95% to 25% O₂), the H₂ yields showed a constant decrease as the O₂ content was reduced. This again can be explained by the reducing body temperature of the gasifier.

Figure 11a,b shows the yield comparisons for 1% and 10% solar heat influxes with reference to gasification without the solar heat input. Moreover, when solar heat was induced at 1%, it was observed that H₂ yield showed a minute change, but it showed a significant rise when 10% solar heat was induced in the first stage of the gasifier, as shown in Figure 9. In the 1% influx case, CO yield was greater in the case of the first stage influx. CO₂, H₂, and CH₄ showed less variation as compared to CO. The second stage influx showed minor changes in almost all gas concentrations. CO increased considerably in the 10% solar heat influx case, especially in the second stage influx case (Figure 9b) because it was the region in proximity to the endothermic gasification reaction. CO₂ yield decreased in both cases with a greater decrease in the second stage. This was due to the increased CO production in both cases, and the increase and decrease in both the gases complement each other. CH₄ also decreased significantly for both cases. H₂ yield overall increased, and interestingly, this increase was greater in the first stage solar heat influx situation. The sudden H₂ and CO increase may be attributed to less CH₄ formation. Overall, we can say the solar input is effective in increasing the CO and H₂ yields and it has a significant role to play in increasing the overall output of the gasifier as less biomass is used for supplying the process heat.

5. Conclusions

We studied the gasification process using computational fluid dynamics in this investigation. Important control parameters such as oxygen content, water content, and solar heat input were varied to investigate their effects on product gas yields, as well as temperature and velocity distributions within the gasifier. Solar energy induction was simulated at two different locations on the gasifier body, with two different solar heat induction magnitudes (1% and 10% heat of total coal input to the gasifier). Increasing the oxidizer’s oxygen content increased the desired product gases in the output gas. Because most important gasification reactions are endothermic, an increase in body temperature was responsible for this improvement. In addition, a higher temperature causes the gasifier unit to start up faster. We determined that 0.763 was the optimal O/C ratio and increasing or decreasing this ratio did not produce a satisfactory result. Increasing the water fraction was another critical factor because it significantly increased the hydrogen content of the product gas.

Solar heat induction in the gasifier body produced interesting results, resulting in a significant change in gasification output due to solar heat. When solar heat was induced in the first and second stages, CO and H₂ yields increased. The 10% solar heat induction in the second stage, on the other hand, resulted in a significant improvement in the product gas composition.