Combustion of Syngas Fuel in Gas Turbine Can Combustor

Abstract

Numerical investigation of the combustion of syngas fuel mixture in gas turbine can combustor is presented in this paper. The objective is to understand the impact of the variability in the alternative fuel composition and heating value on combustion performance and emissions. The gas turbine can combustor is designed to burn the fuel efficiently, reduce the emissions, and lower the wall temperature. Syngas mixtures with different fuel compositions are produced through different coal and biomass gasification process technologies. The composition of the fuel burned in can combustor was changed from natural gas (methane) to syngas fuel with hydrogen to carbon monoxide (H₂/CO) volume ratio ranging from 0.63 to 2.36. The mathematical models used for syngas fuel combustion consist of the k-ε model for turbulent flow, mixture fractions/PDF model for nonpremixed gas combustion, and P-1 radiation model. The effect of syngas fuel composition and lower heating value on the flame shape, gas temperature, mass of carbon dioxide (CO₂) and nitrogen oxides (NO _x ) per unit of energy generation is presented in this paper. The results obtained in this study show the change in gas turbine can combustor performance with the same power generation when natural gas or methane fuel is replaced by syngas fuels.

1. Introduction

Over the past decades domestic and imported oil was used for transportation, and domestic coal and natural gas have been used as the primary fuels for power generation systems. Today the emission regulations for power plant have become more stringent. The concern today with the combustion of fossil fuels is the new emission regulations for power plant with regards to carbon dioxides (CO₂) and nitrogen oxides ( ${\text{NO}}_x$ ). Nitrogen oxides ( ${\text{NO}}_x$ ) are responsible for smog and acid rain, and the carbon dioxides (CO₂) are one of the main green house gases responsible for global warming. Another concern with fossil fuels is the high cost of imported oil. Alternative fuels that can be produced using local feed stocks, burn efficiently, and produce low emissions are needed. With the development of advanced technologies, coal, biomass, or waste products can be used in power generation systems to produce low emissions comparable to the ones obtained with natural gas fuel. This can be achieved through the Integrated Gasification Combined Cycle (IGCC). Different types of gasifiers are used to gasify the solid fuels (coal, biomass or waste products) and produce synthetic gas. The syngas is then cleaned and burned in gas turbine and the hot exhaust gas is used to produce steam for steam turbine. Future power generation systems using Integrated Gasification Combined Cycle (IGCC) will have a higher efficiency and generate lower carbon dioxide and nitrogen oxides emissions. The IGCC systems are used to produce syngas fuel with different compositions from solid fuel feed stocks using different gasification process technologies. The syngas produced through the gasification process consists mainly of hydrogen (H₂) and carbon monoxide (CO) and inert gas such as nitrogen (N₂), water vapor (H₂O), and carbon dioxide (CO₂). Syngas has also less energy density (KJ/Kg) than natural gas. The main characteristics of the syngas fuel are the lower heating value, the H₂/CO ratio, and the fraction (up to 50%) of noncombustibles such as steam, carbon dioxide, and nitrogen. For syngas fuels combustion, the effect of hydrogen content is very important. The burning velocity increases with the hydrogen content because the density of the mixture is very low compared to the density of natural gas. The replacement of methane with syngas with high hydrogen content will help to reduce the CO₂ emissions. On the other hand, the dilution of fuel with nitrogen, water, and carbon dioxide reduces the peak flame temperature and consequently the ${\text{NO}}_x$ emissions.

Gas turbines are designed primarily to be fueled with natural gas (consisting primarily of methane), and supplying them with syngas (fuel gas from biomass, coal, and waste gasification) presents certain challenges that must be addressed. How can we burn efficiently syngas fuel with different chemical compositions and heating values in gas turbine combustors designed for natural gas? In order to meet these challenges, we need to understand the physical and chemical processes of syngas combustion. Information regarding syngas flame shape, flame speed, gas temperatures and pollutant emissions such as ${\text{NO}}_x$ and CO₂ for a range of syngas compositions and heating values is needed for the design of gas turbine combustors. Giles et al. [1] performed a numerical investigation on the effects of syngas composition and diluents on the structure and emission characteristics of syngas counterflow diffusion flame. The counter flow syngas flames were simulated using two representative syngas mixtures, 50%H₂/50%CO and 45%H₂/45%CO/10%CH₄ by volume, and three diluents, N₂, H₂O, and CO₂. The effectiveness of these diluents was characterized in terms of their ability to reduce ${\text{NO}}_x$ in syngas flames. The results indicated that syngas nonpremixed flames are characterized by relatively high temperatures and high ${\text{NO}}_x$ concentrations and emission indices. The presence of methane in syngas decreases the peak flame temperature, but increases the formation of prompt NO significantly. They also concluded that the presence of methane in syngas reduces the effectiveness of all three diluents. Lean premixed combustion of hydrogen–syngas/methane fuel mixtures was investigated experimentally by Alavandi and Agarwal [2]. Methane (CH₄) content in the fuel was decreased from 100% to 0% (by volume), with the remaining amount split equally between carbon monoxide (CO) and hydrogen (H₂), the two reactive components of the syngas. Experiments for different fuel mixtures were conducted at a fixed air flow rate, while the fuel flow rate was varied to obtain a range of adiabatic flame temperatures. The CO and nitric oxide (NO _x ) emissions were measured downstream of the burner, in the axial direction to identify the postcombustion zone and in the transverse direction to quantify combustion uniformity. The results show that increasing H₂/CO content in the fuel mixture decreased both the CO and NO _x emissions. Experimental study on the fundamental impact of firing syngas in gas turbines was performed by Oluyede [3]. The goal of this study was to determine the appropriate amount of reduction in firing temperature needed to maintain the same hot section temperatures as experienced with natural gas firing. The results show that volume fraction of hydrogen content in syngas fuel significantly impacts the life of hot sections as a result of higher flame temperature for hydrogen rich fuels and also the moisture content of combustion products. Correlations were obtained indicating the level of firing temperature reduction, necessary for hot section durability in terms of hydrogen contents and lower heating value of the fuel. The combustion of hydrogen-enriched methane in a lean premixed swirl burner was investigated by Schfere [4]. The results, using methane/hydrogen fuel mixtures, showed that the addition of up to 41% hydrogen significantly extended the lean burning limit. For operating conditions near the lean stability limit, the addition of a moderate amount of hydrogen to the methane/air mixture resulted in a significant increase in the OH concentration and a more robust appearing flame. Pater [5] investigated the thermo acoustic instabilities during turbulent syngas combustion. He used numerical method to predict acoustic fields and instabilities during syngas combustion. The model was used to identify frequencies at which instabilities occur.

The challenges of fuel diversity while maintaining superior environmental performance of gas turbine engines were addressed by Rahm et al. [6]. They reviewed the combustion design flexibility that allows the use of a broad spectrum of gas and liquid fuel including emerging synthetic choices. Gases include ultra-low heating value process gas, syngas, ultrahigh hydrogen, or higher heating capability fuels. The integration of heavy-duty gas turbine technology with synthetic fuel gas processes using lowvalue feed stocks in global power generation marketplace was covered by Brdar and Jones [7]. In their paper they summarized the experience gained from several syngas projects and lessons learned that continue to foster cost reductions and improve the operational reliability of gas turbine. They concluded that further improvements in system performance and plant design are needed in the future. The design of combustion systems using syngas as fuel can take advantage of CFD analysis to optimize the efficiency of the combustion system with respect to the limitations of pollutants emission. The aim of this work is to analyze the fundamental impacts of firing syngas in gas turbine combustor and predict the changes in the firing temperature and emissions with respect to natural gas or methane combustion.

2. Governing Equations

The mathematical equations describing the syngas fuel combustion are based on the equations of conservation of mass, momentum, and energy together with other supplementary equations for the turbulence and combustion. The standard k-ε turbulence model is used in this study. The equations for the turbulent kinetic energy k and the dissipation rate of the turbulent kinetic energy ε are solved. For non premixed combustion modeling, the mixture fraction/PDF model is used. The timeaveraged gas phase equations for steady turbulent flow are:

$\frac{\partial }{\partial x_j}\left(\rho u_i\Phi \right)=-\frac{\partial }{\partial x_i}\left({\Gamma }_{\Phi }\frac{\partial \varphi }{\partial x_i}\right)+S_{\Phi }.$

(1)

Φ is the dependent variable that can represent the velocity u _i , T the temperature, k the turbulent kinetic energy, ε the dissipation rate of the turbulent kinetic energy, and f the mixture fraction. The governing equations are:

2.1. Continuity

$\frac{\partial \overline{\rho u_i}}{\partial x_i}=0.$

(2)

2.2. Momentum Equation

$\frac{\partial \left(\overline{\rho u_i{u}_j}\right)}{\partial x_j}=-\frac{\partial \overline{P}}{\partial x_i}+\frac{\partial \left(\overline{t_{ij}}+\overline{{\tau }_{ij}}\right)}{\partial x_j},$

(3)

where ${\overline{t}}_{ij}$ is the viscous stress tensor defined as:

$\begin{array}{cc}\hfill {\overline{t}}_{ij}& =\mu \left[\left(\frac{\partial \overline{u_i}}{\partial x_j}+\frac{\partial \overline{u_j}}{\partial x_i}\right)-\frac{2}{3}\frac{\partial \overline{u_k}}{\partial x_k}{\delta }_{ij}\right],\hfill \\ \hfill & {\delta }_{ij}=1\text{if}i=j,{\delta }_{ij}=0\text{if}i\ne j.\hfill \end{array}$

(4)

${\overline{\tau }}_{ij}$ is the average Reynolds stress tensor defined as: ${\overline{\tau }}_{ij}=-\overline{\rho u_i^{\prime }{u}_j^{\prime }}$

$\overline{{\tau }_{ij}}={\mu }_t\left[\left(\frac{\partial \overline{u_i}}{\partial x_j}+\frac{\partial \overline{u_j}}{\partial x_i}\right)-\frac{2}{3}\frac{\partial \overline{u_k}}{\partial x_k}{\delta }_{ij}\right]-\frac{2}{3}\left(\overline{\rho k{\delta }_{ij}}\right),$

(5)

where k is the average turbulent kinetic energy defined as: $\overline{k}=(1/2)\overline{u_i^{\prime }{u}_j^{\prime }}$ , μ _t is the turbulent eddy viscosity expressed as: ${\mu }_t=c_{\mu }\overline{\rho k^2/\overline{\epsilon }}$ , where Cμ is constant ( $\text{C}\mu =0.09$ ) and $\overline{\epsilon }$ is the average dissipation rate of the turbulent kinetic energy and defined as follows: $\overline{\epsilon }=\nu \overline{\partial u_i^{\prime }}/\partial x_j\overline{\partial u_i^{\prime }}/\partial x_j$ .

2.3. Turbulent Kinetic Energy Equation

$\frac{\partial \left(\overline{\rho k{u}_j}\right)}{\partial x_j}=\frac{\partial \left[\left(\mu +{\mu }_t/{\sigma }_k\right)\left(\partial \overline{k}/\partial x_j\right)\right]}{\partial x_j}+G_k-\overline{\rho \epsilon },$

(6)

where σ _k = 1 and G _k is the production of the turbulent kinetic energy defined as

$G_K={\mu }_t\left[\left(\frac{\partial \overline{u_i}}{\partial x_j}+\frac{\partial \overline{u_j}}{\partial x_i}\right)\right]\frac{\partial \overline{u_i}}{\partial x_j}-\frac{2}{3}\frac{\partial \overline{u_i}}{\partial x_j}{\delta }_{ij}\left[{\mu }_t\frac{\partial \overline{u_k}}{\partial x_k}+\overline{\rho k}\right].$

(7)

2.4. Dissipation of the Kinetic Energy

$\frac{\partial \left(\overline{\rho \epsilon u_j}\right)}{\partial x_j}=C_{\epsilon 1}\frac{\overline{\epsilon }}{k}{G}_k+\frac{\partial \left[\left(\mu +{\mu }_t/{\sigma }_{\epsilon }\right)\left(\partial \overline{\epsilon }/\partial x_j\right)\right]}{\partial x_j}-C_{\epsilon 2}\overline{\rho }\overline{\frac{{\epsilon }^2}{k}},$

(8)

where $C_{\epsilon 1}=1.44$ , $C_{\epsilon 2}=1.92$ , and σ _ε = 1.3.

2.5. Mixture Fraction f

In non-premixed combustion, fuel and oxidizer enter the reaction zone in distinct streams. The PDF/mixture fraction model is used for non premixed combustion modeling. In this approach individual species transport equations are not solved. Instead, equation for the conserved scalar (f) is solved, and individual component concentrations are derived from the predicted mixture fraction distribution. The mixture fraction equation is given by

$\frac{\partial \left(\overline{\rho f{u}_j}\right)}{\partial x_j}=\frac{\partial \left[\left({\mu }_t/{\sigma }_t\right)\left(\partial \overline{f}/\partial x_j\right)\right]}{\partial x_j}+S_m.$

(9)

The mixture fraction, f, can be written in terms of elemental mass fraction as

$f=\frac{Z_k-Z_{k,\mathit{\text{O}}}}{Z_{k,\mathit{\text{F}}}-Z_{k,\mathit{\text{O}}}},$

(10)

where Z _k is the element mass fraction of some element k. Subscripts F and O denote fuel and oxidizer inlet stream values, respectively. For the mixture fraction approach, the equilibrium chemistry PDF model is used. The equilibrium system consists of 13 species (C, CH₄, CO, CO₂, H, H₂, H₂O, N₂, NO, O, O₂, OH, HO₂). The chemistry is assumed to be fast enough to achieve equilibrium.

2.6. Energy Equation

$\begin{array}{cc}\hfill & \frac{\partial \left(\overline{\left(\rho E+p\right)uj}\right)}{\partial x_j}\hfill \\ \hfill & \hspace{1em}=\frac{\partial \left[\left(k_{\text{eff}}\right)\left(\partial \overline{T}/\partial x_j\right)-{\sum }_j{h}_j{J}_j+\left({\overline{\tau }}_{\text{eff}}{u}_j\right)\right]}{\partial x_j}+S_h,\hfill \end{array}$

(11)

where E is the total energy ( $E=h-p/\rho +v^2/2$ , where h is the sensible enthalpy), k_eff is the effective conductivity ( $k+k_t$ : laminar and turbulent thermal conductivity), J _j is the diffusion flux of species j, and S _h is the term source that includes the heat of chemical reaction, radiation, and any other volumetric heat sources.

2.7. Equation for the P-1 Radiation Model—Radiation Flux Equation

The P-1 radiation model is used in this study to simulate the radiation from the flame. The radiation model is based on the expansion of the radiation intensity into an orthogonal series of spherical harmonics (Cheng [8] and Siegel and Howell [9]). The P-1 radiation model is the simplest case of the P-N model. If only four terms in the series are used, the following equation is obtained for the radiation flux:

$q_r=-\frac{1}{3\left(a+{\sigma }_S\right)-C{\sigma }_S}\nabla G,$

(12)

where a is the absorption coefficient, σ _S is the scattering coefficient, G is the incident radiation, and C is the linear-anisotropic phase function coefficient (Cheng [8] and Siegel and Howell [9]).

The transport equation for G is

$\begin{array}{c}\nabla ·\left(\Gamma \nabla G\right)-aG+4a\sigma T^4=S_G,\\ \Gamma =\frac{1}{\left(3\left(a+{\sigma }_S\right)-C{\sigma }_S\right)}.\end{array}$

(13)

3. Geometry, Boundary Conditions, Mesh, and Numerical Method

The gas turbine can combustor is designed to burn the fuel efficiently, lower the emissions, and keep the combustor wall temperatures low. The basic geometry of the gas turbine can combustor is shown in Figure 1. The size of the combustor is 590 mm in the Z direction, 250 mm in the Y direction, and 230 mm in the X direction. The primary inlet air is guided by vanes to give the air a swirling velocity component (see Figure 1(a)). The boundary conditions of the primary air are as follows: the injection velocity is 10 m/s, the temperature is 300 K, the turbulence intensity is 10%, mixture fraction f = 0 and the injection diameter is 85 mm. The fuel is injected through six fuel inlets in the swirling primary air flow (see Figure 1(a)). The boundary conditions of the fuel are as follows: mass flow rate, 0.001 Kg/s, the temperature is 300 K, the turbulence intensity is 10%, mixture fraction f = 1 and the injector diameter is 4.2 mm. The secondary air or dilution air is injected at 0.1 meters from the fuel injector to control the flame temperature and ${\text{NO}}_x$ emissions. The secondary air is injected in the combustion chamber through six side air inlets each with a diameter of 16 mm (see Figure 1(b)). The boundary conditions of the secondary air are as folllows: the injection velocity is 6 m/s, the temperature is 300 K, the turbulence intensity is 10%, mixture fraction f = 0 and the injection diameter is 16 mm. The can combustor outlet has a rectangular shape (see Figure 1(b)) with an area of 0.0150 m². A quality mesh was generated for the can combustor (see Figure 2). The mesh consists of 106,651 cells or elements (74189 tetrahedra, 30489 wedges, and 1989 pyramids), 234368 faces, and 31433 nodes. The grid quality was checked, and the results showed a maximum cell squish of 0.94, maximum cell skewness of 0.99, and a maximum aspect ratio of 83.17. The finite volume method and the first-order upwind method were used to solve the governing equations. The solution procedure for a single-mixture-fraction system was to (1) complete the calculation of the PDF look-up tables first, (2) start the reacting flow simulation to determine the flow files and predict the spatial distribution of the mixture fraction, (3) continue the reacting flow simulation until a convergence solution was achieved, and (4) determine the corresponding values of the temperature and individual chemical species mass fractions from the look-up tables. The convergence criteria were set to 10⁻³ for the continuity, momentum, turbulent kinetic energy, dissipation rate of the turbulent kinetic energy, and the mixture fraction. For the energy and the radiation equations, the convergence criteria were set to 10⁻⁶.

Geometry of the gas turbine can combustor, (a) primary air (blue) and six fuel inlets, (b) secondary air (blue) and outlet.

Figure 2

Mesh for the basic geometry of gas turbine can combustor.

4. Results

The impact of the variability in the syngas fuel composition and low heating value on the combustion performance and emissions in gas turbine can combustor is performed in this study. Table 1 shows the composition for the five syngas fuel and their lower heating values selected for this CFD analysis. The syngas fuels were produced using different gasification processes (Todd [10]) and using different feed stocks (coal, biomass, waste). The range of the constituents volume fractions for the selected syngas fuels are hydrogen (1) (H₂) = 22.6% −.6%, (2) carbon monoxide (CO) = 23.6%–46.6%, (3) methane (CH₄) = 0.1%–6.9%, (4) carbon dioxide (CO₂) = 5.6%–17.9%, (5) Nitrogen (N₂) = 1.1 %–49.3%, and (6) water (H₂O) = 0.3%–39.8%. The hydrogen to carbon monoxide volume ratio for these five syngasis between 0.63 and 2.36. Table 1 shows also that the lower heating values for the syngas fuels are smaller compared to the lower heating value of the methane. It is also noted that syngas 1 (Schwarze Pumpe) has the highest hydrogen volume fraction (61.9%), syngas 2 (Exxon Singapore) has the highest carbon dioxide volume fraction, syngas 3 (Tampa) has the highest carbon monoxide volume fraction (46.6%), and syngas 5 (Sarlux) has the highest water vapor volume fraction (39.8%).

Table 1:

Syngas compositions.

Constituents	Syngas 1 Schwarze pumpe	Syngas 2 Exxon singapore	Syngas 3 Tampa	Syngas 4 PSI	Syngas 5 Sarlux	Methane (CH4)
H2	61.9	44.5	37.2	24.8	22.7	0
CO	26.2	35.4	46.6	39.5	30.6	0
CH4	6.9	0.5	0.1	1.5	0.2	100
CO2	2.8	17.9	13.3	9.3	5.6	0
N2	1.8	1.4	2.5	2.3	1.1	0
H2O	0.4	0.3	0.3	22.6	39.8	0
Volumetric heating value KJ/m3	12492	9477	9962	8224	—	33570
Lower heating value MJ/Kg	27.8	12.8	12.7	10.4	—	50.1
H2/CO	2.36	1.26	0.8	0.63	0.74	—

The contours of the predicted gas temperature for the combustion of methane in gas turbine can combustor are shown in Figures 3 and 4. The maximum gas temperature for methane combustion is 2200 K. For the validation of the combustion model, the predicted flame temperature for methane combustion was compared to the adiabatic flame temperature. For natural gas or methane fuel and with initial atmospheric conditions (1 bar and 20 C), the theoretical flame temperature produced by the flame with a fast combustion reaction is 2233 K. The predicted maximum temperature of the combustion products or the adiabatic flame temperature compares well with the theoretical adiabatic flame temperature. The peak gas temperature is located in the primary reaction zone. The fuel from the six injectors is first mixed in the swirling air before burning in the primary reaction zone. The gas temperature decreases after the primary reaction zone due to the dilution of the flame with the secondary air.

Figure 3

Temperature contours (X-Z Plane, Y = 0): combustion of methane in gas turbine can combustor.

Temperature contours (X-Y Plane)—Combustion of methane in gas turbine can combustor.

Figure 4 shows the temperature contours (X-Y Plane) for methane combustion in can combustor at different axial positions (Z = 134 mm to 590 mm). The first contour (Z = 134 mm) represents the gas temperature near the six fuel injections. The fuel is injected from the six fuel inlets and mixedwith the swirling air before the start of the combustion. The size of the flame increases downstream and reach a maximum radius at Z = 200 mm. The radius and temperature of the flame decrease after that with the increase of the axial distance Z (Z = 300, 400, and 500 mm). The lowest gas temperature is reached at the exit (Z = 590 mm) of the gas turbine can combustor. A uniform gas temperature distribution is obtained at the exit of the can combustor as shown in Figures 3 and 4.

The contours of the velocity w (Z-component) for methane combustion in gas turbine can combustor is shown in Figure 5. The primary air is injected in the Z-direction with initial velocity of 10 m/s. The primary air is accelerated (up to 16 m/s) at the entrance of the combustors due to the presence of swirlers vanes (see Figure 1). Swirlers curved vanes are used to generate recirculation zone at the entrance of the combustion chamber. Strong recirculation regions are produced in the fuel injection region. This will help to increase the turbulence and mix very well the fuel and air in the primary reaction zone. This will in turn burn the fuel efficiently and reduce the pollutants emissions. Figure 6 shows the contours (X-Y Plane) of the velocity swirling strength for Z = 134 mm to 400. Strong recirculation regions with strong swirling strength are shown at the entrance regions near the fuel injection. The contours show that the velocity swirling strength decreases downstream with the increase of the axial distance Z. The velocity swirling strength contours at Z = 200 shows an annulus region with high swirling strength.

Figure 5

Contours of the velocity w (z-direction) in the y-z plane: combustion of methane in the gas turbine can combustor.

Contours of velocity swirling strength (X-Y plane): combustion of methane in gas turbine can combustor.

The results of methane combustion in gas turbine can combustor are used as a baseline for comparison with syngas fuel combustion. The effect of syngas fuel compositions on gas temperature, CO₂ and ${\text{NO}}_x$ emissions are investigated first. Table 2 shows the fuel, primary and secondary air injection conditions, and the power generated for methane and syngas fuels. It is noted that the first part of the CFD analysis was performed with the same fuel mass flow rate and different power generation. The gas temperature contours (X-Z plane at Y = 0) for methane and syngas fuels combustion in gas turbine can combustor are shown in Figure 7. The gas temperature for the five syngas tested in this study shows a lower gas temperature compared to the temperature of methane. It is noted that all the syngas fuel have lower heating value than the methane. The volumetric heating value of the methane is 2.68 to 4.08 times higher than the volumetric heating value of the syngas fuels tested in this study. In addition to that the flame temperature depends also on the syngas fuels compositions. For example, the maximum gas temperature predicted for syngas 5 (Sarlux) is about 1734 K. The maximum temperature for syngas 5 is 466 K less than the maximum flame temperature for methane. This is due to the high volume fraction of water (39.8%) in syngas 5. The maximum temperature for syngas 3 is 2008 K compared to 2200 K for methane. Syngas 3 has the highest volume fraction of carbon monoxide CO (46.6%). The syngas fuel composition and heating value are not only affecting the maximum flame temperature but also the shape and size of the flame. The flame shape and size are changing when we switch the fuel from pure methane to syngas fuels. The syngas combustion shows a shorter flame compared to the one obtained with methane. Replacing methane with syngas fuels with different hydrogen concentration will affect the combustion process. The reaction zone or the flame is located in the vicinity of the fuel injection region as shown in Figure 7 for all five syngas with hydrogen to carbon monoxide ratio between 0.63 and 2.36. The flame temperature for syngas gas depends not only on the hydrogen to carbon monoxide ratio (combustible constituents) but also on the volume fraction of the noncombustible (CO₂, N₂, H₂O) in the syngas. The higher the hydrogen volume fraction in the syngas is, The higher the gas temperature is and the higher the volume fraction of inert gas in syngas is, the lower the gas temperature is. The contours of carbon dioxide (CO₂) mass fraction for methane and syngas fuel (Schwarze Pumpe) combustion are shown in Figure 8. The figure shows higher CO₂ concentrations in the reaction zone. It is also noted that higher CO₂ concentrations are generated with methane when the fuel mass rate was kept constant. With the same fuel flow rate, the power generated by the syngas fuels tested in this study is about 20% to 55% the power generated by the methane. For the same power generation, the fuel mass flow rate of the syngas fuel and the CO₂ emission will be higher. With a constant fuel mass flow rate, the syngas fuel combustion shows less carbon dioxide formation inside the can combustor for all the five syngas tested in this study. The average carbon dioxide mass fractions at the exit of the combustor generated with methane and syngas fuels with constant fuel mass flow rate are shown in Figure 9(a). A net reduction (30% to 49%) of CO₂ emissions at the exit of the combustor when methane is replaced with syngas fuel (see Figure 9(a)) but at the same time the power generated by the syngas gas decreased also by 20% to 55%. The results shown in Figure 9(a) are obtained with methane and syngas fuel with the same mass fuel rate at the injection and different power generated by the can combustor (see Table 2). For the evaluation of the carbon loading to the environment accounted by the can combustor, the results should be presented as the average mass of CO₂ emitted per unit of energy generation. The average mass flow rate of CO₂ emissions (Kg/s) at the exit of the combustor was calculated. The CO₂ mass flow rate (Kg/s) was normalized with power (KW or KJ/s) to determine the mass of CO₂ emitted per unit of energy generation (Kg/KJ). The results of the mass of CO₂ per unit energy generation are shown in Figure 9(b). The results show clearly that mass of CO₂ emitted per unit of energy generation (Kg/KJ) is higher for the syngas fuels (Exxon Singapore, Tampa, and PSI) with lower heating value (10.4–12.8 MJ/Kg) or power generation (10.4 to 12.8 KW). The only syngas fuel that shows a reduction of the average mass of CO₂ emitted per unit of energy generation (Kg/KJ) compared to methane fuel is Schwarze Pumpe syngas. The Schwarze Pumpe syngas fuel ha a lower heating value of 27.8 MJ/Kg, and the power generated during the combustion of this fuel in the can combustor is 27.8 KW. The average mass of CO₂ emitted per unit of energy generation (Kg/KJ) for the Schwarze Pumpe syngas fuel decreased by 12% compared to methane fuel.

Table 2:

Fuel and air injection conditions and power generated.

Constituents	Syngas 1 Schwarze Pumpe	Syngas 2 Exxon Singapore	Syngas 3 Tampa	Syngas 4 PSI	Syngas 5 Sarlux	Methane (CH4)
Fuel mass flow rate (Kg/s)	0.001	0.001	0.001	0.001	0.001	0.001
Velocity of primary air (m/s)	10	10	10	10	10	10
Velocity of secondary air (m/s)	6	6	6	6	6	6
Total mass flow rate at the exit (Kg/s)	0.0783	0.0783	0.0783	0.0783	0.0783	0.0783
Lower heating value MJ/Kg	27.8	12.8	12.7	10.4	—	50.1
Power KW	27.8	12.8	12.7	10.4	—	50.1

Temperature contours for natural gas and syngas fuel mixtures with the same fuel mass flow rate: effect of Syngas composition (H₂/CO).

Contours of CO₂ mass fraction.

(a) Average carbon dioxide (CO₂) mass fractions at the exit of can combustor. (b) Average mass of carbon dioxide at the exit per unit of energy generation.

The ${\text{NO}}_x$ emissions from syngas combustion were also calculated in this study. The ${\text{NO}}_x$ concentrations emitted from combustion systems are generally low. The ${\text{NO}}_x$ chemistry has negligible influence on the predicted flow field, temperature, and major combustion product concentrations. The ${\text{NO}}_x$ model used in this study was a postprocessor to the main combustion calculation. First the reacting flows are simulated without ${\text{NO}}_x$ emissions until the convergence of the main combustion calculation was obtained; after that the desired ${\text{NO}}_x$ models (thermal, prompt, fuel) were enabled to predict the ${\text{NO}}_x$ emissions. For ${\text{NO}}_x$ emissions prediction (postprocessing), only the equation for NO will be solved, and the solution for the other equations (mass, momentum, energy, radiation, turbulent kinetic energy, dissipation of the turbulent kinetic energy, and mixture fraction) will be turned off. The calculation for ${\text{NO}}_x$ emissions will continue until the NO species residuals are below 10⁻⁶ (convergence of NO solution). The contours of NO mass fractions inside the can combustor obtained with the same fuel mass flow rate or different power (see Table 2) are shown in Figure 10. Like the gas temperature contours, the NO mass fraction inside the combustor decreases when the baseline fuel (CH₄) is replaced with syngas fuel with lower heating value. The thermal NO emissions are function of the gas temperature. The higher is the temperature, the higher is the NO mass fractions. The gas temperature during syngas combustion depends on the lower heating value of the fuel and fuel composition. The temperature increases with the increase of the volume fraction of the combustible constituents in the fuel such as hydrogen, carbon monoxides and methane. On the other hand, the presence of non combustible constituents in the syngas such as water, nitrogen, and carbon dioxides reduces the temperature of the flame and consequently the NO mass fractions. Figure 11(a) shows the average NO mass fraction at the exit of the can combustor. If the fuel mass flow rate is kept constant and the power generated from the syngas fuels is 20% to 55% the power generated by methane fuel, the net NO mass fraction reduction for syngas 1 and syngas 5 is, respectively, 11.5% and 97.6% compared to methane. This reduction is proportional to the amount of inert constituents in syngas fuel. The volume fraction of inert constituents for syngas 1 and syngas 5 is, respectively, 5% and 46.5%. The higher is the amount on non combustible constituents in the syngas fuel the lower is the NO mass fraction at the exit of the can combustor. For non premixed combustion, syngas fuel is used to control the ${\text{NO}}_x$ emissions by diluting the syngas with nitrogen, steam, and carbon dioxides. To assess the amount of ${\text{NO}}_x$ loading to the environment accounted by the can combustor, the results should be presented as the average mass of NO emitted per unit of energy generation. The average mass flow rate of NO emissions (Kg/s) at the exit of the combustor was calculated. The NO mass flow rate (Kg/s) was normalized with power (KW or KJ/s) to determine the mass of NO emitted per unit of energy generation (Kg/KJ). The results of the mass of NO per unit energy generation are shown in Figure 11(b). The results show that mass of NO per unit energy generation is higher for most of the syngas fuels tested in this study (Schwarze Pumpe, Exxon Singapore, and Tampa). The only syngas fuel that shows a 33% reduction of the mass of NO per unit energy generation is the PSI syngas fuel with high water volume fraction (22.6%) as shown in Table 1.

Contours of NO mass fraction.

(a) Average NO mass fractions at the exit of the can combustor. (b) Average mass of NO at the exit per unit of energy generation.

It is noted that all the results presented in Figures 7 to 11 were obtained with a constant fuel flow rate. The syngas fuel composition and lower heating values were changed but the fuel mass flow rate was kept constant at 0.001 Kg/s. In fact when syngas heating value decreases, the mass fuel flow rate should increase to keep the same energy input in the can combustor. For the same fuel input in the can combustor, the fuel mass flow is four to five times greater than for methane or natural gas, due to the lower heating value of the syngas. Methane has a high lower heating value of 50.1 MJ/Kg. The lower heating values for the five syngas tested in this study is between 10.4 MJ/Kg and 27.8 MJ/Kg. Syngas is primarily composed of carbon monoxide and hydrogen but also contains a significant fraction (up to 50%) of non combustibles (nitrogen, steam, carbon dioxide). The volumetric heating values of pure methane, hydrogen, and carbon monoxide are, respectively, 33.5, 10.2, and 12.6 MJ/m³. The hydrogen and carbon monoxide has a lower volumetric heating value about 1/3 of the lower heating value of methane. When combined with nitrogen, water, and carbon dioxide in the gas stream, the syngas fuel volumetric heating value is even smaller (see Table 1). A comparison of the combustion process and emissions between the methane and syngas fuels with the same power and different fuel mass flow rate was investigated (see Table 3). The fuel mass flow rate for syngas 1 (Schwarze Pumpe) was increased from 0.001 Kg/s to 0.0018 Kg/s and syngas 3 (Tampa) was increased from 0.001 Kg/s to 0.0039 Kg/s to match the same heat input for the methane. It is noted that the heat input from methane combustion is 50.1 KW with a fuel mass flow rate of 0.001 Kg/s. The heat input from synags 1 (Schwarze Pumpe) with a fuel flow rate of 0.001 Kg/s and 0.0018 Kg/s is 27.8 KW and 50.1 KW. The heat input from synags 3 (Tampa) with a fuel flow rate of 0.001 Kg/s, and 0.0039 Kg/s is 12.7 KW and 50.1 KW. The fuel mass flow rates for syngas 1 and syngas 3 were increased, respectively, by a factor of 1.8 and 3.9 to match the same heat input for methane. The heat input (KW) was obtained by multiplying the lower heating value of the fuel (MJ/Kg) by the fuel mass flow rate (Kg/s). It is also noted that the primary and secondary air mass flow rates were increased accordingly to the increase in the fuel mass flow rate (see Table 3). Figure 12 shows the temperature contours of methane and syngas 3 with low and high fuel mass flow rate (same power of 50.1 KW). The results with the high fuel mass flow rate for syngas 3 or the same power show the same effects (lower gas temperature and shorter flame) as the results obtained with low fuel mass flow rate. The average masses of CO₂ and NO at the exit of the can combustor per unit of energy generation obtained with the same power were calculated, and the results are presented in Figures 13 and 14. With the same power, Figure 13 shows a reduction of the mass of CO₂ per unit energy generation of about 10.7% to 12.5% for Schwarze Pumpe syngas fuel compared to methane fuel but higher value of CO₂ for Tampa syngas fuel. Figure 14 shows a reduction of the mass of NO per unit energy generated for syngas fuel compared to methane fuel only when the primary and secondary air mass flow rates were increased according to the increase in the fuel mass flow rate (Schwarze Pumpe Syngas 1b and Tampa Syngas 3b) a shown in Table 3.

Table 3:

Fuel and air injection conditions—same power generated.

Constituents	Syngas 1a Schwarze pumpe	Syngas 1b Schwarze pumpe	Syngas 3a Tampa	Syngas 3b Tampa	Methane
Fuel mass flow rate (Kg/s)	0.0018	0.0018	0.0039	0.0039	0.001
Velocity of primary air (m/s)	10	18	10	39	10
Velocity of secondary air (m/s)	6	10.8	6	23.4	6
Total mass flow rate at the exit (Kg/s)	0.079	0.141	0.081	0.305	0.078
Lower Heating value MJ/Kg	27.8	27.8	12.7	12.7	50.1
Power KW	50.1	50.1	50.1	50.1	50.1

Syngas temperature contours: effect of fuel mass flow rate.

Figure 13

Average mass of carbon dioxide at the exit per unit of energy generation—same power 50.1 KW for all the fuels tested.

Figure 14

Average mass of NO at the exit per unit of energy generation—same power 50.1 KW for all the fuels tested.

5. Conclusions

Three-dimensional CFD analysis of syngas fuel combustion in gas turbine can combustor is presented in this study. Five Syngas fuel mixtures with different fuel compositions (H₂/CO = 0.63 to 2.63) and low heating values (8224 KJ/m³ to 12492 KJ/m³) were tested in this study. The syngas fuels are produced by different gasification processes using different feed stocks (coal, biomass, waste). The k-ε model was used for turbulence modeling, mixture fractions/PDF model for non premixed gas combustion and P-1 for radiation modeling. The effect of the syngas fuel composition (H₂, CO, CO₂, CH₄, N₂, H₂O) and fuel heating values on syngas flame shape, gas temperature, carbon dioxide (CO₂), and nitrogen oxides ( ${\text{NO}}_x$ ) emissions was determined in this study

• Baseline fuel (methane) combustion: the results of the gas temperature, velocity field, swirling strength and CO₂ and ${\text{NO}}_x$ emissions show that gas turbine can combustor burns the fuel efficiently, reduces the emissions, and, lower the wall temperature. The predicted maximum temperature of methane fuel combustion compares well with the theoretical adiabatic flame temperature.

• The gas temperature for the all five syngas shows a lower gas temperature compared to the temperature of methane. The gas temperature reduction depends on the lower heating value and the combustible constituents (hydrogen, carbon monoxide, and methane) and non combustibles (inert) constituents in the syngas fuel.

• The results show a reduction (30% to 49%) of CO₂ mass fraction at the exit of the can combustor when methane is replaced with syngas fuel that produces less power (20% to 55% power reduction with syngas fuel). The reduction of CO₂ concentrations depends on the carbon monoxide and methane volume fractions in syngas fuel. For the same power generated by the methane and syngas fuels combustion, the results show a higher mass of CO₂ emitted per unit of energy generation (Kg/KJ) for Exxon Singapore, Tampa, and PSI syngas fuels compared to methane fuel. Schwarze Pumpe was the only syngas fuel that shows a reduction of about 12% of the average mass of CO₂ emitted per unit of energy generation (Kg/KJ) compared to methane fuel.

• With the same fuel mass flow rate but less power generated for the syngas fuel, the results show a reduction (11.5% to 97.6%) of the average NO mass fraction for synags fuel compared to methane. Syngas is used in non premixed combustion to control the ${\text{NO}}_x$ emissions by diluting the synags gas with nitrogen, carbon dioxide, and steam. The diluents reduce the flame temperature and consequently the formation of ${\text{NO}}_x$ . For the same power generated by the methane and syngas fuels combustion, the results show a reduction of the mass of NO per unit energy generated for syngas fuel compared to methane fuel only when the primary and secondary air mass flow rates were increased according to the increase in the fuel mass flow rate to keep the same power.

The results obtained in this study show the change in gas turbine can combustor performance (temperature, flame shape, CO₂ emissions, and NO emissions) when natural gas or methane fuel is replaced by syngas fuels with lower heating value and fuel compositions.