Experimental and Numerical Investigation of the Laminar Burning

Nov 3, 2017 - Numerical calculations of the flame structure, adiabatic flame temperature (Tad), species concentrations, and sensitivity analysis are a...
0 downloads 15 Views 2MB Size
Subscriber access provided by READING UNIV

Article

Experimental and Numerical Investigation of Laminar Burning Velocity and Combustion Characteristics of Biogas at High Pressures Mohammad Hossein Askari, Mehdi Ashjaee, and Sadrollah Karaminejad Energy Fuels, Just Accepted Manuscript • Publication Date (Web): 03 Nov 2017 Downloaded from http://pubs.acs.org on November 4, 2017

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Energy & Fuels is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 33

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

Experimental and Numerical Investigation of Laminar Burning Velocity and Combustion Characteristics of Biogas at High Pressures Mohammad Hossein Askari∗, Mehdi Ashjaee, Sadrollah Karaminejad Mechanical Engineering Department, Tehran University, Tehran, Iran

Abstract

Continuous variation in the composition of gaseous fuels derived from biomass is a challenge in designing efficient combustors for utilizing them. In this study, experimental measurement of laminar burning velocity (ul) of three different compositions of biogas fuel, containing equimolar H2/CO mixtures and N2 ranging from 40% to 60% by volume is conducted. Numerical calculation of flame structure, adiabatic flame temperature (Tad), species concentrations and sensitivity analysis are also performed. Investigations are conducted over a practical range of ∗

Corresponding author. Present address: Mechanical Engineering Department, Tehran

University, N.Kargar, Tehran, Iran. Tel.: +98 9112430179. E-mail address: [email protected] 1

ACS Paragon Plus Environment

Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 33

equivalence ratios (ranging from 0.4 to 1.2), and at high pressures up to 4 bar. The experimental method of schlieren in a high-pressure combustion chamber is utilized for flame speed measurement. Numerical calculations are performed using the premixed code of the Chemkin by utilizing four well-known reaction mechanisms. Laminar burning velocities calculated using USCII mechanism showed the best agreement with the experiments. The results indicated that the mole fraction of H radical increases by equivalence ratio at the whole range considered in this study, while OH radical declares its maximum concentration at the stoichiometric condition. This cause the maximum value of ul occurs at the equivalence ratio of 1.2. Tad increases by increasing pressure, especially near stoichiometric condition and for lower N2 containing fuels. The equivalence ratio of the maximum flame temperature changes from rich state (at φ=1.05) to the stoichiometric state by increasing the N2 content of fuel from 40% to 60%. H2 plays a dominant role in the combustion of biogas fuel at high H2 concentration condition. More than fifty percent of hydrogen burns before the flame front, while CO mainly burns after this position.

Keywords: Laminar Burning Velocity; Biogas; Combustion; Flam Structure

Nomenclature: A

Flame front surface area

Sb

Flame speed

Ar

Pre-exponential factor of reaction

ܵ௨

Stretched laminar burning velocity

b

Temperature exponent

Tad

Adiabatic flame temperature

Cp

Specific heat capacity

Tu

Unburned temperature

Ea

Activation energy

ul

Laminar burning velocity

Lb

Markstein length

ρu

Unburned density

Pu

Unburned pressure

κ

Flame stretch rate

R

Radius of burned gas

ߪ

Gas expansion ratio

2

ACS Paragon Plus Environment

Page 3 of 33

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

1. Introduction The composition of biogas fuel which differs reliant on the manufacturing process and feedstock, mainly comprises of H2, CO, N2 and CO2 [1, 2]. One significant challenge for efficient utilization of biogas fuel is its composition variability which causes variation in its burning characteristics. Therefore, it is vital to determine the precise values of fundamental combustion characteristics of these fuels at different operating conditions, especially at high pressures. Because many industrial combustors such as gas turbines, spark ignited, and dual fuel engines operate at high pressures. One of the most important characteristics of a fuel and oxidizer mixture is laminar burning velocity (ul). ul depends only on the reactant composition, temperature, and pressure, and therefore is a fundamental property of the fuel [3]. Some studies have investigated the ul of different H2/CO mixtures. Fu et al. [4] applied the OH-PLIF and spectrometric techniques to obtain ul values of H2/CO mixtures in a premixed Bunsen flame. They conducted experiments using different compositions of fuel over a H2/CO ratios ranging from 0.25 to 4 and equivalence ratios ranging from 0.5 to 1.2. Their results showed that increasing H2 fraction of fuel increases ul remarkably. Li et al. [5, 6] experimentally studied ul of lean and stoichiometric H2/CO/air mixtures with H2 fraction ranging from 0.3 to 0.7 at pressures ranging from 0.5 to 4 bar. Their results indicated that ul values of H2/CO premixed flames increase with increasing H2 fraction and equivalence ratio. The values of ul show reduction by increasing initial pressure for a specific equivalence ratio and hydrogen fraction. Xie et al. [7] using spherical outwardly expanding flames investigated ul of CO/H2/air mixtures at high pressures ranging from 4 to 6 bar. They performed their experiments with H2 fraction in the fuel changing from 0.2 to 0.8 and the equivalence ratios ranging from 0.5 to 1.0. Their results indicated that critical radius after which flame front becomes unstable decreases with increasing 3

ACS Paragon Plus Environment

Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

H2 content of the fuel. For fuel mixtures with higher H2 content, the critical radius increases with the increase in the equivalence ratio. There are also some works performed to investigate the effect of dilutions on flame characteristics of H2/CO fuels. Natarajan et al. [8] measured ul of H2/CO/CO2 fuel mixtures containing up to 40% CO2 by volume. The experiments were performed using a lean fuel/air mixture at initial temperatures up to 700 K and pressures of 1 and 5 bar. They measured ul using two methods. One using the conventional method of the Bunsen flame and another based on velocity profile of a stagnation flame. These data were compared with numerical simulations based on GRI3.0 and Davis mechanisms. Their results indicated that numerical simulation tends to overpredict the temperature dependence of ul for medium and high H2 content fuels but underpredict it for small H2 content fuels. Vu et al. [9] using schlieren method of constant volume combustion chamber measured ul of three compositions of biogas flames mostly containing CO/H2/CO2/CH4 and some traces of N2. They performed their experiments at pressures up to 3 bar and equivalence ratios ranging from 0.6 to 2. Their results declared that the experimental measurements and predictions using GRI 3.0 mechanism agree well at lean and stoichiometric flames, but they diverge under rich condition. Askari et al. [10] investigated the flame structure, and ul of H2/CO/O2/He premixed flames in a constant volume chamber using a Z-shape schlieren system. They compared the effect of helium and nitrogen addition on ul. Their results indicated that helium leads to a more stable and smooth flame. Wang et al. [11] investigated the effect of H2 and CO2 mole fractions on ul values of CO/H2/CO2/O2 mixtures by using the outwardly spherical propagating flame method. They performed their experiments at 300K under atmospheric and 2 bar pressure. Their results showed that the ul rises by increasing of H2 and reduction of CO2 mole fractions. Hu et al. [12] studied the effect of composition on ul values of four typical biogases used in the efficient gas turbine using a 4

ACS Paragon Plus Environment

Page 4 of 33

Page 5 of 33

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

constant volume chamber and Chemkin package. They carried out their experiments at lean burn condition with equivalence ratios ranging from 0.4 to 1.0. There are also few works conducted to investigate the ul of biogas derived from wood or the effect of high content of N2 on H2/CO fuels. Ouimette and Seers [13] numerically using PREMIX code calculated the ul values of a biogas mixture derived from wood and compared it with the ul values of methane. Their biogas fuel contained 11% H2, 14% CO, 14% CO2 and the rest of it was mainly N2. Their results indicated that low heating value of biogas leads to a slower ul than methane, especially around stoichiometry. Monteiro et al. [14] using constant volume combustion chamber investigated ul and Markstein numbers of three biogas compositions. They performed their experiments at normal temperature and pressure at equivalence ratios ranging from 0.6 to 1.2. Their fuel mixture contained almost 50% N2 and the rest of it mainly CO, H2 and CO2. The results showed that the maximum value of ul is obtained at the stoichiometric equivalence ratio. Burbano et al. [15] investigated the effect of N2 and CO2 dilutions on the ul of H2/CO/air mixtures. The volume fractions of N2 and CO2 were varied from zero to 0.6 and 0.2 respectively. They used contoured slot burner by angle method of schlieren images and performed their experiments at atmospheric condition. The results indicated that increase in the N2 and CO2 dilution fractions considerably reduces the ul. Prathap et al. [16] investigated the effect of N2 dilution ranging from 0% to 60% on the ul of an equimolar H2/CO mixture. They used the shadowgraphy technique with a high-speed imaging camera in a cylindrical combustion chamber to measure ul and Markstein length. They conducted their experiments at atmospheric condition and equivalence ratios ranging from 0.6 to 3.5. The results indicated that dilution with nitrogen decreases ul, increases the flame temperature, and shifts occurrence of peak ul from equivalence ratio of 2.0 under 0% N2 to equivalence ratio of 1.4 under 60% N2 condition. Yepes 5

ACS Paragon Plus Environment

Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

and Amell [17] numerically and experimentally studied the ul of a mixture of 20% H2, 20% CO and 60% N2 using air enriched with oxygen as the oxidizer. The flames were generated utilizing slot burners at ambient conditions (0.83 bar and 298 K). Angle method of schlieren images was used to determine the ul values. Their results indicated that ul values and the flammability limits increase with increasing the concentration of the oxygen. Although several studies are performed to investigate the combustion characteristics of H2 and CO and their combination with some dilutions, few studies have been conducted on combustion of biogas derived from wood or the effect of high concentration of N2 on H2/CO combustion. On the other hand, most of these studies are performed under atmospheric pressure or using numerical simulations. Since many industrial combustors work based on premixed flame propagation at high pressures, understanding the effect of pressure on combustion characteristics of the desired fuel is essential. However, insufficient data is available to understand the combustion characteristics of these compositions of biogas at high pressures. The discrepancy of ul values reported on literature is another necessity for further investigation of biogas combustion. As a result, this study is focused on the evaluation of the ul values and combustion characteristics of biogas/air flames over a wide range of unburned mixture compositions and initial pressures up to 4 bar. Equimolar H2/CO with N2 volumetric fraction ranging from 0.4 to 0.6 is utilized to simulate biogas fuel. The experimental method of schlieren in a high-pressure combustion chamber is employed for ul measurement. Numerical simulation and sensitivity analysis using detail chemical kinetic mechanisms are also performed for a better understanding of combustion characteristics of the fuel. 2. Experimental Procedure 2.1 Experimental Setup 6

ACS Paragon Plus Environment

Page 6 of 33

Page 7 of 33

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

The experiments are carried out using high-speed schlieren imaging in an 11 L steel cylindrical chamber with an optical access of 80 mm. The combustion chamber and primary equipment of the experimental setup are shown in Figure 1. Three cylinders, each containing one component of the biogas fuel (H2, CO, and N2) are used to simulate different compositions of fuel. The purity of all gases is above 99.99%. All fuel components and air are combined in a high pressure 50L mixing chamber to make a homogeneous fuel/air mixture. The volume fraction of each gas in the mixing chamber is adjusted using the partial pressure method. The homogeneous fuel/air mixture is charged to the combustion chamber through the inlet valve. Before starting the charging process, the combustion chamber is evacuated using the vacuum pump. Before each test, the mixture inside the combustion chamber is left for almost 15 minutes to allow the gas to come to rest. After running each experiment, the combustion chamber is air washed several times to remove soot and other combustion products and to cool down the inner walls of the combustion chamber. A high-speed camera at a rate of 5000 frames per second with a resolution of 960×540 was used for recording the flame front propagation images. Three different compositions of fuel utilized in this study which adequately represents the gas produced during the gasification of biomass [1, 2] are shown in Table 1. The N2 content of fuel has changed from 40% to 60% of the final fuel composition. More details of the experimental equipment and schlieren system is explained by Askari and Ashjaee [18]. Table 1. Different fuel mixtures and operating conditions

F1 F2 F3

H2

CO

N2

Pressure (bar)

30% 25% 20%

30% 25% 20%

40% 50% 60%

1, 2, 3, 4 1, 2, 3, 4 1, 2, 3, 4

Equivalence ratio 0.4-1.2 0.4-1.2 0.4-1.2

7

ACS Paragon Plus Environment

Temperature (K) 300±3 300±3 300±3

Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

1. N2 cylinder 2. CO cylinder 3. H2 cylinder 4. gas regulator

5. flashback 6. ball valve 7. high pressure mixing chamber 8. air compressor

Page 8 of 33

9. water trap 10. high-pressure combustion chamber 11. vacuum pump

Figure 1. Experimental setup 2.2 Data Redaction Method The first step for determining ul values from flam propagation images in a combustion chamber is calculating flame front radius variation with time. For calculating time dependency of the flame front radius, a circle should be fitted on the flame front at each time step. At the earlier period of flame propagation, when the flame radius is small, the flame front is disturbed by the spark. It also is disturbed by the inner walls of the combustion chamber when the flame radius is large. Thus, only flame radii ranging from 8 mm to 35 mm (approximately 31% of the chamber radius) are considered in order to eliminate the effect of spark and inner walls of the combustion chamber. On the other hand, the cellularity effects are also considered for choosing appropriate data for calculating burning velocity, and the data reduction is limited to un-cellular flame propagation. After determination of flame front radius variation versus time, in order to obtain the ul values, it is essential to use modification to eradicate the effect of the stretch. There are 8

ACS Paragon Plus Environment

Page 9 of 33

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

linear and nonlinear equations for calculating unstretched laminar flame speed. In this study in order to obtain more precise results, the non-linear equation is used to calculate the unstretched flame propagation speed. At linear method that is based on asymptotic studies [19-21], the flame speed stretch correction is obtained by equation (1):

S b = S b0 − σ L b .κ

(1)

At nonlinear method proposed by Kelly and Law [22] this correction is obtained by equation (2): 2

 Sb   Sb  Lbκ  0  ln  0  = − 0 Sb  Sb   Sb 

(2)

where ܵ௕ is the stretched laminar flame speed, ܵ௕଴ is the unstretched laminar flame speed, Lb is the Markstein length. The unstretched laminar flame speed is correlated to ul, as,

ul =

(3)

Sb0

σ

where ߪ is the gas expansion ratio and is defined as, ߪ=

ߩ௨ ߩ௕

(4)

In this equation, ߩ௨ is unburned gas density calculated at the pressure and temperature of the combustion chamber just before spark, and ߩ௕ is burned gas density calculated by simulating the freely propagating flame using CHEMKIN package. A comparison between obtained results using linear and nonlinear approaches at 3 bar and a fuel mixture containing 20% H2, 20% CO, and 60% N2 is shown in Figure 2. As is indicated in this figure, differences between linear and nonlinear results are considerable and linear method tends to overestimate burning velocity. During the flame propagation, the formation of cellular structures is inhibited by the strong stretch of a small flame radius. However, the suppressions 9

ACS Paragon Plus Environment

Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 33

decline with the expanding flame, and then the cellular structures appear on the flame surface, and the flame speed begins to accelerate rapidly with decreasing stretch. After development of the flame surface into cellular structure, the data can't be used to determinate the ul. This point is called the critical flame radius (Rcr). For the condition shown in Figure 2, the critical radius is 20 mm (Rcr = 20 mm).

Figure 2. Stretched laminar flame speed vs. stretch rate using linear and nonlinear approaches at 3 bar and a fuel containing 20% H2, 20% CO, and 60% N2 3.2 Uncertainty Analysis In this section, the process used for uncertainty evaluation of the measured ul is described. The approach delineated by Moffat [23] is utilized in the analysis. The total uncertainty can be evaluated by adding the systematic uncertainty and the repeatability error as follows, U tot = B

2 sys

t  +  M −1,95σ   M 

2

(5)

where Utot is the total uncertainty in measured burning velocity. The first term in the radical (Bsys) is the total systematic or bias uncertainty and the second term is the repeatability error 10

ACS Paragon Plus Environment

Page 11 of 33

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

which is obtained by repeating a test numerous times and calculating the resulting deviation. In this equation, M is the number of repeated experiments, σ is the standard deviation among the repeated tests, and tM-1, 95 is the value of student distribution at 95% confidence which is reported in the related tables, and (M – 1) is degrees of freedom. The number of repeated experiments exceeds five times. The systematic uncertainty is calculated by deriving appropriate correlations between the ul and different variables xi (such as pressure). By calculating the partial derivative of the ul relative to each variable and multiplying it in the errors in the measurement of corresponding variable (ui) the systematic uncertainty is obtained as,  ∂u ( x )  = ∑  l i ui  ∂xi i =1   N

Bsys

2

(6)

where ul(xi) is the correlation between the ul and the different variables. In this study, total uncertainties in ul is changed from 1.6% to 5.4%, depending typically on the fuel composition, equivalence ratio, and initial pressure. 3. Numerical Simulation 3.1 Computational Method In this work the 1-dimentional spherically expanding flame is modeled utilizing CHEMKIN PREMIX code [24]. This code by combining Newton iteration scheme and time integrating method finds the solutions of governing conservation equations [25]. The solver uses an automatic coarse-to-fine mesh refinement based on predefined criteria to enhance the convergence speed of the solution by providing optimal mesh settlement. In this study results are obtained using 6 continuations with the gradient and curvature parameters set to 0.004 and 0.008 respectively. These parameters led to more than 2500 grid points. Adiabatic flame temperature is also obtained by means of EQUI codes of CHEMKIN package. Four well-known chemical 11

ACS Paragon Plus Environment

Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 33

kinetic mechanisms of GRI3.0 [26], USCII [27], Li [28] and Davis [29] are used for detail simulation of combustion phenomena. 3.2 Sensitivity Analysis The sensitivity analysis method is utilized to investigate the effect of reaction rates on fuel burning rate. On the other hand, the sensitivity of the burning rate of fuel which is a representative of its burning velocity, respect to a slight change in reaction rates is obtained using sensitivity analysis. The locally normalized sensitivity coefficient is defined as [30],

 k j ∂ci   ∂ ln ci  S = =  c ∂k   ∂ ln k  i j j    

(7)

These coefficients form the normalized sensitivity matrix which represents the fractional change in concentration ci caused by a fractional change of the parameter kj. The sensitivity matrix accounts for the change of a single variable as a result of the change of individual parameters. In this equation kj which is reaction rate is taken as the input parameters and ci is the fuel component’s concentration. 4. Validation For validation of experimental setup and data reduction method, results are compared with the experimental data available in the literature [15-17] and is presented in Figure 3. Study is carried out at atmospheric condition (1bar and 300K) for two fuel mixtures containing equimolar H2/CO, and N2 content of 40% and 60% at equivalence ratios ranging from 0.6 to 1.2. The Figure 3 illustrates that measured results in this study show good agreement with all literature data. For fuel containing 40% N2 the present results show better agreement with the data reported by Burbano et al. [15]. While for fuel containing 60% N2 data reported by Prathap et al. [16] show best agreement with present study. The slight differences between these data can be attributed to 12

ACS Paragon Plus Environment

Page 13 of 33

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

the differences in pressure and temperature conditions. In comparison with the data from Yepes et al.[17], the values of the burning velocity are higher because they performed their experiments at lower operating pressure (0.85 bar).

Figure 3. Comparison between obtained experimental results and literature [15-17] at 1 bar and 300 K 5. Results and Discussion Experimental and numerical results are presented in this section. At first, the effect of equivalence ratio on ul and flame structure is investigated using the experimental method and detailed chemical kinetic simulation. Verification of numerical simulation with experimental data is also illustrated in this section. After that, the effect of fuel composition and pressure are presented. Finally, Tad and expansion ratio at different operating conditions and fuel compositions are investigated. 5.1 Effect of Equivalence Ratio The variations of ul with equivalence ratio for F1 (X_N2=0.4) and F3 (X_N2=0.6) fuels at different pressures (2 bar and 4 bar) are illustrated in Figure 4(a) and (b). By increasing the 13

ACS Paragon Plus Environment

Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 33

equivalence ratio from 0.6 to 1.2, ul values increase 44% at 2 bar and 47% at 4 bar. Increasing in the ul values at rich condition is in contrast with the most common hydrocarbon fuels where their ul is shown to be maximized at equivalence ratio of approximately 1.05 [31]. Enhancement of the ul values by rising equivalence ratios in the range of 0.6 to 1.0 can be contributed to increasing of Tab. While, at rich condition (equivalence ratio in the range of 1.0 to 1.2) it is mainly due to chemical nature of hydrogen and carbon monoxide, for which the maximum reaction rates are achieved at rich fuel/air condition. The ul value of hydrogen is shown to be maximized at rich condition around equivalence ratio of 1.7 [32, 33] and that of pure carbon monoxide with very small amounts of H2 or water vapor is maximized at φ = 2.6 [34]. It should be noticed that CO does not react with dry air or pure oxygen in the absence of hydrogen containing additives. It is difficult to ignite and sustain a dry CO/O2 or CO/Air flame because the direct reaction between CO and O2 (CO+O2 → CO2+O) has a high activation energy (48 kcal/mol) and therefore it is a prolonged process even at high temperatures. Furthermore, the O atoms produced in this reaction does not lead to any rapid chain-branching reactions[35]. On the other hand, N2 is an inert gas and doesn’t participate in the reaction directly. Therefore, hydrogen containing reactions initiate the combustion of biogas and play the dominant role in determining ul values.

14

ACS Paragon Plus Environment

Page 15 of 33

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

a

b

Figure 4. Laminar burning velocity vs equivalence ratio at (a) 2bar (b) 4bar The numerical results obtained using different chemical kinetic mechanisms are also presented in Figure 4. As is indicated in this figure, numerical results show good agreement with the experiments. For equivalence ratios ranging from 0.6 to 1.0 the results obtained using USCII chemical kinetic mechanism indicate better consistency with the experiments compared to other mechanisms (Davis, Li, and GRI3.0). While for equivalence ratios within the range of 1.0