Effects of CO2 Dilution on Methane Ignition in Moderate or Intense Low-oxygen Dilution (MILD) Combustion: A Numerical Study*

CAO Zhenjun (曹甄俊) and ZHU Tong (朱彤)**

1 College of Mechanical Engineering, Tongji University, Shanghai 200092, China

2 Shanghai DFYH Tech Services Co., Ltd., Shanghai 201114, China

1 INTRODUCTION

Due to the increasing concern about sustainability and climate change, the improvement of energy utilization efficiency to reduce fossil fuel consumption and pollutant emission is an important issue in the field of combustion science. In recent years, combustion of fossil fuels under the prevailing conditions of high temperature and low oxygen concentrations has been proved to be superior to conventional combustion technologies. For instance, the reaction zone is spread over a larger volume, leading to the lower flame peak temperature and reduced CO, CO2and NOXamount,as well as a higher thermal efficiency [1-4]. This technology is referred to as moderate or intense low-oxygen dilution (MILD) combustion technology [5], flameless oxidation (FLOX) [6], or high temperature air combustion (HiTAC) [7].

To minimize pollutant emissions, exhaust gas recirculation (EGR) is preferred in MILD combustion.An intrinsic problem with MILD combustion system,especially for premixed configurations, is the potential for spontaneous ignition of the fuel-air mixtures prior to reaching the primary combustion zone. Such an occurrence would result in flashback and flame stabilization at the injector, leading to off-design performance and, in the extreme, physical damage to combustor hardware [8]. Furthermore, in many cases it seems that strong EGR would increase the possibility of extinction in combustor. Consequently, understanding the effects of EGR dilution on autoignition in homogeneous mixture is not only a topic of fundamental importance, but also crucial for the development of the new combustion model with the capabilities to predict the interactions between turbulence and slow chemical reactions in MILD combustion.

As a promising gas fuel, methane has been widely used in industrial furnaces based on the economical and environmental considerations. Methane fuel autoignition, as a fundamental problem in combustion science, has been extensively investigated for decades.Spadaccini and Colket [8] studied the ignition delay times for mixtures of typical natural gas at the equivalence ratio of 0.45-1.25 for temperatures from 1300 to 2000 K and pressures from 0.6 to 1.5 MPa. Holton et al. [9] investigated the autoignition delay times for natural gas in a reactor at the atmospheric pressure and the equivalence ratio ranging between 0.5 and 1.25, and temperature from 930 K to 1140 K. Petersen et al. [10] performed an experimental shock-tube study and chemical kinetic modeling of methane/propane/air ignition delay times over a wide range of C3H8fraction (10%-40%), stoichiometry (0.5-3.0), temperature(1042-1615 K), and pressure (0.6-2.8 MPa). Later,Healy et al. [11, 12] provided considerably wider range of methane/ethane/propane mixtures ignition times in the temperature range of 740-1580 K and pressures 1-4 MPa. Recently, it has been found that several additives can remarkably decrease the autoignition delay time of natural gas at homogeneous charge compression ignition engine [13, 14]. However, few studies were reported on the measurements of ignition delay time in methane/air flames in MILD combustion. Moreover, to the best of authors’ knowledge, the chemical kinetic effect of EGR on methane oxidation process has not been well understood.

Hence, special concerns would be required to address the chemical effects of EGR dilution in MILD combustion conditions. Although N2is the major component of EGR gases, the thermo-physical properties such as heat capacity and diffusion coefficient of N2are similar to those of air, but significantly different from those of CO2. Furthermore, CO2concentration in exhaust gases is much higher in MILD combustion system compared with traditional combustion technologies. In this framework, we investigate the effects of CO2dilution on methane/air autoignition characteristics in MILD combustion regime with detailed chemical reaction mechanism. In this paper, the definition of ignition delay time and research methodologies are presented, the effects of equivalence ratio,temperature, dilution ratio on the system ignition are examined, and an empirical expression in form of Arrhenius-type is developed to capture the variation of ignition delay times over the range investigated.

2 COMPUTATIONAL SPECIFICATIONS

2.1 Definition of ignition delay time

Autoignition is a spontaneous process whereby a combustible mixture undergoes a chemical reaction leading to ignition and combustion in the absence of an external ignition such as a flame or spark. An essential parameter associated with autoignition is the ignition delay time, which is the period of time between the creation of the combustible mixtures and ignition. From a fundamental perspective, the ignition delay time (or, in other word, the induction time for ignition) is an important parameter for the validation and optimization of chemical kinetic mechanisms.

Several definitions of ignition delay times were reported in the literature. For example, in the SENKIN code [15], a default definition of ignition delay time for homogeneous mixture is the moment when the temperature exceeds the initial temperature by 400 K.In the study by Donato and Petersen [16], the time of occurrence of the peak concentration of OH or CH was applied as the definition. When ignition improvers are introduced, as reported by Golovitchevet al.[17], the definition of conventional ignition delay time using the maximum H or OH concentration may not be valid. In this paper, following the studies of Golovitchev and co-workers [17, 18], the ignition delay time is defined as the time for reaching the maximum rate of temperature rise (the inflection point). An example of temperature history to define the ignition delay time is presented in Fig. 1.

2.2 Model description

Figure 1 Temperature temporal profile used to define the ignition delay time (CH4/air mixture, T=1200 K, P=0.1 MPa,φ=0.6)

It is generally accepted that autoignition characteristics are primarily controlled by chemical kinetics[19]. In this regard, numerical modeling of the autoignition delay times was performed using the Sandia SENKIN code [15] in conjunction with CHEMKIN [20] software package (assuming adiabatic, constant pressure conditions). The GRI-Mech 3.0 [21] which provides the most comprehensive and standardized set of mechanisms for methane combustion was employed to calculate the ignition delay time. It consists of 53 species and 325 elementary reaction steps.

SENKIN is a Fortran computer program that predicts the time-dependent chemical kinetics behavior of a homogeneous gas mixture in a closed system. The basic governing equations are described as follows:

Conversation of species:

whereYk=mk/mis the mass fraction of thekth species,Wkthe relative molecular mass of thekth species,v=V/mthe specific volume,mthe total mass of the mixture,kω˙ the net chemical production rate of speciesk, andVthe volume of the system.

Conversion of energy:

wherehkis the specific enthalpy of thekth species,and the mean specific heat of the mixture is

The net chemical production ratekω˙ of each species results from a competition between all the chemical reactions involving that species. Each reaction proceeds according to the law of mass action and the forward rate coefficients are in the modified Arrhenius form:

whereEis the activation energy,βthe temperature exponent,Athe pre-exponential constant, andRthe universal gas constant.

3 RESULTS AND DISCUSSION

3.1 Characteristics of ignition delay time

Figure 2 shows multiple temperature profiles as a function of the residence time in the reactor for CH4/air/CO2mixtures at different dilution ratios. The dilution ratio,X, defines as the mole fraction of CO2in the mixture. In general, the predicted temperature profiles kept the similar tend qualitatively. That is after a short time interval when the system temperature slowly increases, it abruptly jumps to its final chemical equilibrium state. The final temperature decreases with the increment of dilution ratio. The different shapes of the temporal temperature profiles indicate different evolution of the oxidation process with various dilution fractions. It shows that the influence of CO2dilution on oxidation is obvious. It is also seen that the ignition delay time is prolonged with the increasing of dilution ratio. This suggests that CO2addition delays the onset of the CH4oxidation process.

Figure 2 Temperature profiles in the process of ignition(T=1300 K, P=0.1 MPa, φ=1.0)X: 0; 0.1; 0.2; 0.3

In a MILD combustion system, generally, a pair of regenerative burners, operating alternatively, is applied as a unit. The flame can be shifted from one burner to the other on the principle of short-term heat storage and release of the regenerators. Depending on burner and regenerator design, the switching time,normally between 30 and 60 s, is adopted to reduce the heat loss from exhaust gases. In fact, due to the dynamic operation carried out by the switching devices, it is difficult to identify flame properties with an acceptable high accuracy. A situation more relevant to MILD combustion is the periodically ignition and extinction. With respect to these phenomena, it is noteworthy that high dilution ratio would lead to relatively long autoignition delay time and even make stable combustion more difficult to be achieved, especially at short switching times.

Figure 3 Ignition delay times for CH4/air/CO2 mixtures for temperature 1000-1200 K under stoichiometric condition 1—X=0; 2—X=0.1; 3—X=0.2; 4—X=0.3; 5—X=0.4;6—X=0.5; 7—X=0.6

In view of special concern of safe combustion of methane, it is interesting to investigate the effects of preheated temperature and CO2dilution on methane ignition. As illustrated in Fig. 3, ignition delay time increases rapidly at decreasing initial temperature.Specifically, when a low-mediate temperature and a high dilution ratio were adopted, relatively long ignition delay times (sometimes more than 1 s) are required for ignition of mixtures. For instance, the ignition delay time is ～1.10 s at the condition of 1000 K whenX=0.However, the ignition delay time increases drastically to ～3.29 s at the condition of 1000 K whenX=0.6. In particular, relative long ignition delay time would increase the possibility of explosion or extinction in combustor. Consequently, it is essential to properly design preheated temperature and dilution ratio in a MILD combustion system for fire safety.

The C H bond dissociation energy of methane(～439 kJ·mol-1) probably is the highest among all hydrocarbons. It is responsible for low chemical reactivities. Thus, in the following discussions, attention is only given to the condition of MILD combustion, particularly at temperature between 1200 and 1600 K under atmospheric conditions.

Figure 4 Effect of equivalence ratio on the ignition delay of CH4/air/CO2 mixtures (T=1300 K, P=0.1 MPa)◇ X=0; ■ X=0.1; ○ X=0.2; ◆ X=0.3; △ X=0.4; ● X=0.5;□ X=0.6

Figure 4 reports the predicted ignition delay times as a function of equivalence ratio over a dilution ratio range fromX=0 toX=0.6. As expected, the results indicate that at all equivalence ratios, the ignition delay time increases when the mole fraction of CO2increases. The reason is that as the dilution ratio increases, the evolution of oxidation process is relatively slower. For a given dilution degree and preheating temperature, in terms of methane oxidation, a positive correlation of ignition delay time with equivalence ratio was observed. Furthermore, Fig. 4 shows that a higher dilution ratio of CO2leads to greater slope and a substantially higher level of ignition delay time.

Figure 5 Ignition delay times for CH4/air/CO2mixtures under stoichiometric condition (P=0.1 MPa, φ=1.0)■ X=0; △ X=0.1; ● X=0.3; □ X=0.6

Figure 6 Production rate of CH4 in CH4/air/CO2 flames (T=1300 K, P=0.1 MPa, φ=1.0)total rate of production; R11; R52; R53; R98

Figure 5 depicts the Arrhenius plot of the predicted ignition delay time of CH4/air/CO2mixtures as a function of the initial temperature under stoichiometric conditions and at dilution ratios of 0, 0.1, 0.3, and 0.6,respectively. It is indicated that ignition delay time is exponentially related with the reciprocal of initial temperature. The ignition delay time is dramatically influenced by the CO2addition. Consequently, in case of constant equivalence ratio, ignition delay time mainly depends on the initial temperature and dilution ratio.Here, it shows that initial temperature and dilution ratio have opposite effects on ignition delay time.

The production rate of CH4in the flame is shown in Fig. 6. The dominant reactions contributing to CH4can be identified from the rate of production analysis as follows:

The free radicals such as H, O and OH are extremely important in the oxidation of methane. The main consumption reactions of CH4in the flame are through the attack by O, H and OH radicals to produce the methyl (CH3) radical. In Fig. 7, it is evident that H,O, and OH as the reactants rapidly increase and then decrease to some constant values after ignition. The mole fractions of H, O and OH decrease as CO2is added and this will suppress the oxidation of methane.

Figure 7 Mole fraction profiles of O, H, and OH in the process of ignition (T=1300 K, P=0.1 MPa, φ=1.0)1—X=0; 2—X=0.1; 3—X=0.2; 4—X=0.3; 5—X=0.4; 6—X=0.5; 7—X=0.6

Table 1 Selected reactions involving CO2 based on GRI-Mech 3.0 (Rate constants are expresses as k=ATβexp[-E/(RT)]with units of J, cm3, mol and s)

In more detail, an artificial species (Ω), which has the same thermodynamic, transport properties as those of the added CO2but it does not participate in any reaction, is introduced to further investigate the chemical effect of CO2on methane oxidation process.The dilution effects of CO2were evaluated by addition ofXwhich is assumed as an inert species. A list of reactions involving CO2in GRI-Mech 3.0 is given in Table 1, including their rate coefficients.

As illustrated in Fig. 8, for a given dilution degree and pre-heating temperature, the predicted ignition delay times in CO2are higher than in Ω. In particular, the predicted gap between ignition delay times of two diluents increases progressively with the dilution ratio. For instance [Fig. 8 (a)], the gap is ～120 μs for dilution ratio of 0.2, and ～2690 μs for dilution ratio of 0.6. The reason for this phenomenon is that the chemical effects of CO2addition can further slow down the methane oxidation reactions. Since the CO2addition can alter chemical reactions in flame, it is reasonable to conclude that the chemical role of CO2dilution cannot be neglected.

Figure 8 Ignition delay times over a wide range of dilute ratios (T=1300 K, P=0.1 MPa)■ CO2; ○ Ω

Figure 9 Profiles of temperature and mole fractions of CO and CO2 in the process of ignition with different diluents(Mole fraction of added CO2 or X equals to 0.3, T=1300 K,P=0.1 MPa, φ=1.0)CO2; Ω

An analysis of the temperature and major species profiles involving CO2and Ω dilution was carried out under the stoichiometric conditions. Fig. 9 reports the profiles of temperature and mole fractions of CO and CO2as a function of the residence time with different diluents. For a given dilution degree, the temperatures in CO2are higher than in Ω, as illustrated in Fig. 9 (a).For instance, the final temperatures with CO2and Ω dilution are ～2263 K and ～2053 K, respectively.Although the destruction of CO2to form CO is considered as an endothermic reaction, the oxidation of CO would make the final temperature higher in CO2dilution than in Ω due to the chemical effects of CO2in high temperature conditions. Figs. 9 (b) and 9 (c)clearly demonstrated the phenomena of CO2destruction and CO oxidation during the process of ignition.This suggests that chemical effects of CO2are important in the prediction of flame temperature. As shown in Fig. 9 (b), the final CO2mole fraction in CO2dilution is higher than in Ω dilution. Furthermore, in case of CO2dilution, the predicted CO mole fraction increases drastically and presents a maximum near the ignition point and then decreases to some constant value. However, in case of Ω dilution, CO mole fraction increases and then keeps the same value. These are important results since it demonstrates that the chemical effects of CO2have a direct influence on the converting CO to CO2and CO2decomposition.

Generally, methane combustion can be simply characterized as a two-step process: the first step involves the breakdown of methane to carbon monoxide,with the second step being the final oxidation of carbon monoxide to carbon dioxide. Thus, the oxidation of carbon monoxide is extremely important to the oxidation of methane. As illustrated in Table 1, CO2is not inert but directly participates in chemical reactionsslow and does not contribute significantly to the formation of CO2, but rather serves as the initiator of the chain sequence. The actual CO oxidation step, (R99),is also a chain-propagating step, producing H atoms.Reaction R99 has been identified as the primary pathway for the chemical effect of CO2in the literature[22]. The competition of CO2for H radical through the reverse reaction of R99 with the single most importantsignificantly reduces the concentration of O, H and OH, and plays a chemical inhibiting role that reduces the overall combustion rate [23].

According to the previous analysis, we can find that there are two main fundamental effects of CO2on ignition delay time: (1) changing the heat capacities and transport properties of the mixture, (2) the direct chemical effects of CO2. It is therefore of importance to take into account of the effects of CO2dilution in MILD combustion.

3.2 Fitting data to an overall correlation

Additional work has been conducted to derive kinetic time models in Arrhenius-type formulas. Since the mixture consists of CH4, air, and CO2, the mole fraction of CO2can be simply determined by the mole fraction of CH4and O2. As a result, the Arrhenius-type formula is a function of temperature, mole faction of O2and mole faction of CH4:

where τignis the ignition delay time or general kinetic time (s);4CHX and2OX are the mole fractions for CH4and O2in the mixture, respectively; and E, A, x and y are constants. In fact, E in this equation is commonly referred to as ignition activation energy. Take the natural logarithm of Eq. (5), which is a multi-parameter linear relation suitable for use in a multi-parameter linear curve-fitting routine.

The correlation is given by the following expression according to the 140 observation data calculated by GRI-Mech 3.0:

Comparing the correlation to the results from the mechanism for the full range of conditions specified in this study yields an r2value of 0.987, and the average relative error is about 5%.

Figure 10 compares the predictions from detailed mechanism to correlation computations for temperature ranges from 1200 K to 1600 K under stoichiometric conditions. Generally, the agreements between predictions and calculation are very good for different CO2dilution ratios. The correlation correctly reproduced the kinetics of CH4/air/CO2ignition routes.

It is worth to note that the above discussion relates to the prediction of a pure kinetic time for autoignition of CH4/air/CO2mixture. It is widely recognized that auto-ignition in practical combustors is strongly influenced by the interactions between flow field and chemical reaction. That is to say the ignition of mixture commonly comprises a series of physical and chemical processes, i.e., the time for fuel and air mixing, heating, diffusing and pre-flame reaction. Unfortunately, direct implementation of detailed chemical schemes to a three-dimensional computational fluid dynamics (CFD) analysis is far beyond the capability of present computers.

However, the predicted results of autoignition delay times only depend on the initial state of the mixture and the chemistry scheme employed, regardless of the turbulent flow. Consequently, these results are meaningful and can be applied to MILD combustion as “reference time”, for example, to predict ignition delay time in turbulent reacting flow.

4 CONCLUSIONS

The effect of CO2addition on the ignition of methane/air has been predicted under MILD combustion conditions using a zero-dimensional SENKIN code from CHEMKIN package. CO2not only significantly affects the autoignition delay time, but also influences the oxidation process and the final state of combustion system. The chemical effects of CO2were evaluated by comparison with an inert species X. This study is summarized with the following conclusions:

Figure 10 Comparison of the predicted results with correlation (P=0.1 MPa, φ=1.0)■ mechanism; correlation

(1) In general, both increment of equivalence ratio and volumetric percentage of CO2increase autoignition delay times. However, ultra-high dilution ratio will further prolong the ignition delay time and increase the possibility of explosion or extinction.

(2) The inhibition of chemical reaction with CO2is due to the decrease of O, H and OH, eventually leading to slower oxidation process of methane.

(3) An empirical correlation in form of Arrheniustype formulas was derived to calculate the ignition delay times in all of the conditions. A good match was found between the results from mechanism and the correlation all over the investigated conditions.

NOMENCLATURE

Apre-exponential constant, cm, mol, s

cpspecific heat at constant pressure, J·mol-1·K-1

Eactivation energy, J·mol-1

hspecific enthalpy, J·kg-1

kfrate constant for a reaction, depends on reaction order

mmass, kg

ppressure, Pa

Runiversal gas constant, 8314.3 J·kmol-1·K-1

Ttemperature, K

ttime, s

Vvolume, m3

Wrelative molecular mass

Ymass fraction

βtemperature exponent

τignignition delay time, s

φequivalence ratio

Subscript

kspecies

1 Yang, W., Blasiak, W., “Flame entrainments induced by a turbulent reacting jet using high-temperature and oxygen-deficient oxidizers”,Energy Fuels, 19 (4), 1473-1483 (2005).

2 Gupta, A.K., Bolz, S., Hasegawa, T., “Effect of air preheat temperature and oxygen concentration on flame structure and emission”,J.Energy ResourcesTechnol., 121 (3), 209-216 (1999).

3 Yang, W., Blasiak, W., “Numerical simulation of properties of a LPG flame with high-temperature air”,Int.J.Therm.Sci., 44 (10),973-985 (2005).

4 Yuan, J., Naruse, L., “Effects of air dilution on highly preheated air combustion in a regenerative furnace”,Energy Fuels, 13 (1), 99-104(1998).

5 Cavaliere, A., de Joannon, M., “Mild combustion”,Progr.Energy Combust.Sci., 30 (4), 329-366 (2004)

6 Wünning, J.A., Wünning, J.G., “Flameless oxidation to reduce thermal NO-formation”,Progr.Energy Combust.Sci., 23 (1), 81-94(1997).

7 Yang, W., Blasiak, W., “Numerical study of fuel temperature influence on single gas jet combustion in highly preheated and oxygen deficient air”,Energy, 30 (2-4), 385-398 (2005).

8 Spadaccini, L.J., Colket Iii, M.B., “Ignition delay characteristics of methane fuels”,Progr.Energy Combust.Sci., 20 (5), 431-460 (1994).

9 Holton, M.M., Gokulakrishnan, P., Klassen, M.S., Roby, R.J., Jackson, G.S., “Autoignition delay time measurements of methane, ethane, and propane pure fuels and methane-based fuel blends”,J.Eng.Gas Turbines Power, 132 (9), 091502 (2010).

10 Petersen, E.L., Kalitan, D.M., Simmons, S., Bourque, G., Curran,H.J., Simmie, J.M., “Methane/propane oxidation at high pressures:Experimental and detailed chemical kinetic modeling”, Proc. Combust. Inst., 31 (1), 447-454 (2007).

11 Healy, D., Curran, H.J., Simmie, J.M., Kalitan, D.M., Zinner, C.M.,Barrett, A.B., Petersen, E.L., Bourque, G., “Methane/ethane/propane mixture oxidation at high pressures and at high, intermediate and low temperatures”, Combust. Flame, 155 (3), 441-448 (2008).

12 Healy, D., Curran, H.J., Dooley, S., Simmie, J.M., Kalitan, D.M.,Petersen, E.L., Bourque, G., “Methane/propane mixture oxidation at high pressures and at high, intermediate and low temperatures”,Combust. Flame, 155 (3), 451-461 (2008).

13 Noorpoor, A.R., Ghaffarpour, M., Aghsaee, M., Hamedani, A., “Effects of fuel additives on ignition timing of methane fuelled HCCI engine”, J. Energy Inst., 82 (1), 37-42 (2009).

14 Morsy, M.H., Chung, S.H., “Effect of additives on ignition of methane at homogeneous charge compression ignition engine-like conditions”,Proc. Inst. Mech. Eng. D J. Automobile Eng., 221 (5), 605-619 (2007).

15 Lutz, A.E., Kee, R.J., Miller, J.A., “Senkin: A Fortran program for predicting homogeneous gas phase chemical kinetics with sensitivity analysis”, Sandia National Laboratories Report, SAND87-8248(1987).

16 Donato, N.S., Petersen, E.L., “Simplified correlation models for CO/H2chemical reaction times”, Int. J. Hydrogen Energy, 33 (24),7565-7579 (2008).

17 Golovitchev, V.I., Pilia, M.L., Bruno, C., “Autoignition of methane mixtures—The effect of hydrogen peroxide”, J. Propul. Power, 12(4), 699-707 (1996).

18 Golovitchev, V.I., Chomiak, J., “Evaluation of ignition improvers for methane autoignition”, Combust. Sci. Technol., 135 (1), 31-47 (1998).

19 Westbrook, C.K., “Chemical kinetics of hydrocarbon ignition in practical combustion systems”, Proc. Combust. Inst., 28 (2), 1563-1577(2004).

20 Kee, R.J., Rupley, F.M., Miller, J.A., Coltrin, M.E., Grcar, J.F.,Meeks, E., Moffat, H.K., Lutz, A.E., Dixon-Lewis, G., Smooke,M.D., Warnatz, J., Evans, G.H., Larson, R.S., Mitchell, R.E., Petzold,L.R., Reynolds, W.C., Caracotsios, M., Stewart, W.E., Glarborg, P.,Wang, C., Adigun, O., Houf, W.G., Chou, C.P., Miller, P., Ho, P.,Young, D.J., CHEMKIN Release 4.0, Reaction Design, Inc., San Diego, CA (2004).

21 Smith, G.P., Golden, D.M., Frenklach, M., Moriarty, N.W.,“GRI-Mech 3.0”, [2010-10-10], http://www.me.berkeley.edu/gri-mech/

22 Park, J., Hwang, D.J., Park, J.S., Kim, J.S., Keel, S.I., Cho, H.C.,Noh, D.S., Kim, T.K., “Hydrogen utilization as a fuel: Hydrogen-blending effects in flame structure and NO emission behaviour of CH4-air flame”, Int. J. Energy Res., 31 (5), 472-485 (2007).

23 Liu, F., Guo, H., Smallwood, G.J., “The chemical effect of CO2replacement of N2in air on the burning velocity of CH4and H2premixed flames”, Combust. Flame, 133 (4), 495-497 (2003).