Effects of CO2 and N2 Dilution on the Combustion Characteristics of H2/CO Mixture in a Turbulent, Partially Premixed Burner.

Herein, the influences of CO2 dilution, N2 dilution, and CO2/N2 (in which half of the N2 is replaced by CO2) dilution on the combustion characteristics of a turbulent, partially premixed CO/H2-air flame were experimentally investigated in terms of the flame structure, flame temperature, and CO and CO2 concentrations in flames. TDLAS (tunable diode laser absorption spectroscopy) technique and an infrared gas analyzer were used for such purposes. CO2 dilution not only increases more momentum but also reduces the reaction rate. This results in a much longer flame length than that under N2 dilution. Compared with N2 dilution, the L H (axial length of the high-temperature reaction zone) values for the same levels of CO2 dilution and CO2/N2 dilution are much longer. The highest CO concentration in the CO2 diluted flame is higher than that in the CO/H2 flame and that in the CO2/N2 diluted flame is higher than in that the N2 diluted flame. The sizes of the main chemical reaction zone in CO2 and CO2/N2 diluted flames are larger than that in the N2 diluted flame. The inflection points in the rates of variation of the flame temperature and the CO and CO2 concentrations verify that CO2 dilution creates lower intensities and lower rates of chemical reactions, compared with N2 dilution.

Introduction

With the development of industry, the consumption of fossil fuels such as coal, petroleum, and natural gas increases while the problem of the efficient utilization of energy arises.[1−3] Coal is the most important energy in China. To realize the efficient and clean utilization of coal resources, it is one of the most effective technological approaches to apply the coal gasification technology to obtain producer gas.[4,5] The main combustible components of producer gas are CO and H2 while the noncombustible components are CO2 and N2.[6,7] The producer gas in industries is usually mixed with water gas and mainly used as the fuel for heating furnaces, heat treatment furnaces, and IGCC system boilers.[8−10] Due to the sensitivity of producer gas to the variety of coal or biomass sources and the gasification process, its components are not constant, which may make great effects on the combustion characteristics of producer gas.

To get the fundamental understanding of the combustion characteristics of the H2/CO flame, some investigations are performed to measure the laminar flame speed with different CO to H2 ratios or under high pressure.[11−17] CO2 and N2 are the main dilutions for the H2/CO mixture and have great effects on the combustion characteristics. Furthermore, the addition of CO2 in the fuel simulates the EGR (exhaust gas recirculation) system which reduces the cost of CCS (carbon capture and storage) and the pollution of NO[18,19] Thus, it is essential and economical to study the effects of CO2 and N2 dilution on flames. It has been proved that CO2 dilution significantly influences the flame speed, flame stability, flame structure, and heat release characteristics of laminar H2/CO, H2/N2, CH4/O2, and MTHF flames.[20−28] The effects of CO2 dilution on the flame characteristics, flame structure, emissions, temperature, and flame dynamics in turbulent CH4/air, CO/H2/CO2, H2/O2, and CH4/H2 flames were also discussed.[29−35] On the other hand, the impacts of N2 dilution on the flame characteristics have been analyzed in some investigations and their results were mainly related to the burning velocity, flame stability, flame structure, and flame chemiluminescence in laminar CH4/air and H2/CO flames.[36−43] The addition of CO2 has effects on the flame structure in H2/CO2/N2 and H2/O2 counterflow diffusion flames.[44,45] Furthermore, the effects of the stagnation plate, flame temperature, and fuel properties on the combustion characteristics and emissions of diffusion flames were experimentally studied.[46−48] In most of the aforementioned studies, the flame structure, flame stability, and flame speed are the most important parameters for the combustion characteristics of CO2 diluted laminar or turbulent premixed flames and N2 diluted laminar premixed flames. In a variety of industrial heating applications, the flame temperature, species concentration, and flame length of nonpremixed or partly premixed flames in a turbulent industrial burner are mainly of concern.

The flame temperature and species concentration in flames are important for combustion since they characterize the combustion process. The measurements of flame temperature and species concentration in the turbulent burner of industrial furnaces are significant for monitoring the high-temperature zone and verify some reaction mechanisms under a complex combustion environment. Tunable diode laser absorption spectroscopy (TDLAS) is a novel developed real-time online detection technique providing high sensitivity, fast response, and high precision, which has been utilized for temperature and species concentration measurement in flames.[49−52] In the previous study of the present authors, a combustion diagnostic system based on TDLAS was developed and used to measure the flame temperature in turbulent partly premixed CO/CO2 and CO/CO2/H2 flames, which simulates the combustion of converter gas in an industrial burner.[53]

In the current work, rather than premixing fuel gas and air, the fuel was partially premixed with air in a turbulent industrial burner. Three dilutions including N2, CO2, and CO2/N2 (in which half of the N2 was replaced by CO2) were added to the fuel gas (CO/H2 mixture) to simulate the combustion of producer gas and water gas and investigate their impacts on the flame characteristics. To identify the effects of different dilutions, the flame structure, flame temperature, CO, and CO2 concentrations in flames were studied experimentally. To analyze the effects caused by the different dilutions, the differences in the corresponding flame length, the abrupt change in the high-temperature zone, and the inflection points in the derivatives of temperature and concentrations were discussed.

Experimental Section

Combustion Diagnostic System Based on TDLAS

In this study, the combustion diagnostic system, which is shown in Figure , is utilized to investigate the flame structure, flame temperature, and CO2 concentration in flames.

Figure 1

Schematic diagram of the combustion diagnostic system.[54] Reprinted with permission from [Liu et al.]:[Springer Nature].

The CO, H2, CO2, and N2 gases are supplied by gas cylinders from Praxair with purity of 99.9, 99.999, 99.999, and 99.999%, respectively. The oxidant gas is offered by an air compressor (Soret Gas Equipment KS-75A), while a refrigeration dryer (FLD-1D) is used as the supporting equipment to eliminate the effects of water vapor in the compressed air. The flow rates of air and fuel are controlled by mass flow controllers (MFCs) with an accuracy of ±1% S.P. and displayed on the operating screen together with the pressure.

A turbulent partly premixed industrial burner is shown in Figure . The gas and air are partly premixed in a cup and then ignited by an electric ignition device. The further mixing of gas and air and the beginning of combustion reactions occurs in the zone between the spark and nozzle.

Figure 2

Structure of the turbulent partly premixed industrial burner.[54] Reprinted with permission from [Liu et al.]:[Springer Nature]. The burner sleeve inside the combustion chamber has a length of 250 mm. The ignition place is 150 mm apart from the nozzle.

The laser diode controller, DFB laser, combiner, and splitter are used to generate two beams of lights of specific wavelengths. The detector and the BNC adapter convert the received optical signal into an electrical signal and transmit it to the data processing center on the computer, while the other BNC adapter linked to the laser diode controller synchronizes the output signal with the input signal. The integrating sphere (THORLABS IS200-4) between the laser probe and the detector is used to reduce the error caused by the nonuniform distribution of the incident light source or the beam deviation, resulting in more reliable measurement results. The internal structure of the integrating sphere is shown in Figure . In the current work, the measurement error of the flame temperature and CO2 concentration is within ±2% F.S.

Figure 3

Internal structure of the integrating sphere.[54] Reprinted with permission from [Liu et al.]:[Springer Nature]. The integrating sphere has an inner wall coated with the material with the diffuse reflection coefficient close to 1.0. Several apertures are opened as the input port of light and the reception port of the detector. The light inside the sphere is reflected many times through the inner wall, forming a light with uniform distribution.

TDLAS Theory: Two-Line Direct Absorption Spectroscopy

TDLAS is governed by the law of Beer–Lambert.[55] The fractional transmission can be expressed as eq ,where I0 and I are the incident and transmitted laser intensity, P is the static pressure, L is the gas medium length, S(T) is the line strength of the transition, which is only a function of temperature, Φ(υ) is the line shape function, and X is the concentration of the absorbing gas. The line shape function Φ(υ) represents the relative light intensity of the unit frequency interval. The area underneath the line shape function is normalized so that ∫–∞+∞Φ(υ)dυ = 1. The integrated absorbance across the absorption feature can be obtained by eq ,

Species concentration X can be obtained using eq ,

For the temperature measurement in two-line direct absorption spectroscopy, the flame temperature can be achieved by comparing the line strength of two different transitions with different temperature dependences.[56] The line strength S(T) can be expressed according to the known line strength at a reference temperature T0, which is defined by eq ,[50]where Q(T) is the molecular partition function, E″ is the lower-state energy, c is the speed of light, h is Planck’s constant, and k is Boltzmann’s constant.

The temperature can be deduced from the measurement of the integrated absorbance ratio for two different temperature-dependent transitions, and the ratio of these two integrals is given by eq ,where Pabs is the partial pressure of the absorbing species, S(T0, v) is the line strength of the transition centered at v for the reference temperature T0, and T is the temperature of the gas which can be obtained using eq ,The quantity hc/k has a numerical value of 1.438 cm·K. A1 and A2 are the integrated areas of the absorption lines.

In this work, the two absorption lines are determined based on the following rules:[57]

Both lines need sufficient absorption over the selected temperature range.

The absorption ratio should be single-valued with temperature, and the line strengths of the two lines should be similar.

The two lines should have sufficiently different lower state energies E″ to yield an absorption ratio that is sensitive to the probed temperature.

The two lines should be free of significant interference from nearby transitions.

The two lines are desired to have the same line shape function.

The two lines are close enough to be scanned by a single laser.

To select two suitable absorption lines, the plotted spectral simulation of the chosen absorption lines (P = 1 atm, X = 10%, and L = 45 mm) is shown in Figure using the HITRAN database. As H2O is one product of the CO/H2–air flame, the absorption lines of H2O are also illustrated in Figure to choose the absorption lines with the strong absorbance of CO2 and weak absorbance of H2O. To guarantee the high signal-to-noise ratio (SNR) and minimal interference from ambient water vapor, the strong absorbance of CO2 and high-temperature sensitivity are required. Considering the rules and requirements above, the spectral lines of 1996.89 and 2004.02 nm (marked in Figure ) are chosen as light beams to probe the absorption of CO2. The lower state energy (E″) for 1996.89 nm is 994.19 cm-1 and the E″ for 2004.02 nm is 106.13 cm–1.

Figure 4

Simulation of the CO2 and H2O absorption lines at different temperatures.[54] Reprinted with permission from [Liu et al.]:[Springer Nature].

Operating Conditions and Data Analysis

In this study, experiments were carried out under several different conditions, as shown in Table . High purity CO2, N2, and CO2/N2 (in which half of the N2 was replaced by CO2) were used as different diluents in CO/H2 gas. In Table , Q is the flow rate of different gases, V% represents the volume fractions of CO, H2, CO2, and N2 in their mixtures, Φ is the equivalence ratio, and D is the dilution.

Table 1

Combustion Conditions Used in This Study (Cases 1 to 13)

case	Qfuel	QCO	VCO%	QH2	VH2%	QDCO2	VCO2%	QDN2	VCO2%	Qair	Φ
	L/min	L/min	%	L/min	%	L/min	%	L/min	%	L/min
1	37.0	28	76	9	24	0	0	0	0	300	0.29
2	40.2	28	70	9	22	3.2	8	0	0	300	0.29
3	44.0	28	64	9	20	7.0	16	0	0	300	0.29
4	48.6	28	58	9	19	11.6	24	0	0	300	0.29
5	54.4	28	51	9	17	17.4	32	0	0	300	0.29
6	40.2	28	70	9	22	0	0	3.2	8	300	0.29
7	44.0	28	64	9	20	0	0	7.0	16	300	0.29
8	48.6	28	58	9	19	0	0	11.6	24	300	0.29
9	54.4	28	51	9	17	0	0	17.4	32	300	0.29
10	40.2	28	70	9	22	1.6	4	1.6	4	300	0.29
11	44.0	28	64	9	20	3.5	8	3.5	8	300	0.29
12	48.6	28	58	9	19	5.8	12	5.8	12	300	0.29
13	54.4	28	51	9	17	8.7	16	8.7	16	300	0.29

The CO concentration in the flame was sampled and measured using an infrared gas analyzer with the linearity error of ±2% F.S. (IGS-09S, Beijing Stead Automation Equipment., Ltd). The flame temperature and CO2 concentration in the flame were measured using two-line scanned-wavelength direct absorption spectroscopy. The scanning rate of the laser is 2 × 106 Hz, and the signal frequency was 100 Hz. To obtain sufficiently accurate results, 2 × 106 signal-shots were averaged for each position. Thirty positions along the axial direction at the burner centerline were selected for the temperature measurements, while 15 positions were used for CO2 concentration measurements.

Figures and 6 show the representative measurements (Plot 0) of the absorption spectra of CO2 along with the corresponding Voigt-fitting profiles (Plot 1). The CO2 absorption in this section was calculated from the difference between the CO2 absorption of the environment and the flame to reduce the influence of the surroundings. The instability of the turbulent flame led to the deviations between the measurements and the Voigt-fitting. Plot 2 shows the residuals of the two plots, which represent the normalized difference between the measured data and the Voigt-fitting profile. Higher residuals indicate that the turbulent flame fluctuated violently.

Figure 5

Measured absorption spectra of CO2 at 1996.89 nm near 1100 K. The Voigt fit to the experimental data is plotted for comparison, along with the residuals.

Figure 6

Measured absorption spectra of CO2 at 2004.02 nm near 1100 K. The Voigt fit to the experimental data is plotted for comparison, along with the residuals.

Results and Discussions

Flame Structure

The lean fuel combustion was studied by maintaining an equivalence ratio of 0.29. Then, a turbulent partially premixed burner was utilized to simulate the combustion in industrial furnaces. Finally, three diluents, CO2, N2, and CO2/N2 (in which half of the N2 was replaced by CO2) were added to the fuel gas (CO/H2 mixture) to simulate the combustion of producer gas and water gas. The flow rates of air, CO, and H2 were kept constant and the volume fractions of the diluents were raised from 8 to 32%. Images of the flames were captured using a camera. The camera was set at the side of the flame with an exposure of 1/10 s. As can be seen in Figure , overall the flames comprised a bright blue zone followed by a purple-orange zone. The bright blue zone resulted from the entrainment of the fuel gas mixed with air while the purple-orange zone indicated the existence of soot particles. As the flame diameters derived by image processing software ranged 38–40 mm, the flame diameters changed little following the addition of the diluents. Comparing the images of the CO/H2 flame and the flame with 32% dilution, this change is very obvious.

Figure 7

Variations in the lean fuel flame structure with different diluents ((a) CO2 dilution, (b) N2 dilution, and (c) CO2/N2 dilution).

Figure shows that the flame length increased with the increasing dilution. Compared with N2 dilution, the flame length under CO2 dilution was longer. The flame length under CO/N2 dilution was between the two diluents above. In this study, the flame length was defined as the distance between the nozzle and the highest point along the axial flame; this was obtained by measuring the visible length of the flame. Figure shows flame lengths derived by image processing software which are calculated by averaging the flame lengths of 20 images, and the mathematical relationship between the flame length and dilution ratio obtained by Origin software. The increasing rate of the flame length, , was denoted as , where the amount of increase in the flame length, ΔLf, was calculated as ΔLf = LD – L0. Here, LD is the flame length for each variation of the diluent, L0 is the length of the CO/H2 flame (98 mm), and VD% is the volume fraction of each variation of the diluent. ΔLf and are presented in Table . Figure and Table indicate that, compared with the dilution of N2, both of ΔLf and for the CO2 and CO2/N2 diluted flames were larger. As it was difficult to predict the mixing degree of CO/H2 and the diluent in the turbulent partially premixed burner, investigations into the flame speed in the turbulent premixed flame, (),[58] and the flame height in the turbulent jet diffusion flame, (),[59] were not sufficient to determine how the diluent affected the flame length. In this study, the flow rates of CO, H2, and air were kept constant and the amount of each diluent was equivalent; the only variable was the momentum. According to the momentum formula , the total flow rate, v, and area, S, were kept constant, whereas the density of the mixed gas varied with each different diluent. As the density of CO2 (1.98 g/L) at normal atmospheric temperature is greater than that of N2 (1.25 g/L), the increase in momentum following CO2 dilution was higher than that for dilution with an equivalent amount of N2, thus resulting in a longer flame length.

Figure 8

Flame lengths and the mathematical relationship between the flame length and dilution ratio for different diluents.

Table 2

ΔLf and for Different Diluents and Levels of Dilution

dilution	CO2				N2				CO2/N2
volume fraction (%)	8	16	24	32	8	16	24	32	8	16	24	32
ΔLf (mm)	9	22	38	57	4	12	23	36	6	16	29	45
	1.12	1.37	1.58	1.78	0.50	0.75	0.96	1.12	0.75	1.00	1.21	1.41

Unlike flame propagation kinetics, chemical reactions are related to the equilibria in the thermodynamic reactions. First, N2 does not participate in the reactions in flames in the range of 1200 K, meaning that it only diluted the reactants and affected the chemical reaction kinetics in flames. CO2 is an important product of CO/H2 flames. The effects of CO2 dilution were compared to those of N2 dilution to estimate the influences of kinetic factors and then the effects of the variation of thermodynamic reactions caused by CO2 dilution on the chemical reaction rate and flame length. Considering 32% N2 (17.4 L/min) dilution as an example, according to the momentum formula , as the momentum increases caused by the two diluents were equal, the calculated CO2 addition amount was 13.8 L/min and the dilution ratio was 27%. Assuming that CO2 dilution only affected the momentum change, the flame length obtained under 27% CO2 dilution should have been the same as that under 32% N2 dilution. However, the flame length obtained under a CO2 dilution ratio of 24% had exceeded the flame length of that under 32% N2 dilution. The main chemical reactions in the CO/H2 flame that could be affected by CO2 dilution are CO + OH = CO2 + H (1) and CO + O2 = CO2 (2).[60] Under the condition of lean fuel combustion, there was sufficient O2. The flow rates of air, CO, and H2 were kept constant in this study. This implies that CO, O2, H, and OH were not the factors affecting the progress of chemical reactions. This indicates that CO2 dilution inhibited the chemical reactions in the flame, which reduced the reaction rate and increased the flame length.

Flame Temperature

The flame temperature and its change in the high-temperature zone have a great influence on the performance of industrial furnaces. In this work, the temperature distributions along the flame axial centerline were measured under several different conditions (Case 1, 2, 3, 4, 5, 6, 7, 8, 9, 11, and 13).

Figure shows the flame temperature distribution along the centerline of the flame axis as the amount of each diluent was increased. As the amount of each diluent increased, the flame temperature decreased. This occurred due to the dilution effect of the diluent on the reactants, which reduced the intensity of the reaction and the heat release. The highest flame temperatures for each diluent are listed in Table . TH is the highest temperature recorded in each flame. Under CO2 dilution, the flame temperature was much lower than that with the same amount of N2, whereas the flame temperature under CO2/N2 dilution was between the two. Since the specific heat of CO2 at normal atmospheric temperature is higher than that of N2, the mixed gas under 16 and 32% CO2 dilution will absorb more heat, resulting in the flame temperature 2 and 5 K higher than that under N2 dilution theoretically, respectively. In addition, CO2 dilution is not conducive to the exothermic reaction of CO oxidation, which will also reduce the intensity of the reaction, resulting in a lower flame temperature.

Figure 9

Centerline flame temperature distributions under different diluents and different levels of dilution.

Table 3

TH under Different Dilutions of CO2, N2, and CO2/N2

flame	CO/H2	CO/H2/CO2				CO/H2/N2				CO/H2/CO2/N2
dilution	none	CO2				N2				CO2/N2
VD (%)	none	8	16	24	32	8	16	24	32	16	32
TH (K)	1228	1202	1167	1123	1064	1218	1195	1168	1135	1181	1092

Taking the derivative of the flame temperature by Origin software, the rate of variation of the temperature at different positions in the axial direction was obtained, as shown in Figure . It is interesting to note that as the axial distance increased, two inflection points appeared in the derivative of the flame temperature. In this work, the inflection point was defined as the point where there was a significant difference in the temperature variation rate. At the first point, the flame temperature reached the peak value, and the intensity and rate of the chemical reaction in the flame both reached their maximums. The temperature variation rate dropped sharply at the other point, indicating rapid decreases in the chemical reaction intensity and reaction rate. The variation rate from the second point to the flame front tended to stabilize, indicating that the temperature variation was mainly affected by air dilution. Although other inflection points were observed for Case 1, 2, and 6, these were caused by the rapid temperature decrease from the flame to the exhaust zone. According to the flame temperature and its variation rate, the flame could be divided into two parts in this study. The area before the second inflection point was defined as the high-temperature reaction zone, meaning that nearly all fuel was consumed and almost all chemical reactions were completed. The subsequent area was defined as the postcombustion zone as no major reaction occurred. The high-temperature reaction zone increased following the addition of diluents. The axial length of the high-temperature reaction zone (LH) and variations in LH are presented in Table . Compared with under N2 dilution, the LH was longer under identical levels of CO2 and CO2/N2. This indicates that the intensities and rates of the chemical reactions in flames under CO2 dilution were lower than those under N2 dilution. To verify the effects of CO2 on the chemical reactions and understand the reaction process, the species concentrations in these flames are discussed in the next section.

Figure 10

Derivative to the flame temperature under different diluents and levels of dilution.

Table 4

LH for Each Dilution and Levels of Dilution

flame	CO/H2	CO/H2/CO2				CO/H2/N2				CO/H2/CO2/N2
dilution	none	CO2				N2				CO2/N2
VD (%)	none	8	16	24	32	8	16	24	32	16	32
LH (mm)	98	112	126	140	161	98	105	119	133	119	147

CO and CO2 Concentration in Flames

Species concentrations in flames are important parameters for characterizing the combustion process. In this work, the CO and CO2 concentrations along the axial direction at the flame centerline under several different conditions (Case 1, 3, 7, and 11) were measured. The effects of the three diluents on CO and CO2 concentrations and their variations were investigated.

Figure shows the comparison of CO and CO2 concentration distributions along the flame axial centerline. The CO concentration reached its peak at the outlet of the burner, while the profile of the CO2 concentration distribution was similar to a Gaussian curve. The highest CO concentration (CH) values in the undiluted and diluted flames are presented in Table . Theoretically, the CO concentration should have decreased under the three different diluents. However, the CH measured in the CO2 diluted flame was higher than in the CO/H2 flame, and the CH measured in the CO2/N2 diluted flame was higher than that in the N2 diluted flame. The addition of CO2 inhibited the oxidation of CO to CO2, increasing the CO concentration.

Figure 11

CO and CO2 concentration distributions along the flame axial centerline for different diluents and different levels of dilution.

Table 5

CH Values in Flames for Different Diluents and Different Levels of Dilution

flame	CO/H2	CO/H2/CO2	CO/H2/N2	CO/H2/CO2/N2
dilution	none	CO2	N2	CO2/N2
VD (%)	none	16	16	16
CH (%)	38.24	41.78	28.62	31.93

There were two inflection points in the distributions of CO and CO2 concentrations in flames with different diluents. By evaluating the derivatives of the CO and CO2 concentrations, the variation rates of the CO and CO2 concentrations at different positions along the flame axial direction were obtained, as shown in Figure . The variation rate of CO concentration decreased first in the N2 (16%) and CO2/N2 (8%/8%) diluted flames, while that of the CO2 concentration increased first. Both variation rates then flattened. The inflection points were in the mutation region of the variation rate. Following dilution with CO2 (16%), the profile of CO2 concentration variation rate was identical to those of the other two diluted flames, while the variation rate of the CO concentration decreased at the burner outlet. This was due to the inhibition of high CO2 concentration on CO oxidation, which reduced the chemical reaction intensity. In the zone before the two inflection points, the CO and CO2 concentrations varied drastically, indicating that strong CO oxidation reactions were occurring in the flame. As the intensity of the chemical reaction then became very low, variations in CO and CO2 concentrations in this zone were mainly affected by air dilution, which stabilized their variation rates. Since the interval distance between the concentration measurement positions was 14 mm, the inflection point was located in the area 14 mm before the mutation point, as marked in Figure and presented in Table . PIP is the distance between the inflection point and nozzle, which can be defined as the size of the main reaction zone. The PIP under pure CO2 dilution was larger than that under N2 dilution. When 50% of the N2 was replaced with CO2, the PIP was also larger than that under pure N2 dilution. The sizes of the main chemical reaction zone in the CO2 and CO2/N2 diluted flames were larger than that of N2 dilution. This indicates that CO2 dilution had a lower intensity and rate of the chemical reactions in flames, compared with that under N2 dilution. In addition, the inflection points for flame temperature were also in the mutation areas for the CO and CO2 concentrations, which verified the effect of CO2 on the chemical reactions.

Figure 12

Derivatives of CO and CO2 concentrations for different diluents and different levels of dilution.

Table 6

PIP Values for Different Diluents and Different Levels of Dilution

flame	CO/H2	CO/H2/CO2	CO/H2/N2	CO/H2/CO2/N2
dilution	none	CO2	N2	CO2/N2
VD (%)	none	16	16	16
PIP (mm)	84–98	112–126	98–112	112–126

Conclusions

Inert gases can significantly influence the combustion characteristics of industrial gases. In this study, the effects of using different diluents in a turbulent partially premixed CO/H2–air flame were experimentally investigated. The effects of CO2, N2, and CO2/N2 dilution were determined in terms of the flame structure, flame temperature, and CO and CO2 concentrations in the flames. The flame temperature and CO2 concentration were measured along the axial direction at the flames’ centerline using the TDLAS technique, by employing the spectral lines of 1996.89 and 2004.02 nm, respectively. The CO concentration was measured using an infrared gas analyzer. The major results of this investigation are as follows:

The flame length increased with increasing levels of dilution for each diluent, owing to the increasing momentum. The flame length was longer under CO2 dilution than that under the same level of N2 dilution. As CO2 has a higher density than N2 at normal atmospheric temperature and inhibits chemical reactions in the flame, CO2 dilution not only increases more momentum but also reduces the reaction rate, thus resulting in a much longer flame length than that for N2 dilution.

The flame temperature decreased with increasing levels of dilution. The highest flame temperature under CO2 dilution was much lower than that under the same level of N2 dilution. Compared with N2 dilution, the LH values under the same levels of CO2 and CO2/N2 dilution were much longer. These results indicate that the intensities and rates of the chemical reactions in the flames under CO2 dilution were lower than those under N2 dilution.

The highest CO concentration in the CO2 diluted flame was higher than that in the CO/H2 flame and that in the CO2/N2 diluted flame was higher than that in the N2 diluted flame. The sizes of the main chemical reaction zone in the CO2 and CO2/N2 diluted flames were larger than that under N2 dilution. The addition of CO2 inhibited the oxidation of CO to CO2, increasing CO concentration.

There were inflection points in the rates of variation of the flame temperature, CO concentration, and CO2 concentration in flames; these points were nearly in the same position. This is the validation of the effects of CO2 dilution on the flame temperature and species concentration in flames.