Quantifying acetylene mole fraction in rich flat laminar premixed C2H4/air flames using mid-infrared polarization spectroscopy

Mid-infrared laser polarization spectroscopy (IRPS) has been applied to measure the mole fraction of acetylene in rich premixed laminar C2H4/Air flat flames at equivalence ratios (Φ) of 1.7, 2.1, and 2.3, and under atmospheric pressure. The detection was conducted by probing the ro-vibrational P(19) transition at ~ 3.1 μm. The total collisional broadening coefficient of C2H2 was approximately 0.074 cm−1 atm−1 and varied within a range of 0.5% under different flame conditions, which made the effect of linewidth not obvious in the CH4/air flame. The calculated mole fraction of C2H2, using the Chemkin model, at Φ = 1.3 and 1.5 was used to calibrate the recorded IRPS signal intensities at different Height Above Burner (HAB). A single scaling factor was then used to quantify the measured C2H2 at highly sooting conditions, Φ = 1.7, 2.1, and 2.3, with a Limit of Detection (LoD) of 35 ± 5 ppm. The first observed C2H2 mole fraction appeared at HAB of 3 mm and measured as 2003 ppm, 2217 ppm, and 2495 ppm, for Φ = 1.7, 2.1, and 2.3, respectively. The mole fraction increased as the HAB increased to reach the maximum value of 2296 ppm, 2807 ppm, and 3478 ppm, for Φ = 1.7, 2.1, and 2.3, respectively, up to HAB of 5 mm. It was observed that the C2H2 mole fraction reaches a plateau region at HAB of ~ 8 mm. The production of C2H2 has been observed to be subject to a critical gas temperature of 1400 ± 30 K. The critical soot inception temperature, where the first incepted soot particles are observed, is the same as the gas temperature where χC2H2max\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${\chi }_{{{\mathrm{C}}_{2}\mathrm{H}}_{2}}^{\mathrm{max}}$$\end{document} was detected, namely at 1500 ± 30 K. These measurements and calibration procedure demonstrate a plausible technique to probe other flames and to better understand soot inception and its correlation with C2H2.


Introduction
Acetylene plays an important role in the formation and nucleation of soot particles and polycyclic aromatic hydrocarbons (PAHs) in all hydrocarbon flames due to the H-abstraction/C 2 H 2 -addition sequences in soot formation and growth [1]. The concentration of C 2 H 2 can vary considerably over very small spatial and very short temporal scales. Therefore, the detection of C 2 H 2 in flame conditions can contribute to understanding physical and chemical processes during soot formation. It can also help to control the emission of soot particles which have adverse effects on the environment and health. However, due to the harsh environment background in the sooting flames, such measurement is not trivial.
Laser polarization spectroscopy (PS), which is a coherent laser diagnostic technique, has been developed to offer non-invasive detection of minor combustion species with higher accuracy [2]. Compared with other nonlinear laser techniques, PS has the advantage of high intensity, and its experimental setups are relatively simple. The background influence can be reduced by a high-quality pair of crossed polarizers so that the signal from the target species can be discriminated from the surrounding environment [3].
PS has been applied in the UV/visible range based on electronic transitions of OH, C 2 , NH [4][5][6], and two-photon electronic transitions of NH 3 and CO [7]. However, only limited species can be measured in this spectral range. Many important species, such as CO 2 , C 2 H 2 , and CH 4 , cannot be probed by UV/visible PS as they do not possess accessible transitions. Mid-infrared laser polarization spectroscopy (IRPS) has been introduced as a tool to detect molecules in the mid-IR (2-5 um) via ro-vibrational transitions [3]. By probing molecular vibrational transitions, it is expected that the influence of molecular collisions on the signal will be reduced.
IRPS has been successfully applied in various hydrocarbon flames to detect infrared active species, such as CH 4 , CO 2 , HCl, and HCN, in the mid-IR (2-5 um) range via rovibrational transition [8,9]. For example, Li et al. performed the first IRPS measurement in laminar flames to detect CO 2 and H 2 O at atmospheric pressure [10], followed by the measurement of OH and H 2 O at low-pressure laminar flames [11]. However, the application of IRPS on C 2 H 2 , which is the main precursor of soot particles and PAHs, is scarce. Recently, Sun et al. [12] reported a quantitative measurement of C 2 H 2 in a sooty flame, however, this was only performed at one fixed location in the flame with the calibration performed separately in a different location. To the authors' knowledge, there is no spatially resolved systematic IRPS measurement of C 2 H 2 in the flame that has been previously reported in the literature.
This work reports on the use of IRPS to quantitively measure C 2 H 2 in C 2 H 4 /air laminar premixed sooty flames, using a McKenna burner at atmospheric pressure. The fundamental rovibrational transitions of C 2 H 2 were probed at the wavelength of around 3.1 μm in the near IR range. The IRPS spectra were recorded at the height above the burner surface (HAB) from 3 to 11 mm for spatially resolved measurement. The Chemkin model, at equivalence ratios of 1.3 and 1.5, was used to determine the calibration factor between the measured IRPS signals and the C 2 H 2 mole fractions, C 2 H 2 . The paper divides into the experimental details, calibration, and the quantitative measurement of C 2 H 2 along with the soot volume fraction at different heights above the burner. Finally, the collisional effect from the major combustion species is evaluated and discussed.

Experimental setup
The experiment setup was organized into three parts, which are a laser system to generate the infrared radiation at a spectral region around 3.1 μm, a McKenna burner to produce premixed C 2 H 4 /air sooty flame, and a laser diagnostics system with signal detection. Figure 1 shows a schematic diagram of the setup used in this work. The gas temperature at different HAB was measured by a 75 μm-diameter uncoated Pt/Pt-R thermocouple (Omega). The measured values were corrected based on the heat radiation loss method [13,14]. The fine-wire thermocouple was selected to minimize conductivity and radiation effects and other catalytic effects at the junction. Flat premixed C 2 H 4 /air flames were produced in a watercooled McKenna burner, which had a porous plate with a diameter of 60 mm. A steel plate of 80-mm diameter was employed at a height of HAB = 21 mm to stabilize the flames. To ensure the homogeneous mixing of gases, the mixture of C 2 H 4 and air was transferred through a 10-m Teflon pipe from the mass flow meter to the burner. The total flow rate of C 2 H 4 and air was controlled at 5 L/min. Additionally, nitrogen gas was controlled at the flow rate of 1 L/ min and used as co-flow to shield the flames. The burner was traversed vertically with a resolution of 1 mm to perform the measurement at different HAB.
The second harmonic of an injection-seeded Nd:YAG laser (Quantel, YG980), operated at 532 nm, was used to pump a tuneable dye laser (Quantel, TDL90) running with the mixture of laser dyes LDS 798 and LDS 765 purchased from Luxottica/Exciton. The output of the dye laser was then difference frequency mixed with a part of the Nd:YAG fundamental infrared 1064 nm beam in a mixing unit with a lithium niobate (LiNbO 3 ) crystal. Subsequently, a tunable IR laser beam was generated at the wavelength of around 3.1 µm with the power of 1.5 mJ per pulse. The linewidth (FWHM) of the seeded Nd:YAG laser and the tuneable dye laser is 0.005 cm −1 and 0.05 cm −1 , respectively. This results in a linewidth (FWHM) of ~ 0.05 cm −1 of the mid-IR beam at 3.1 μm. A wavelength meter (Bristol Instrument, 821B) was used to monitor the change in the wavelength of the dye laser during wavelength scanning.
A 632.8 nm HeNe laser was further overlapped with the IR beam in the first CaF 2 beam splitter (BS1) to visualize the IR beam path and facilitate alignment. After passing the second beam splitter (BS2), ~ 4% of the IR beam was reflected and used as the probe beam, while the remaining transmitted IR beam was used as the pump beam.
The pump and the probe beams were focused by f = 500 mm and f = 750 mm CaF 2 lens, respectively. A quarter-wave plate was placed in the path of the pump beam to generate a circular polarization, which enhances the P and R branches of C 2 H 2 detection.
The burner was placed between two crossed IR polarizers (YVO4) in the path of the probe beam. The probe beam was crossed with the strong pump beam at the detection volume in the center of the burner with a crossing angle of approximately 5 ○ . After passing the detection volume, the pump beam was blocked by a beam dump, while the probe beam emerged parallel using an f = 750 mm CaF 2 lens (L3) and then focused by an f = 100 mm CaF 2 lens (L4) to a liquid-nitrogen-cooled indium-antimonide (InSb) detector (Hamamatsu Photonics, P5968-060).
Additionally, in order to monitor the input laser power, ~ 4% of the probe beam was reflected to an amplified IR detector (Thorlabs, PDA10PT-EC) using a beam splitter (BS3). After being triggered by the Nd:YAG laser, both the laser power signal and the IRPS signal of C 2 H 2 could be simultaneously collected and stored in a 1-GHz oscilloscope to avoid errors associated with the input power fluctuation.

Numerical analysis
According to Demtröder [15], the PS signal can be given by the following equation when two polarizers are perfectly crossed and there is no other birefringent optics between the two polarizers: where I probe is the intensity of the probe beam; ξ is the leakage through the crossed polarizers; l is the interaction distance between the pump and probe beams along the probe beam path; L(ω) is the line shape function; Δα is the induced dichroism, which can be expressed as follows: where N is the number of molecules or radicals; f B is the Boltzmann fraction; σ is the polarization-independent absorption cross-section; I pump is the intensity of the pump beam; pump is the pulse width of the pump beam; ω is the laser wavelength, and ξ JJ' is the polarization numerical factor.
At atmospheric pressure and following IR excitation, the collision broadening is significantly larger than the Doppler broadening. Thus, the Lorentzian function was selected to demonstrate the line shape: [16] where ω 0 is the center wavelength; ∆ω self and ∆ω foreign are the line-broadening due to the self and the buffer gas collisions, respectively; and n is the Lorentzian power fitting factor, which is 3 for non-saturated pump energy in this work. The integration over the fitting lines and the maximum intensity at the center wavelength was used to present the integral and maximum IRPS intensities, respectively.

Chemical kinetic model
The flames with lean conditions were simulated using the premixed laminar burner-stabilized flame model in the Chemkin-Pro Software package. The mechanism developed at the University of Southern California (USC) was selected to conduct chemical kinetic analysis for (1) simulating the mole fraction of C 2 H 2 . The mechanism used in this work is the "High-Temperature Combustion Reaction Model of H 2 /CO/C1-C4 Compounds" with 784 reactions including a comprehensive reaction model of ethylene combustion [17]. The maximum number of grid points was 250 to provide grid-independent results and other solver parameters of this model were set as their default values. The Chemkin simulation was conducted, by imposing the measured temperature profiles, for Φ = 1.3 and 1.5. The Chemkin model is used to calibrate the simulated C 2 H 2 mole fractions, C 2 H 2 , from the experimental IRPS signals to obtain the quantitative calibration factor. Then, the IRPS intensities of the sooty flame were calibrated to their mole fractions by the calibration factor. The mole fractions of C 2 H 2 as a function of the height above the burner were well resolved at different sooty flame conditions. Figure 2 presents the dependence of the IRPS signal intensity on the pumping laser power. The dependence was measured by varying the pump laser power with different neutral density filters and using the mixture of 2.5% acetylene and N 2 . In this work, the maximum laser energy of the 3.1 μm mid-IR beam was 1.5 mJ, which contributes to the maximum pump laser energy of 1.44 mJ. The quadratic dependence in Fig. 2 indicates the laser power is still within the unsaturated regime because the PS signal is expected to be the square of the pump laser power ( I Unsaturated IRPS ∝ I 2 pump ) under unsaturated conditions and to be independent of the pump laser power under saturated conditions [4]. Shown in Fig. 3, a typical example of measured and simulated IRPS spectra of C 2 H 2 are collected in a flame with Φ = 2.1 and HAB = 5 mm. For the simulated spectrum of C 2 H 2 , the parameters of line position and intensity were extracted from the HITRAN database [18], and the line profile was characterized by the Lorentzian function. The measured spectra were compared with the simulated spectra to identify the location of the transition lines. According to Fig. 3, a strong vibrational transition P (19) line is marked and selected as the candidate line in this work due to its high intensity and lower interference from the background noise, under flame conditions. As described in Eqs. (1) and (2), some weak transitions with small absorption cross-section, σ, may not be detectable using unsaturated IRPS, where the pump energy was not sufficiently high.

Results and discussion
To obtain a high-resolution IRPS signal of the P(19) line, a low-speed wavelength scan was performed within a narrow spectral range from 3248.3 to 3248.8 cm −1 . Figure 4 shows the IRPS scan of the P(19) line in flame with Φ = 2.3 at HAB of 5 mm, 7 mm, and 10 mm. The scan wavelength interval was 0.00167 cm −1 between two consecutive measurements.
It is clear from Fig. 4(b) to (c) that although the IRPS intensity decreases with the increasing HAB, the signal linewidth at different HAB is very similar. Under flame conditions, the linewidth is both temperature-and pressure-dependent due to  where 0 is the center wavelength; c is the speed of light; k is the gas constant; m is the molecular mass. The bimolecular collision broadening of a spectral line can be corrected for the temperature effect using [20,21], where 0 is the collision broadening coefficient at the reference temperature T 0 . According to Eqs. (4) and (5), the Doppler width increases with the temperature, while the collision width has an opposite trend. At atmospheric pressure, the collision broadening is larger than the Doppler broadening for the non-saturated IRPS signal; therefore, the signal linewidth is dominated by collision broadening and decreases with increasing temperature. However, this change is not obvious in flames when the temperature is more than 1000 K [12]. In order to study the effects of linewidth, the IRPS intensity of the P(19) line was estimated by two methods, which are the integration of the signal intensity over the whole wavelength and the maximum intensity at the center wavelength. The comparison between the two methods is On the other hand, as a non-linear technique, the IRPS signal intensity is proportional to the laser power cubed ( I Unsaturated IRPS ∝ I 2 pump ⋅ I probe ). The slightly noisy signal can be caused by the unsaturated laser power and pulse-to-pulse laser energy, due to the nature of frequency mixing used to generate the mid-IR tunable laser radiation. Figure 5 shows the effect of pump beam scattering on the IRPS signal. Compared to pump-off, the background level of pump-on is increased by 15%. As the energy of the probe beam is much lower than the pump beam, the scattered light is mainly contributed by the pump beam and further results in the increase of background level. Although pump scattering can be reduced by increasing the crossing angle between the probe beam and the pump beam, however, a large crossing angle may cause a weaker IRPS signal due to a smaller detection volume. Therefore, the crossing angle used in the experiment is 5° to achieve relatively less pump scattering, but sufficient IRPS signal intensity. On the other hand, compared to the IRPS signal, the fluctuation of the background level is only 1% of the maximum IRPS intensity. Thus, the effect of pump scattering can be neglected in this work.
To be able to calibrate the IRPS signal intensity in flame conditions, the mole fraction of C 2 H 2 was varied by changing the HAB in Φ = 1.3 and 1.5. For these two Φ's, Chemkin is expected to produce reliable results [22,23] and the mole fraction of C 2 H 2 is high enough to be used for the calibration. Figure 6 shows the square roots of the experimental IRPS intensities as a function of the simulated mole fractions of C 2 H 2 for Φ = 1.3 and 1.5 flame conditions. The experimental signals at different HAB are compared with the Chemkin model in Fig. 6(a) and (c) and present good agreement. Two methods, i.e., the integral IRPS intensity and the maximum IRPS intensity, are presented. The integral IRPS intensity is presented by summing the signal intensity over the whole wavelength, while the maximum IRPS intensity is estimated by summing the signal intensity at the center wavelength ( 0 ± 0.09 cm −1 ). The line fitting in Fig. 6 (b) and (d) shows the calibration equation. Both methods yield a consistent trend, however, the maximum IRPS intensity shows a larger variance.
According to Eqs. (1) and (2), the IRPS intensity is proportional to the square of the mole fraction of the investigated species. Therefore, both quantities are connected by a single calibration factor. Based on a simple linear fitting, the calibration factor, C 1 , for line maximum was found to be 2.48 × 10 -3 . The line integration calibration factor, C 2 , was 1.45 times C 1 . The confidence level and an R-squared value were 95% and 0.977 and 0.939 for the line integration method and the line maximum method, respectively.
As the integral IRPS intensity has a better agreement between the experimental and modeling results, it has been selected to infer the C 2 H 2 mole fraction from the IRPS intensity for the sooty flames in the subsequent measurements, namely Φ = 1.7, 2.1, and 2.3.
On the other hand, the difference between calibration using the integral IRPS intensity and the maximum IRPS intensity is caused by the effects of linewidth, which is decreased with increasing temperature. It should be mentioned that although the maximum IRPS intensity does not consider the effect of linewidth, its calibration still can be fitted well by a linear equation with a relatively high R-squared value. This indicates that the effect of linewidth is not obvious in the C 2 H 4 /air flames, which is consistent with the work reported by Sun et al. [12]. Figure 7 shows the mole fraction of C 2 H 2 as a function of the height above the burner surface for the sooty flame with Φ = 1.7, 2.1, and 2.3. The C 2 H 2 mole fraction at each data point is calibrated from the corresponding line integration calibration factor, C 2 . The HAB was elevated from 3 to 10 mm for spatial-resolved measurement. The error bars are calculated based on the three repetitive measurements. As fuel-rich flames can promote the formation of C 2 H 2 , more C 2 H 2 species are detected in the flame with a higher equivalent ratio. From Fig. 7, the first observed C 2 H 2 mole fraction appeared at HAB of 3 mm and measured as 2003 ppm, 2217 ppm, and 2495 ppm, for Φ = 1.7, 2.1, and 2.3, respectively. The mole fraction increased as the HAB increased to reach the maximum value of 2296 ppm, 2807 ppm, and 3478 ppm, for Φ = 1.7, 2.1, and 2.3, respectively. According to the background level and the slope from Fig. 6(b), the Limit of Detection (LoD) was calculated as 35 ± 5 ppm.
In addition, the location of the peak mole fraction of C 2 H 2 also varies with the flame conditions. When the equivalent ratio increases from 1.7 to 2.3, the peak mole fraction of On the other hand, when increasing HAB, the signal reduction of C 2 H 2 could be contributed by the formation of PAHs and soot in the upward direction. It was observed that the mole fraction reaches a plateau region at HAB of ~ 8 mm. This mole fraction profile is similar to the results in C 2 H 4 /air flames with Φ from 1.77 to 2.37 reported by Otti et al. [24], where the fuel-rich flames exhibit a greater peak mole fraction of C 2 H 2 and the mole fraction decays with increasing HAB until ~ 7 mm.
On the other hand, Fig. 7 also compares the scaled IRPS results with the Chemkin simulated results for Φ = 1.7, 2.1, and 2.3. Compared with the measured results, the simulated results have a higher peak mole fraction of C 2 H 2 and its location also shifts to the higher HAB. This discrepancy becomes more obvious when increasing the equivalent ratio. At Φ = 2.3, the simulated peak mole fraction of C 2 H 2 and its location are approximately 750 ppm and 2 mm higher than the measured results, respectively. The clear discrepancy between the measured and computed C 2 H 2 values is evident of the  limitation to applying Chemkin in rich flame conditions because the prediction of soot formation is unavailable in this Chemkin model. As C 2 H 2 is consumed during the formation of soot, the higher mole fraction of C 2 H 2 was predicted in the simulated results, especially in the highly sooting conditions. A similar discrepancy was also observed by Bennett et al. in ethylene laminar flame [25].
One of the advantages of IRPS is its quantitative nature. This is because molecular collisions are not so pronounced as, for example, in Infrared Laser Induced Fluorescence (IRLIF). Molecular collisions may impact the IRPS signal in two ways, namely, in the form of reorientation and energy transfer. The former results in losing the orientation memory created by the pump beam and the latter changes the overlapping integral between the laser line width and the absorption line width, because of the collisional line broadening. For the short effective pulse length, estimated as 4 nsec, the numbers of kinetic collisions in C 2 H 2 -N 2 and C 2 H 2 -CO 2 systems are less than 5 and less than 1, respectively. In other words, molecular collisions are dominated by N 2 , and prior knowledge of the mixture fractions may not be necessary.
To be able to evaluate the effect of the collisional broadening of the major species, we assume a linear combination of the collisional broadening based on C 2 H 2 -N 2 and C 2 H 2 -CO 2 . The effective collisional broadening coefficient, taken into consideration only CO 2 and N 2 , is approximately given as follows: where N 2 and CO 2 are the mole fraction of N 2 and CO 2 , respectively, and C 2 H 2 −N 2 and C 2 H 2 −CO 2 are broadening coefficients caused by N 2 and CO 2 , respectively. For the P (19) transition, the collisional broadening coefficients C 2 H 2 −N 2 and C 2 H 2 −CO 2 is 0.073 cm −1 atm −1 and 0.082 cm −1 atm −1 , respectively [26,27]. Figure 8(a) shows the spatial-resolved temperature profile measured by the thermocouple, while Fig. 8(b) presents the spatial-resolved profiles of the mole fraction of N 2 and CO 2 for the flame with Φ = 1.5. Figure 8(c) shows the value of C 2 H 2 −x as a function of HAB for the flame with Φ = 1.5 and 2.3. Although the mole fraction of N 2 decreases as the increasing HAB while CO 2 has an opposite trend, the mole fraction of N 2 is more than 20 times higher than that of CO 2 , which makes the collisional broadening coefficients C 2 H 2 −N 2 dominates the total collisional broadening coefficients C 2 H 2 −x . Therefore, according to Fig. 8(c), the values of C 2 H 2 −x are around 0.074 cm −1 atm −1 for both flames, which is very similar to the collision broadening coefficient of N 2 . In addition, comparing the selected flames with Φ = 1.5 and 2.3, the C 2 H 2 −x of Φ = 1.5 is slightly higher than that of Φ = 2.3, which is caused by more CO 2 produced in the less sooty flame. However, this variance is only less than 0.5%. It further reveals the change of linewidth is not obvious in the CH 4 /air flame. Figure 9 shows the spatially resolved profiles of temperature, C 2 H 2 mole fraction, and soot volume fraction for the premixed C 2 H 4 / air flame under sooty conditions with Φ = 2.1 and 2.3. The lowest detectable value C 2 H 2 mole fraction was 2217 ppm, and 2495 ppm, for Φ = 2.1, and 2.3, respectively, observed at HAB of 3 mm. The gas temperature at HAB = 3 mm was determined to be 1400 ± 30 K. As the LoD was evaluated to be ~ 60 times smaller than the lowest recorded C 2 H 2 mole fraction, we can conclude that the production of C 2 H 2 is subject to a critical gas temperature of 1400 ± 30 K. The mole fraction of C 2 H 2 increases to the maximum when the maximum flame temperature is achieved. For example, at Φ = 2.3, the mole fraction of C 2 H 2 is found to reach the maximum at the flame temperature of ~ 1500 ± 30 K. The soot volume fraction was reported by Axelsson et al. using the LII technique under similar gas flows and flame conditions as in this study [28]. The soot particles are found to start forming at the location where the mole fraction of C 2 H 2 reached the maximum. The soot volume fraction increases significantly with the decaying trend of the C 2 H 2 mole fraction. It corroborates the role of C 2 H 2 in the formation of soot and the inception temperature of soot formation is ~ 1500 ± 30 K. This is in a very good agreement with the recent study of Algoraini et al. who report a soot inception temperature in low-pressure C 2 H 4 /Air premixed flames as 1465 ± 66 K. [14] It is interesting to conclude that the critical soot inception temperature is the same as the gas temperature where max C 2 H 2 was detected. In addition, even in highly sooting conditions, e.g., when Φ = 2.3, the soot volume fraction at HAB = 10 mm is 280 ppb, the C 2 H 2 mole fraction is still detectable. It further provides confidence that the IRPS method reported in this paper is suitable to measure the combustion species in such sooting flame conditions. It is also the first time that the measurement of C 2 H 2 is performed systematically using nonintrusive techniques along with the formation of soot particles, where the calibration was evaluated at flame conditions.

Conclusion
Quantitative measurement of C 2 H 2 was demonstrated in the premixed rich C 2 H 4 /air flame under atmospheric pressure using IRPS. The high-resolution IRPS signal was recorded by probing the ro-vibtational transition P(19) within a narrow spectral range from 3248.3 to 3248.8 cm −1 . The C 2 H 2 mole fraction was calibrated from the square root of the recorded IRPS intensity using a linear equation. The calibration factors for the two methods, i.e., the line integration and line maximum, were 3.60 × 10 -3 and 2.48 × 10 -3 , respectively. More C 2 H 2 species were detected in the flame with a higher equivalent ratio. The first observed mole fraction appeared at HAB of 3 mm and measured as 2003 ppm, 2217 ppm, and 2495 ppm, for Φ = 1.7, 2.1, and 2.3, respectively. The mole fraction increased as the HAB increased to reach the maximum value of 2296 ppm, 2807 ppm, and 3478 ppm, for Φ = 1.7, 2.1, and 2.3, respectively. The location of the maximum mole fraction of C 2 H 2 was found to shift to a higher location within fuel-rich flames. When the Φ increased from 1.7 to 2.3, the peak mole fraction of C 2 H 2 shifted from the HAB of 4 mm to 5 mm. After HAB for max C 2 H 2 , the C 2 H 2 mole fraction was decreased to a plateau region at HAB of ~ 8 mm. In addition, the clear discrepancy between the measured and computed C 2 H 2 mole fraction is evident of the limitation to applying Chemkin in rich flame conditions because the prediction of soot formation is unavailable in this Chemkin model. The C 2 H 2 LoD was determined as 35 ± 5 ppm with the present setup at atmospheric pressure conditions. The total collisional pressure broadening coefficient of C 2 H 2 was found to be dominated by the N 2 -broadening coefficients. Its value was ~ 0.074 cm −1 atm −1 and varied within a range of only 0.5% under the other flame conditions. For Φ = 2.1 and 2.3, the mole fraction of C 2 H 2 increased to the reach max C 2 H 2 , at the maximum gas temperature. It was found that the critical soot inception temperature is the same as the gas temperature where max C 2 H 2 was detected, namely at 1500 ± 30 K. The detected C 2 H 2 signal in the flame with high soot volume fraction provides confidence that the IRPS is capable to measure the combustion species in sooty flames. The quantitative and high-sensitive IRPS measurement reported in this work demonstrates the potential applications in flame environments to help understand physical and chemical processes. IRPS can be used as a tool to perform quantitative measurements of other important species such as NH and OH with alternative fuels in the future.