Measurements of the laminar burning velocity of hydrogen–air premixed flames

Measurements of the laminar burning velocity of hydrogen–air premixed flames

international journal of hydrogen energy 35 (2010) 1812–1818 Available at www.sciencedirect.com journal homepage: www.elsevier.com/locate/he Measur...

951KB Sizes 5 Downloads 63 Views

international journal of hydrogen energy 35 (2010) 1812–1818

Available at www.sciencedirect.com

journal homepage: www.elsevier.com/locate/he

Measurements of the laminar burning velocity of hydrogen–air premixed flames Jhon Pareja a,*, Hugo J. Burbano a, Yasuhiro Ogami b a

Science and Technology of Gases and Rational Use of Energy Group, Faculty of Engineering, University of Antioquia, Calle 67 N 53, 108 Bloque 20, 447 Medellı´n, Colombia b Institute of Fluid Science, Tohoku University, 2-1-1 Katahira, Aoba-ku, Sendai, Miyagi 980-8577, Japan

article info

abstract

Article history:

Experimental and numerical studies on laminar burning velocities of hydrogen–air

Received 16 October 2009

mixtures were performed at standard pressure and room temperature varying the equiv-

Received in revised form

alence ratio from 0.8 to 3.0. The flames were generated using a contoured slot-type nozzle

5 December 2009

burner (4 mm  10 mm). Measurements of laminar burning velocity were conducted using

Accepted 6 December 2009

particle tracking velocimetry (PTV) combined with Schlieren photography. This technique

Available online 4 January 2010

provides the information of instantaneous local burning velocities in the whole region of the flame front, and laminar burning velocities were determined using the mean value of

Keywords:

local burning velocities in the region of non-stretch. Additionally, average laminar burning

Hydrogen combustion

velocities were determined using the angle method and compared with the data obtained

Laminar burning velocity

with the PTV method. Numerical calculations were also conducted using detailed reaction

PTV

mechanisms and transport properties. The experimental results from the PTV method are in good agreement with the numerical results at every equivalence ratio of the range of study. Differences between the results obtained with the angle method and those with the PTV method are reasonably small when the effects of flame stretch and curvature are reduced by using a contoured slot-type nozzle. ª 2009 Professor T. Nejat Veziroglu. Published by Elsevier Ltd. All rights reserved.

1.

Introduction

One of the most important parameters for the analysis of the combustion phenomena is the laminar burning velocity, SL. Information on SL is used for the study of the structure and stability of premixed flames, validation of reaction mechanisms, and the analysis of turbulent premixed combustion. For the analysis of turbulent premixed flames, data on laminar burning velocities is especially necessary for obtaining the flame characteristic time, the turbulent Karlovitz number, the turbulent Damko¨hler number, and the relationship between the turbulent burning velocity and the turbulence intensity.

SL is defined as the flame propagating speed of onedimensional flat flame. However, such kind of flame is an idealization since actual flames are affected by heat losses and stretch, thus other types of flames have to be used for measurements. Burning velocities of hydrogen–air premixed flames have been extensively studied in the past using two typical measurement methods, the spherical bomb method and the burner stabilized flame method [1–6]. The spherical bomb technique has been used for obtaining recent experimental data on laminar burning velocities. However, in this technique flame propagation is unsteady, the duration of the combustion is very short, pressure changes with time, and observation of the flame front, to determine if it

* Corresponding author. Tel.: þ57 4 219 85 45; fax: þ57 4 211 90 28. E-mail address: [email protected] (J. Pareja). 0360-3199/$ – see front matter ª 2009 Professor T. Nejat Veziroglu. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.ijhydene.2009.12.031

international journal of hydrogen energy 35 (2010) 1812–1818

Nomenclature SL f Le q U

laminar burning velocity, cm/s equivalence ratio of the mixture Lewis number half flame angle,  average velocity of the unburned mixture, cm/s

is smooth, is difficult. Additionally, since extrapolations to the non-stretched flame are needed, SL cannot be obtained directly from experimental data. Recently, Chen et al. [7] showed theoretically, numerically and experimentally that there is a critical flame radius, strongly dependent on the Lewis number, only above which extrapolations are valid. At large Lewis numbers, the critical radius is larger than the minimum flame radius used in previous experimental measurements, leading to invalid flame speed extrapolations. Their results also show that there is a strong dependence of flame trajectory on ignition energy, leading to the occurrence of the flame speed reverse phenomenon, which greatly narrows the experimental data range for flame speed extrapolation. In the case of the burner stabilized flame technique, long duration experiments and continuous observation of the flame are achieved to determine the stability of the flame front, while pressure and temperature of the mixture remain constant. SL can be determined either by the flame area or by the flame angle. However, only average values of SL can be obtained by using this technique since local burning velocities vary along flame front due to effects of flame stretch at flame tip and heat losses near the burner walls. At atmospheric conditions data on SL measured with the same method agree with each other for different equivalence ratios. However, values measured with the spherical bomb technique are much lower than those measured with the burner stabilized flame technique. Qin et al. [8] measured the SL of hydrogen–air mixtures using a burner stabilized flame technique which allows determining local burning velocities on the region of non-stretch flame with particle tracking velocimetry (PTV) and instantaneous Schlieren photography. However, a circular-type nozzle burner was used; therefore the flame was affected by stretch due to the flame curvature in the direction of the burner axis. Their results at standard pressure and temperature are in agreement with the former studies using the burner stabilized flame method. Recently, Ogami and Kobayashi [9,10] have measured the SL of methane–air premixed flames and of H2/O2/He premixed flames using PTV technique and planar laser induced fluorescence for the OH radical (OH-PLIF) with a slot-type nozzle burner in order to remove the effect of the flame curvature mentioned previously. Due to the advantages and improvements cited above, in this study, measurements of the laminar burning velocity of hydrogen–air premixed flames were performed at standard pressure and room temperature using a burner stabilized flame with PTV technique and instantaneous Schlieren photography in order to determine local burning velocities along the flame front on the region of non-stretch; this technique will be referred as PTV method in this paper. Flames were generated for different equivalence ratios using a slot-type nozzle burner

Sr q1 Ur Ka

1813

local burning velocity, cm/s angle between the tangential line of the flame front and the streamline,  magnitude of the local velocity vector at the front edge of the preheating zone, cm/s. Karlovitz number

in order to remove the stretch due to the flame azimuthal curvature. PTV technique was improved taking advantage of recent advances in photography by using a high-resolution camera that allowed determining more accurately the center of the particles seeded. Experimental data on SL obtained with the PTV method was compared with previous reported data and with numerical calculations using existing reaction mechanisms. Additionally, average SL was also calculated using the burner stabilized flame method (with the flame angle) and compared with data obtained from the PTV method.

2.

Experimental and numerical methods

2.1.

Experimental setup

Fig. 1 shows a schematic diagram of the experimental setup implemented. Flames were generated using a small burner with a contoured slot-type nozzle (4 mm  10 mm) in order to keep laminar Reynolds numbers at every equivalence ratio studied as well as to reduce the effects of flame stretch and curvature in the direction of the burner axis. The design of the burner nozzle allowed obtaining a nearly uniform exit velocity profile, which gave a defined triangular flame with fairly straight edges. A high-intensity discharge Xenon lamp as well as a set of lenses and mirrors were used to generate Schlieren images of the flame, and photographs were taken using a highresolution CCD camera (Ikegami SKC-133, 1300  1030 pixels). A double-pulsed Nd-YAG laser (Continuum Minilite-PIVD) was used for particle tracking velocimetry. The laser beam was transformed into a vertical sheet with a thickness of less than 100 mm by conical and cylindrical lenses. SiO2 particles with specific gravity and mean diameter of 2.0 and 2.0 mm, respectively, were added to the mixture as Mie scattering particles, and their images were obtained by a high-resolution CCD camera (NIKON D50; 3008  2000 pixels).

2.2.

Flame instabilities

Laminar flames are prone to corrugations of the flame front mainly due to intrinsic flame instabilities, i.e., hydrodynamic instability and diffusive-thermal effect. Hydrodynamic instability is caused by the thermal expansion across the flame front; essentially this instability is always present. However, the hydrodynamic instability is restrained by the diffusive-thermal effect when the Lewis number, Le, is larger than unity [11]. Fig. 2(a) and (b) shows instantaneous Schlieren photographs of H2/air premixed flames at equivalence ratios of 1.0 and 0.6, respectively. At f ¼ 0.6, flame becomes unstable and

1814

international journal of hydrogen energy 35 (2010) 1812–1818

Fig. 1 – Schematic diagram of the experimental setup.

the flame tip occasionally splits due to the intrinsic flame instabilities since Le of this mixture is very low. At f ¼ 1.0, Le is less than unity, however the diffusive-thermal effect is enough to restrain the hydrodynamic instability and keep the flame smooth and stable. With the increase in the equivalence ratio, the flame remains stable since Le of the mixture increases due to the high thermal diffusivity of hydrogen [8]. In this study, the measurements of SL were performed varying the equivalence ratio of the mixtures from 0.8 to 3.0, in which the flames were smooth and stable.

simulations considered four detailed reaction mechanisms, specifically, the mechanism of Mueller et al. [14], Allen et al. [15], Conaire et al. [16], and the H2/O2 reactions in GRI-Mech 3.0 [17]. Transport properties were evaluated using the multicomponent diffusion model, and thermal diffusion (Soret effect) was included in the calculation. A re-gridding strategy was implemented; the number of grid points was increased until SL converged. SL was calculated for the same range of equivalence ratios as in the experimental study.

2.3.

3.

Results and discussion

3.1.

Determination of the laminar burning velocity

Numerical method

Numerical calculations were conducted using one-dimensional premixed flame code, PREMIX [12], and the thermochemical data and transport properties were evaluated using the CHEMKIN-II package [13]. For comparison, present

Laminar burning velocities were determined by the burner stabilized flame method and by the PTV method combined

Fig. 2 – Instantaneous Schlieren photographs of H2/air flames. (a) f [ 1.0. (b) f [ 0.6.

international journal of hydrogen energy 35 (2010) 1812–1818

with Schlieren photography. In the case of the burner stabilized flame technique, the flame propagates toward the unburned mixture at an angle of q as shown in Fig. 3. The velocity component of the unburned mixture which is normal to the flame front is identical to the laminar burning velocity, therefore SL is calculated as follows: SL ¼ U sin q

(1)

where U is the average velocity of the unburned mixture. This method is usually called as the angle method. However, local burning velocities vary along the flame front due to the effects of flame stretch and heat loss, and only average values can be obtained by this method. In this study, with the aim of obtaining a well-defined triangular flame for determining q more accurately, a burner with a contoured slottype nozzle was used to generate the flames. A contoured nozzle burner is preferred over straight cylindrical tube burner for the following reason. The exit velocity profile of a long cylindrical tube is parabolic and hence the flame angle to the incoming flow varies along the flame height. Whereas, a contoured nozzle produces a nearly uniform exit velocity profile, which gives a defined flame with fairly straight edges [18,19]. Besides, the use of a slot-type nozzle enables to remove the effects of flame stretch and curvature in the direction of the burner axis that appear when a circular-type nozzle is employed. In order to measure q, Schlieren images are utilized to determine the flame front. As shown in Fig. 2(a), two flame edges can be identified: the Shadowgraph edge and the Schlieren edge. Shadowgraph edge is dependent on the distance between the flame and the photographic plate and is hence unreliable to determine the flame front. On the other hand, Schlieren edge is not dependent on the distance mentioned before, and Schlieren image serves to determine the flame front and thus the laminar burning velocity since the flow lines remain parallel to the burner axis until the Schlieren edge is reached [20]. Due to the high-resolution achieved in the flame Schlieren image, flame front was selected as the center line of the Schlieren edge.

Fig. 3 – Burner stabilized flame technique (angle method).

1815

As pointed out above, local burning velocity, Sr, varies along the flame front on an actual flame. Sr decreases near the burner exit due to heat losses to the burner wall and increases at the flame tip due to the flame stretch, however Sr remains constant in a region between the flame base and the flame tip [18,21]. Therefore, a PTV technique combined with Schlieren photography was implemented to measure Sr in the constant region and thus to determine SL more accurately. First a set of binarized particle images collected by the CCD camera is analyzed using a PIV analysis package (AEA Technology, VISIFLOW), and a map of instantaneous velocity vectors over the cross section of the flame was generated as shown in Fig. 4(a). Then, from this map, the local streamlines of particles are calculated by successive interpolation. Along each streamline, the point where the direction of a velocity vector changes abruptly is determined by calculation. After that, the flame front from the Schlieren image is superposed on the streamlines, Fig. 4(b). At the intersection of the calculated streamline and the flame front, after calculating the angle qr, between the tangential line of the flame front and the streamline, Sr is determined as follows: Sr ¼ Ur sin qr

(2)

where Ur is the magnitude of the local velocity vector at the front edge of the preheating zone, i.e., the location of the change in local velocity vector. Fig. 5 presents an experimental result for a stoichiometric H2/air premixed flame at standard pressure and room temperature. This result shows nearly constant values of Sr on the region of measurement. Additionally, Karlovitz number, Ka, was estimated from the local velocity gradient as can be seen in Fig. 5. In this region, Ka are lower than O(103), thus flame curvature and stretch are minimum. That is why SLis considered to be the mean value of Sr. In all experimental conditions, similar magnitudes of Ka were obtained, hence the effect of flame stretch was sufficiently small. Fig. 6 shows the experimental results of the laminar burning velocity of H2/air mixtures for different equivalence ratios at standard pressure and room temperature along with data from previous studies. Three groups of data can be observed, namely, data obtained with the burner stabilized flame method [2,4,6], those with the spherical bomb method [1,3,5], and with the PTV method [8]. Data on SL measured with the same method agree with each other for different equivalence ratios. However, values measured with the spherical bomb technique are much lower than those measured with the burner stabilized flame technique. Despite the fact that present measurements were performed with a burner stabilized flame, experimental data on SL is lower than those obtained in previous studies with the burner stabilized flame method. This difference is associated with the fact that in those previous measurements SL was determined as an average value; meaning that effects of stretch, curvature, and heat losses were included. In contrast, as it has been explained above, in the present work a contoured slot-type nozzle burner was used in order to reduce as much as possible the effects of stretch and curvature in the direction of the burner axis, and SL was determined from the data of the local burning velocity in the region of non-stretch without

1816

international journal of hydrogen energy 35 (2010) 1812–1818

Fig. 4 – Example of experimental results at f [ 1.0. (a) Velocity vectors obtained by PTV. (b) Streamlines superposed on flame front.

ratios in the case of H2/air mixtures, the critical radius is larger than the minimum flame radius used in previous experimental measurements, leading to invalid flame speed extrapolations. Their results also showed that there is a strong dependence of flame trajectory on ignition energy, leading to the occurrence of the flame speed reverse phenomenon, which greatly narrows the experimental data range for flame speed extrapolation. At f ¼ 1.0, SL value of the present work (237 cm/s) is in good agreement with the experimental data of Qin et al. [8]. However, at other equivalence ratios their experimental data are larger than data of the present work although they used the same method as the present study. This may be explained by the fact that they employed a circular-type nozzle burner and therefore the flame was affected by stretch due to the flame curvature in the direction of the burner axis. Additionally, PTV method was improved in the

500

0.08

450

0.07

400

0.06 0.05

350

Sr

300

0.04

S L = 237 cm / s

250

0.03 0.02

200 150

0.01

100

0

50 0 0.8

Karlovitz number, Ka

Local burning velocity Sr (cm/s)

considering the values near the flame base and the flame tip since those were affected by heat loss and stretch respectively. Referring to Fig. 6, data on SL obtained with the spherical bomb method were lower than those obtained in the present work, especially at rich equivalence ratios. This difference may be explained by the fact that values of SL with the spherical bomb method are the result of extrapolations to the non-stretched flame, which means that SL was not obtained directly from the experimental data. Additionally, Chen et al. [7] have reported recently some restrictions related to the extrapolations made to obtain the non-stretch SL from the spherical bomb technique. They showed theoretically, numerically and experimentally that there is a critical flame radius, strongly dependent on the Lewis number, only above which linear and non-linear extrapolations are valid. At large Lewis numbers, i.e., rich equivalence

-0.01

Ka

-0.02 1.1

1.4

1.7

2

2.3

2.6

2.9

3.2

3.5

3.8

X (mm)

Fig. 5 – Local burning velocities, Sr, and local Karlovitz number, Ka, on the region of non-stretch flame at f [ 1.0.

1817

international journal of hydrogen energy 35 (2010) 1812–1818

400

Laminar burning velocity, SL (cm/s)

350 300 250 200

Pres ent wo rk Qin et al.(8) Tse et al.(5) Kwo n et al. (3) Aung et al.(1) Liu and Mac Farlane. (4) Günther and J anis c h. (2) Wu and Law. (6) mec hanis m o f Mueller et al.(14) mec hanis m o f Allen et al. (15) mec hanis m o f Co naire et al. (16) GRI-Mec h v er. 3. 0 (17)

PTV Metho d

Experim en t

150

Bo mb Met ho d

Burner Metho d

100 50

Calcu lation

0 0

0.5

1

1.5

2

2.5

3

3.5

Equivalence ratio, φ Fig. 6 – Laminar burning velocities of H2/air flames at different equivalence ratios.

spherical bomb method. As in the present work, it is only intended to evaluate and report which reaction mechanism better reproduces the present experimental, a deep study is necessary to understand why they generate similar values of SL at low equivalence ratios and different values at high equivalence ratios, since these reaction mechanisms have some differences in rate coefficients and transport properties. In Fig. 7 the experimental data of the average laminar burning velocity of H2/air mixtures obtained with the angle method as well as the results of SL obtained with the PTV method are shown. Schlieren images were used to determine the flame front and the half flame angle, and average laminar burning velocities were calculated using equation (1). Maximum difference between the data obtained with the angle method and those with the PTV method is 11 cm/s at f ¼ 1.9. Therefore, using the angle method to determine SL with the improvements mentioned above, for reducing the effects of stretch and curvature, yields reliable data.

350

Laminar burning velocity, SL (cm/s)

present study by using a high-resolution camera that allowed obtaining particles images with diameters up to 20 pixels. Therefore center of the particles was determined more precisely with the high-resolution particles images, and thus velocities of the particles were calculated more accurately. The accuracy of the velocity measured from PTV depends on the accuracy of the time step spacing and the spatial accuracy of the particle locations. With respect to the first one, the trigger used to control the two laser pulses is quite accurate (0.1 ms) in comparison with the spatial accuracy. Concerning the spatial accuracy, for the whole recording field of this study 1 pixel represents 2.6 mm. The accuracy of the algorithm for determining the center of the particles is below the pixel size; a value of 0.1 pixel has been reported by Dalziel [22]. From Eq. (2) the relative error of the local burning velocity measurement can be estimated as 0.7% which represents an accuracy of 2.3 cm/s for the highest Sr measured in this work. However, although Sr in the non-stretch region is nearly constant, there are variations of its values measured in this work. The maximum variation was found to be 3.6% and this should be considered the accuracy of the method since SL is determined as the mean value of Sr. In Fig. 6 error bars of the present data are presented for all equivalence ratios. Fig. 6 also shows numerical calculations of SL using the mechanisms of Mueller et al. [14], Allen et al. [15], Conaire et al. [16], and the H2/O2 reactions in GRI-Mech 3.0 [17]. Present results are in good agreement with the numerical results using the mechanism of Mueller et al. [14] and Allen et al. [15]. Although the mechanism of Conaire et al. [16] is one of the newest mechanism, present experimental data differ with the numerical results conducted with this mechanism, this is because it was adjusted using the experimental data of Tse et al. [5]. Present results also differ with the numerical results using the GRI-Mech 3.0 [17]; this is because parameters for the H2/O2 reactions of this mechanism were optimized using data obtained with the

300

250 PTV M ethod

200

Angle M ethod

150

100 0.5

1

1.5

2

2.5

3

3.5

Equivalence ratio, φ

Fig. 7 – Comparison between the data on laminar burning velocities of H2/air flames obtained with the PTV method and with the angle method at different equivalence ratios.

1818

4.

international journal of hydrogen energy 35 (2010) 1812–1818

Conclusions

Measurements of laminar burning velocities for H2/air flames stabilized by a slot-type nozzle burner at standard pressure and room temperature were conducted using a technique based on particle tracking velocimetry and Schlieren photography. Additionally, average laminar burning velocities were calculated using the angle method. Numerical calculations were also performed using existing detailed reaction mechanisms for comparison with the present experimental results. The following results were obtained: 1. At low equivalence ratios (f < 0.8) H2/air flames showed flame instabilities due to interactions between hydrodynamic instability and diffusive-thermal effect. 2. Particle tracking velocimetry technique was improved by using a high-resolution CCD camera, which allowed reducing the error to determine local burning velocities along the flame front. 3. Experimental results for all equivalence ratios are in good agreement with the numerical results using the mechanisms of Mueller et al. [14] and Allen et al. [15]. 4. Reducing the effects of stretch and curvature of the flames by using a contoured slot-type nozzle burner yielded reliable data of the laminar burning velocities when determined with the angle method.

Acknowledgments Mr. Burbano and Mr. Pareja are very grateful to the Science and Technology of Gases and Rational Use of Energy Group, University of Antioquia, TURBODIESEL S.A., and COSERVICIOS S.A. for their financial support that made possible their research internship at Tohoku University. They are also very grateful to Professor Hideaki Kobayashi for allowing them to use the facilities of the Reacting Flow Laboratory during the development of this work.

references

[1] Aung KT, Hassan MI, Faeth GM. Effects of pressure and nitrogen dilution on flame/stretch interactions of laminar premixed H2/O2/N2 flames. Combust Flame 1998;112(1–2): 1–15. [2] Gu¨nther R, Janisch G. Measurements of burning velocity in a flat flame front. Combust Flame 1972;19(1):49–53. [3] Kwon OC, Faeth GM. Flame/stretch interactions of premixed hydrogen-fueled flames: measurements and predictions. Combust Flame 2001;124(4):590–610.

[4] Liu DDS, MacFarlane R. Laminar burning velocities of hydrogen–air and hydrogen air–steam flames. Combust Flame 1983;49(1–3):59–71. [5] Tse SD, Zhu DL, Law CK. Morphology and burning rates of expanding spherical flames in H2/O2/inert mixtures up to 60 atmospheres. Proc Combust Inst 2000;28(2):1793–800. [6] Wu CK, Law CK. On the determination of laminar flame speeds from stretched flames. Proc Combust Inst 1985;20(1): 1941–9. [7] Chen Z, Burke MP, Ju Y. Effects of Lewis number and ignition energy on the determination of laminar flame speed using propagating spherical flames. Proc Combust Inst 2009;32(1): 1253–60. [8] Qin X, Kobayashi H, Niioka T. Laminar burning velocity of hydrogen–air premixed flames at elevated pressure. Exp Thermal Fluid Sci 2000;21(1–3):58–63. [9] Ogami Y, Kobayashi H. A study of laminar burning velocity for H2/ O2/He premixed flame at high pressure and high temperature. In: sixth ASME/JSME thermal engineering joint conference 2003 [TED-AJ03-520]. [10] Ogami Y, Kobayashi H. Laminar burning velocity of stoichiometric CH4/air premixed flames at high-pressure and high-temperature. JSME Int J Ser B 2005;48:603–9. [11] Sivashinsky GI. Instabilities, pattern formation and turbulence in flames. Annu Rev Fluid Mech 1983;15: 179–99. [12] Kee RJ, Grcar JF, Smooke MD, Ja Miller. Fortran program for modeling steady laminar one-dimensional premixed flames. Sandia National Laboratories 1993 [Sandia Report SAND898009]. [13] Kee RJ, Rupley FM, Miller Ja. Fortran chemical kinetics package for the analysis of gas phase chemical kinetics. Sandia National Laboratories 1991 [Sandia Report SAND858240]. [14] Mueller MA, Kim TJ, Yetter RA, Dryer FL. Flow reactor studies and kinetic modeling of the H2/O2 reaction. Int J Chem Kinet 1999;31:113–25. [15] Allen MT, Yetter RA, Dryer RL. High pressure studies of moist carbon monoxide/nitrous. Combust Flame 1997;109: 449–70. ´ , Curran HJ, Simmie JM, Pitz WJ, Westbrook CK. A [16] Conaire MO comprehensive modeling study of hydrogen oxidation. Int J Chem Kinet 2004;36(11):603–22. [17] Smith GP, Golden DM, Frenklach M, Moriarty NW, Eiteneer B, Goldenberg M, et-al. GRI-Mech Home Page, http:/www.me. berkeley.edu/gri-mech/; 1999. [18] Law CK. Combustion physics. Cambridge: Cambridge University Press; 2006. [19] Natarajan J, Kochar Y, Lieuwen T, Seitzman J. Pressure and preheat dependence of laminar flame speeds of H2/ CO/CO2/O2/He mixtures. Proc Combust Inst 2009;32(1): 1261–8. [20] Rallis CJ, Garforth AM. The determination of laminar burning velocity. Prog Energy Combust Sci 1980;6:303–29. [21] Lewis B, Von Elbe G. Combustion, flame and explosions of gases. London: Academic Press Inc; 1961. [22] Dalziel SB. Decay of rotating turbulence: some particle tracking experiments. Appl Sci Res 1992;49(3):217–44.