Volume 9 Issue 9 - July 24, 2009 PDF
Counter
A comprehensive study of two interactive parallel premixed methane flames on lean combustion

Ho-Chuan Lin1, Tsarng-Sheng Cheng2, Bi-Chian Chen1, Chun-Chin Ho1, Yei-Chin Chao1,*

1Institute of Aeronautics and Astronautics, National Cheng Kung University, Tainan 701, Taiwan, ROC
2Department of Mechanical Engineering, Chung Hua University, Hsinchu 300, Taiwan, ROC
ycchao@mail.ncku.edu.tw

Proceedings of the Combustion Institute, 32:995–1002, 2009
SCI Category: Engineering, Mechanical, Ranking 3 /107 = 2.8%

 
Font Normal   Font Enlarge
Abstract

This study aims to numerically and experimentally investigate interactive parallel lean premixed methane flames issued from twin rectangular slot burners with variable jet spacing. The flowfield and combustion chemical reactions are predicted by detailed numerical simulation with Skeletal and GRI-v3.0 mechanisms and validated by experimental particle image velocimetry (PIV) and flame measurements. When moved closer, these twin flames become interactive and both flames tilt outward in appearance with a wider operational range of lean and velocity conditions. There are three different interacting stages: entrainment, recirculation and reverse flows according to jet-to-jet spacing. At the reverse flow stage, a stagnating flowfield termed lateral impingement can be identified along the symmetric axis between the flames, which is similar but not identical to that found in the counter-flow flames. This reverse flow pattern provides a hot and slow postflame flowfield and transports the residual OH radicals from the main flames to heat and burn the fuel escaping from the stand-off gap between the flame base and burner rim.

Introduction

Lean combustion is generally considered as a timely solution to the more stringent environmental regulations and global weather concerns in a new era. However, the instability associated with the lean flame significantly keeps the lean combustion technique from being widely accepted as a major combustion technique for general applications. The past studies indicate some methods to improve the combustion instability such as vortex, swirling, interacting side-by-side diffusion burner array or matrix [1-3], hydrogen addition [4,5] etc. The swirling flow is a typical flowfield for gas turbines. The method using side-by-side fuel jet array looks promising, however, the difficulty of oxygen supply does create a limit to bring these two diffusion flames closer so that the enhancement of extending blowout limit is not as good as expected. The fourth method attracts more attention recently because it also improves the NOx emission. Nevertheless, the cost of hydrogen is high and expected to be higher in the future. In this study, we propose the premixed and partially premixed side-by-side jet array as a reasonable solution to the stability problem associated with lean premixed flames. The present study intends to explore the detailed flame structure and stabilization mechanism of interactive twin jet flames, based on the measurements of flow and thermal field and detailed numerical simulation of flame speed, heat release rate and chemical reactions on lean combustion conditions.

Methods of analysis

For experimental studies we used two 50 mm X 5 mm rectangular slot burners, and twin flames set in dimensionless jet spacing L/d = 2 ~ 6 with the mean inlet flow velocity ranging from 0.7 to 3 m/s and the equivalence ratio in the range of 0.55–1.38 were investigated. The temperature distributions were measured by thermocouples, and the global flame appearance was captured by a digital CCD camera. The flowfield of the interactive flames was carefully examined by using the PIV technique. This PIV digital camera was equipped with a mechanical shutter to prevent the PIV image from flame-illumination contamination. To numerically simulate the laminar premixed methane twin flames, the relevant governing equations were solved by using the CFD numerical code for flow, heat transfer, mixing and chemical reaction computations with database from CHEMKIN Rev. 3.0 package and with Skeletal and GRI Rev. 3.0 mechanisms for preliminary and detailed chemical calculations

Results and Discussion

As illustrated in Fig. 1 the flame offset angle, which is defined by the ratio of the offset distance of flame tip from jet central axis to the flame height, becomes larger as the burners approach each other from jet spacing of L/d = 4 down to 2 and with no apparent change to the offset angle as the inlet speed increases from 1.0 to 1.5 m/s. The comparison of the computed flame shape with respective to flame image on the left top corner of Fig. 1 shows satisfactory agreement. The variation of the flow pattern in the inboard gap is believed to be strongly related with the enhancement of the inboard flame stabilization, especially for lean conditions. As noted in Fig. 1 the flowfield near the inboard gap changes from entrainment flow for L/d = 4, recirculation flow for L/d = 3, and finally up to an ‘‘impingement” reverse flow for L/d = 2. The ‘‘impingement” reverse flow depicts a characteristic stagnating flow. The postflame flowfield for the L/d = 4 flame has a less pronounced local pressure peak outstanding from the yellow background. This local pressure peak is due to the limited space in this minor impinging area. For the case of L/d = 3, a local recirculation region of entrained flow at the symmetric axis is formed and the pressure peak at the symmetric axis becomes pronounced, pushing the flame outward more. When the jet spacing is L/d = 2, the postflame streamlines impinge each other strongly and a reverse flow is generated, creating a flame and a flowfield structure similar to those found in the counter-flow flame. At this case entrainment of ambient air is completely blocked.
Fig. 1. Effects of jet-to-jet spacing on flame and flowfield characteristics.

The counter-flow can also create a hydrodynamic effect to enhance the flame stabilization. Figure 2 shows the lateral impingement demonstrated by overlapping streamlines and PIV uniform vectors in the background of the XOH (left two pictures) and XCO (right one) mole concentration. It also shows the calculated streamlines and the first three streamlines designated as IB-1 IB-2 and IB-3 for inboard rim and OB-1, OB-2 and OB-3 for outboard ones respectively. This impingement area has an obvious stagnation point, designated as a lateral or an angled counter-flow, and the computed streamlines are carefully verified by the PIV measurement as denoted in uniform velocity vectors (show only the direction). The differences between the inboard and outboard regions in the flowfield pattern significantly affect the characteristics of the flame and the reaction near the burner rim as depicted by the color-coded OH and CO concentration distributions in Fig. 2. The characteristics of the thermal and the flame near the burner rim will directly influence the flame stabilization. The OH concentration in one spot of the IB-1 streamline is rich enough to support the reaction of OH + CH4 CH3+H2O because of the additional supply from stored HO2 and H2O2 through the reactions of H + HO2 2OH and H + H2O2 H2O+ OH. As shown in Fig. 2, the surplus of inboard OH mole fraction is extended deeply into the gap of the two burners along the reverse flow, which implies a high temperature zone with active chemical species does exist between the two flames and provides an environment to stabilize the flame. As observed, the weighting of heat release contribution shifts from reaction of OH + CH4 CH3+H2O to HO2 + CH3 OH + CH3O in the jet-to-jet gap area.
Fig. 2  Lateral impingement demonstrated by streamlines with OH and CO mole concentration background.

In summary of the stabilization mechanism of the twin interactive flames, a schematic drawing compared the inboard and outboard flame stabilization is presented in Fig. 3 of the jet spacing of 2d. At the reverse flow stage, a stagnating flowfield is generated along the symmetric axis between the flames. The transition of the postflame flowfield is believed to be the main mechanism to enhance the flame stabilization, especially in lean conditions. By regulating the amount of entrainment flow, this lateral impingement flowfield provides a zone with low dissipation of both the heat and the chemical species in this area, which would favor the flame stabilization. The wall proximity effect also turns HO2 into H2O2 to accommodate more incoming HO2 leading to additional OH supply for the combustion. The OH radicals are diffused and transported by the reverse flow into the impinging area and react with the escaping fuel to generate the additional heat in the gap region. It dramatically helps the flame base to be stabilized near the burner rim with the additional supply of OH, especially for lean combustion. On the contrary, the outboard flame, as shown in Fig. 3, suffers from the quench effect due to the cold entrainment flow which dissipates the heat as well as the active species near the rim tip.
Fig. 3 Schematic comparison of the differences of flowfield, thermal and chemical behaviors between inboard and outboard sides

Conclusions

In this study the parallel interactive premixed methane flames are identified numerically and experimentally as a phenomenon of lateral impingement of postflames. These parallel stagnation flames possess a few typical characteristics of counter- flow flames such as the flame stretching and the enhanced flame stabilization observed in this study. The stabilization of the interactive twin flames is enhanced by a number of crucial factors such as lateral stagnating flow, low dissipation of thermal and interacting chemical species supplied from the main flames. This flame stabilization mechanism is comprehensively explained in this study.

References

[1] B.-J. Lee, J.-S. Kim, S. Lee, Combust. Sci. Tech. 176 (2004) 482–497.
[2] F.G. Roper, Combust. Flame 34 (1979) 19–27.
[3] K. Seigo, H. Satoshi, U. Yoshito, M. Syuichi, A. Katsuo, 20th ICDERS, Canada, 2005.
[4] E.R. Hawkes, J.H. Chen, Combust. Flame 138 (2004) 242–258.
[5] J.-Y. Ren, F.N. Egolfopoulos, Combust. Sci. Tech. 174 (2002) 181–205.
< Previous
Next >
Copyright National Cheng Kung University