Elsevier

Combustion and Flame

Volume 160, Issue 3, March 2013, Pages 656-670
Combustion and Flame

Effect of diluents on soot precursor formation and temperature in ethylene laminar diffusion flames

https://doi.org/10.1016/j.combustflame.2012.11.004Get rights and content

Abstract

Soot precursor species concentrations and flame temperature were measured in a diluted laminar co-flow jet diffusion flame at pressures up to eight atmospheres while varying diluent type. The objective of this study was to gain a better understanding of soot production and oxidation mechanisms, which could potentially lead to a reduction in soot emissions from practical combustion devices. Gaseous samples were extracted from the centerline of an ethylene–air laminar diffusion flame, which was diluted individually with four diluents (argon, helium, nitrogen, and carbon dioxide) to manipulate flame temperature and transport properties. The diluted fuel and co-flow exit velocities (top-hat profiles) were matched at all pressures to minimize shear-layer effects, and the mass fluxes were fixed over the pressure range to maintain constant Reynolds number. The flame temperature was measured using a fine gauge R-type thermocouple at pressures up to four atmospheres. Centerline concentration profiles of major non-fuel hydrocarbons collected via extractive sampling with a quartz microprobe and quantification using GC/MS + FID are reported within. The measured hydrocarbon species concentrations are vary dramatically with pressure and diluent, with the helium and carbon dioxide diluted flames yielding the largest and smallest concentrations of soot precursors, respectively. In the case of C2H2 and C6H6, two key soot precursors, helium diluted flames had concentrations more than three times higher compared with the carbon dioxide diluted flame. The peak flame temperature vary with diluents tested, as expected, with carbon dioxide diluted flame being the coolest, with a peak temperature of 1760 K at 1 atm, and the helium diluted flame being the hottest, with a peak temperature of 2140 K. At four atmospheres, the helium diluted flame increased to 2240 K, but the CO2 flame temperature increased more, decreasing the difference to approximately 250 K.

Introduction

Many practical combustion devices rely on non-premixed combustion, allowing for more robust combustion and achieving higher fuel efficiency by operating at overall equivalence ratios well below the lean premixed flammability limit. The unfortunate consequence of non-premixed combustion is typically higher levels of soot emissions. It continues to be of utmost interest to mitigate pollutant production while maintaining fuel efficiency in diffusion flame dominated devices. Particulate matter (soot) is of great concern because it plays a role in the radiation balance and global climate change and is detrimental to human health, as it is a known carcinogen and mutagen and is believed to cause long-term illnesses [1], [2], [3], [4]. Soot formation is a complex process involving interactions between chemistry, transport and fluid mechanics, and these interactions are still not entirely understood, particularly at high pressure where transport rates are higher due to steeper concentration gradients. This study aims to gain a better, more complete, understanding of the soot formation process by measuring soot precursors at elevated pressures. It is crucial to understanding the effects of pressures on soot and soot precursor formation as most practical combustion devices operate at elevated pressures to increase their thermodynamic efficiency.

The fuel, ethylene, was diluted in these studies to reduce the soot yield, maintaining these flames below their smoke point at 8 atm, and allowing investigation of effects of flame temperature and transport rates. Using an additive (such as an inert diluent) has three general effects as described by Liu et al. [5]: (1) thermal effect due to the change in the flame temperature; (2) direct chemical effect (which may or may not be present) due to the active participation of the diluent in the chemical reaction concerning soot formation and oxidation; and (3) dilution effect since the concentrations of the reactive species responsible for soot formation are reduced. McLintock [6] investigated the effect of adding inert diluents (to either the fuel or the oxidizer stream) on the production of soot. Using a parabolic fuel flow exit velocity profile, he found a dramatic difference in soot loading between carbon dioxide and helium when used as a soot suppressant. When using carbon dioxide as the diluent, he reported a strong soot suppressing effect, but when using helium there was virtually no noticeable effect on soot suppression. Glassman [7] showed similar soot suppressing tendencies as those reported by McLintock [6]. Although their results differ slightly, it is clear that CO2 is a more efficient soot suppressant than helium. Pfefferle and McEnally [8] and Gulder and Snelling [9] investigated effects of different levels of nitrogen dilution on flame shape and hydrocarbon species profiles and found that maximum centerline concentrations of every measured non-fuel hydrocarbon monotonically decreased and shifted to higher heights in the flame with increasing dilution. Liu et al. [5] investigated the chemical effects of carbon dioxide addition to both the fuel and oxidizer stream and found that the chemical effects of CO2 are significantly higher when CO2 is added to the oxidizer stream by inhibiting pyrolysis of the fuel to acetylene and reducing flame temperatures.

Research conducted with a similar diffusion flame burner and pressure vessel by Berry Yelverton and Roberts [10] measured the effects of pressure, dilution and velocity profiles on a global measure of soot production, the smoke point. In these investigations, they found that while using a top-hat flow fuel exit velocity profile, dilution had a substantial effect on the smoke point height at atmospheric and elevated pressures, while varying the specific diluent had little effect. However, they also were able to show that while using a parabolic fuel flow exit velocity profile there was a substantial difference in smoke points between the various diluents, with carbon dioxide being a much better soot suppressant than helium, in agreement with [6], [7]. A top-hat flow exit velocity profile was used in the current work to suppress pressure oscillations and improve flame stability at elevated pressure.

In this study, major hydrocarbon species concentrations and two-dimensional temperature profiles were measured in an ethylene laminar diffusion flames with four diluents (helium, nitrogen, argon and carbon dioxide) at atmospheric and elevated pressures. The objective was to understand the influence of dilution and the type of diluent on soot precursor formation and flame temperature. The diluents were chosen such that they cover a broad range of diffusivities and heat capacities. With addition of any of these diluents to the fuel stream, there will be net change in the viscosity, which could affect the shear layer growth and entrainment. The variation in the specific heat capacities should affect the flame temperature. Each of the diluents with their difference in physical properties is also expected to have a considerable effect on properties in the flame.

Section snippets

Flame shape

Species concentrations were measured in 15 different flames, four each for argon, helium and nitrogen diluted flames and three for carbon dioxide diluted flame at pressures of 1, 2, 4 and 8 atm. However, temperature was measured via thermocouple for only 12 of these flames; pressure was limited to four atmospheres due to increased soot deposition on the thermocouple junction at higher pressures. It should be noted that the He-diluted flames are lifted in this study, due its low density

Experimental setup

A laminar jet diffusion flame burner (modeled after a Burke-Schumann [16] over ventilated flame) was used to create the ethylene–air flames. The burner was housed in a high-pressure vessel (designed by Li [17]). A full description of the pressure vessel and its operation can be found in Refs. [12], [18], [19]. The flames were established on a 4.0 mm ID fuel tube (with a 15° knife edge lip machined to minimize turbulent eddies around the fuel tube tip) with a co-flow diameter of 60.3 mm. The

Flame temperature

Fig. 1, Fig. 2, Fig. 3. show the in flame 2D temperature contour data collected using an R-type (Pt–Pt/13% Rh) thermocouple with the four diluents up to four atmospheres. The data are plotted as a function of the physical distances in the radial and axial directions. The color scales are uniform for all the four flames to easily ascertain the effects of the type of diluent and pressure on flame temperature. It should be noted that the fuel mass flux and the dilution level (82.5 vol.%) were kept

Conclusions

Major hydrocarbon species concentrations were measured in an ethylene diffusion flame diluted with four diluents (helium, argon, nitrogen and carbon dioxide) at 82.5 vol.% from one to eight atmospheres. Flame temperatures were measured up to four atmospheres in helium, argon, nitrogen at 82.5% and carbon dioxide at 78% dilution by volume. The following are notable observations and conclusions drawn from these measurements:

  • 1.

    The concentrations of ethylene increase first and then decrease with the

Acknowledgments

This material is based upon work supported by, or in part by, the U.S. Army Research Laboratory and the U.S. Army Research Office under Grant – W911NF-10-1-0118.

References (25)

  • M.L. Comstock

    Clin. Lab. Med.

    (1998)
  • W.K.C. Morgan et al.

    Ann. Occup. Hyg.

    (1997)
  • F. Liu et al.

    Combust. Flame

    (2001)
  • I.S. McLintock

    Combust. Flame

    (1968)
  • I. Glassman

    Symp. (Int.) Combust.

    (1998)
  • O.L. Gulder et al.

    Combust. Flame

    (1993)
  • C.H. Kim et al.

    Combust. Flame

    (2008)
  • L.L. McCrain et al.

    Combust. Flame

    (2005)
  • D.S. Bento et al.

    Combust. Flame

    (2006)
  • C.S. McEnally et al.

    Combust. Flame

    (1997)
  • H. Guo et al.

    Proc. Combust. Inst.

    (2002)
  • C.S. McEnally et al.

    Combust. Flame

    (2000)
  • Cited by (74)

    View all citing articles on Scopus
    View full text