• 1 view
  • No download information available

Quantifying emissions and flame geometry of lab-scale air-assisted flares

  • Author / Creator
    Mobaseri, Sina
  • Flaring has been a routine practice in the petroleum and petrochemical industry and flaring levels have remained virtually constant over the past ten years despite the efforts to reduce or eliminate the activity. Injecting an inert fluid into the combustion area in order to induce complete combustion of hydrocarbons and suppress smoke is referred to as assisting and the fluid of choice is commonly steam or air. Emissions from air-assisted flares have been studied but flame shape characteristics and the effects of operating parameters such as fuel heating value, burner size, etc. on emissions and flame geometry remain to be systematically studied. Two geometrically similar stainless steel burners with a scaling factor of 2:1 were built with a tube-in-tube design where the larger burner (2" burner) measured 50.8mm in outer tube and 25.4mm in inner tube outside diameter. Air was delivered through the inner tube and fuel flowed in the annular region between the two tubes. Propane was used as hydrocarbon fuel and it was diluted with CO2 maintaining a total flow rate of either 10 or 20SLPM with propane mixture fractions of 100, 70, 50 and 30% by volume to achieve different fuel heating values. Air flow was increased from zero up to 225SLPM or flame blow-off, whichever was achieved first. The plume of combustion products was captured through an exhaust hood above the flame and directed into a duct, downstream of which samples were drawn through a probing tube and directed to the diagnostic equipment suite for black carbon (BC), NOx and CO2 concentrations to be measured which were subsequently converted to per unit mass of fuel emission indices (EI) through a carbon-based closed mass balance technique. Digital instantaneous pictures of the flame were taken continuously at a rate of 3-4Hz and a software package was devised to process the photographs and extract flame intermittency contours. Flame length and width were defined as the height and width of the box bounding the 50% intermittency contours of the flames. Results of the 2" burner with 20SLPM pure propane as fuel showed BC emissions of 0.35g/kgFuel to remain unchanged with increasing air assist up to a mass flow ratio (MFR) of 1.3 and to decrease by multiple orders of magnitude past this point (e.g., two orders of magnitude by MFR=2.5) down to being fully suppressed by further increasing the air assist. At lower flow rates of air assist a second flame was observed to sit on the tip of the inner tube which blew off with increasing MFR. Interestingly, the blow-off point was observed to be concurrent with the onset of BC suppression. NOx emission index increased monotonously from 1.7g/kgFuel at zero assist up to 2.3g/kgFuel at MFR=8.8. Flame length went up from 74.0cm at zero assist up to 92.2cm at MFR=3.3, then decreased by further increasing assist, while flame width starting at 16.5cm decreased monotonically after inner flame blow off with assist flow rate. At about the same point where flame length decay began, a narrowing of the flame just above the burner occurred. This "neck" became narrower and closer to the burner as assist was further increased. The same overall patterns were observed with lowering fuel heating value or flow rate for emissions but fuels with lower heating values had generally lower emission indices than pure propane and lowering fuel flow resulted in BC suppression onset MFR to be delayed. Additionally, the peaking pattern was not observed in 10SLPM fuel flow cases and flame lengths were not observed to significantly increase before going down. In the smaller burner BC suppression started at a much smaller MFR of 0.3 and changing fuel flow rate did not impact the observed flame length or BC emission patterns. A dilution-corrected air assist-fuel mixture fraction was introduced and an empirical exponent of 0.2 best fit the different fuel dilutions.

  • Subjects / Keywords
  • Graduation date
    Fall 2021
  • Type of Item
  • Degree
    Master of Science
  • DOI
  • License
    This thesis is made available by the University of Alberta Libraries with permission of the copyright owner solely for non-commercial purposes. This thesis, or any portion thereof, may not otherwise be copied or reproduced without the written consent of the copyright owner, except to the extent permitted by Canadian copyright law.