Estimating maximum steam assist rate for arbitrary flare gas mixtures

CanmetENERGY L e a d e r s h ip in e c o ln n o v a tio n Estimating Maximum Steam Assist Rate for Arbitrary Flare Gas Mixtures P. Gogolek 2011 AFRC Combustion Symposium Natural Resources Canada Ressources naturelles Canada Canada Motivation • Need analytical approach to controlling steam-assist rate. • Heat content fails to take account of hydrogen effect, differences between nitrogen and carbon dioxide. • Mass ratio of steam to flare gas (SFR) did not seem to be an effective correlating factor. CanmetENERGY r L e a d e r s h ip in e c o ln n o v a tio n ■ 1 t | Natural Resources Canada Ressources naturelles Canada Canada Basic Argument • Steam can snuff the flare. • At some rate of steam-assist lower than the snuff rate, there is a drop in the Combustion Efficiency (CE) and in the Destruction Efficiency (DE). • Can we use the FL information to develop a correlating factor. CanmetENERGY r L e a d e r s h ip in e c o ln n o v a tio n B ^ B 1 t B Natural Resources Canada Ressources naturelles Canada Canada Flammability for Methane with Dilution METHANE, PERCENT OXYGEN IN ORIGINAL ATMOSPHERE, PERCENT • Taken from Coward and Jones [1952] • Effect of H20 is between that of N2 and C02 (l*,F*) nmetENERGY Calculation Method ( ) 1 • Data are taken from report of Zabetakis • Critical points for steam dilution are interpolated from those for C02 and N2. K l0 = 0.7Fc*ft + 0.3F ; • Flammability limit is p j^ _ p * + j * \ • Define Steam Capacity as \ \ I* S = -- CanmetENERGY r L e a d e r s h ip in e c o ln n o v a tio n ■ 1 t M Natural Resources Canada Ressources naturelles Canada Canada Tabulated Data PG Shell FL* H 20 d H 20 FL* H 20 dH 20 H2 13.87 60.6 12.04 62.6% CH4 5.48 37.3 3.95 32.5% CO 2.50 59.5 2.59 58.7% E th yle n e 12.04 44.7 12.00 38.8% E th a n e 8.43 36.1 8.73 37.6% A c e ty le n e 14.44 56.4 20.63 56.2% P ro p a n e 9.24 33.5 7.84 34.4% P ro p yle n e 9.32 34.3 9.83 33.1% Iso -B u ta n e 10.45 26.8 11.35 31.5% N -B u ta n e 10.45 26.8 11.35 31.5% B u tenes 12.25 31.6 12.27 32.9% B u ta d ie n e 13.12 35.0 12.27 32.9% C 5+ 10.82 28.1 21.83 31.1% • These data are compiled from Zabetakis plots ("tip of the nose") CanmetENERGY L e a d e r s h ip in e c o ln n o v a tio n Natural Resources Canada Ressources naturelles Canada Canada Calculation Method ( ) 2 Flammability limits are combined according to LeChatelier' s Principle 1 FLr m -1 FL, Introduce mixing rule for Steam Capacity 3 w -1 l l Algebra gives the critical steam volume fraction s Ir * m X F L m m 1 + <5m CanmetENERGY r L e a d e r s h ip in e c o ln n o v a tio n B ^ B 1 t B Natural Resources Canada Ressources naturelles Canada Canada Results - Marathon TXC M arathon TXC v2 r 1 .0 0 ■ 1 1 0 .9 0 A 0 .8 0 SP ■ A 19 ,r0 .7 0 u A A ll A ■ A- ■ A8 i A ♦ B 0 .6 0 ■ C D 0 .5 0 E - F G 0 .4 0 0 0 .2 0 .4 0 .6 1 .2 0 .8 RSVF. M Natural Resources Canada Ressources naturelles Canada w V y C l f ' l 0 d l Id /" J L ld . Results - Marathon TXC again Marathon TXC v2' 1 .0 0 Using Shell values 0 .9 0 0 .8 0 CE,% ▲ * ■ A19 0 .7 0 A A ll > ■ A8 I A ♦ B 0 .6 0 ■ C D 0 .5 0 E - F G 0 .4 0 0 0 .2 0.4 0 .6 RSVF. 0 .8 1 .2 IGY ^ In n o v a tio n Natural Resources Canada Ressources naturelles Canada Canada Results - Marathon Detroit M arathon Detroit v2 1 .0 0 ■ a ■ ■ 0.98 ■ ■ 0.96 ♦ ♦ ♦ 0.94 A ■ aa CE, % 0.92 ■ 0.90 ■ \ ■ ■ 0 .8 8 ■ * ♦ A 0 .8 6 ■ B a 0.84 r ■ ■ D 0.82 ■ ■ E 0.80 1 0 0 .2 1 0.4 0 .6 0 .8 ♦ 1 i GY 1 .2 RSFV, - M Natural Resources Canada Ressources naturelles Canada m o v a tio n { V y C l f ' l 0 d l Id /" J L ld . Results - Flint Hills Austin Flint Hills AU v2 1 0 0 A M i l * ■ 98 ■ 96 ♦ 94 A* 92 ♦ \P os 90 U 8 8 8 6 ♦ AU-A ■ AU-B 84 A AU-C 82 ■ AU-D 80 I 0 0 .1 0 .2 0 .3 0 .4 0 .5 0 .6 0 .7 0 .8 0 .9 RSVF, i ^ Natural Resources i Ressources naturelles Canada Results - Flint Hills Louisiana Flint Hills LOU v2 1 0 0 ♦ 95 A 90 * VP ov 85 U 80 ♦ LOU-A ■ LOU-B 75 A LOU-C 70 I 0 0 .2 0.4 0 .6 RSVF, l^ l Natural Resources Ressources naturelles 0 .8 1 .2 'a t i o n Canada Results - TCEQ 100 II---------------- £* • -m ^ 90 0 X * ♦ x- x ■ . 80 X X 70 o ♦ + + + X 7S 60 P 50 LU o 40 30 20 10 ■ S3.1 S4.1 > S5.1 ♦ S6.1 XS7.1 • S8.1 + S9.1 -S10.1 - S11.1 OS12.1 S13.1 A S14.1 ♦ 1C X 0 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 GY 0.9 RSVF M Natural Resources Canada Ressources naturelles Canada novation Cl f 'l Cl /"J { V y d l Id L ld . Weaknesses • Effect of water on FL is estimated based on a single example (methane). • Unlikely that a single number can represent the effects of mixing for different tip designs. • Needs real time analysis for control. CanmetENERGY r Leadership in ecolnnovation ■ 1 t | Natural Resources Canada Ressources naturelles Canada Canada Strengths • Values are tabulated for most compounds sent to flares. • and can estimate the rest? • No special treatment of hydrogen or propylene. • Not sensitive to (reasonable) variation of the pure component data (l*,F*) CanmetENERGY r Leadership in ecolnnovation ■ 1 t | Natural Resources Canada Ressources naturelles Canada Canada Way forward • Look for simple way to incorporate difference between centre steam and upper steam. • Look at requirements for supplemental gas on cost/benefit basis. • Look for simplified analysis to use for real time control. CanmetENERGY r Leadership in ecolnnovation B ^ B 1 t B Natural Resources Canada Ressources naturelles Canada Canada Conclusion • FL can be used as a mixture property to explain gross behaviour of steam-assisted flares. • Only one ‘fudge factor' to be applied. • can elaborate for different tip designs. • Should be robust to new data sets for different gas mixtures and similar tip designs. CanmetENERGY r Leadership in ecolnnovation ■ 1 t | Natural Resources Canada Ressources naturelles Canada Canada Acknowledgements • Funding for this work provided by the PERD project "Burners and Sensors for Opportunity Fuels" • Concept of RSVF developed during the IFC, support of the members gratefully acknowledged. • Numerous discussions with GM, BD, ZK, SE, RC, and so many more. CanmetENERGY r Leadership in ecolnnovation B ^ B 1 t B Natural Resources Canada Ressources naturelles Canada Canada CMA Study - steaming propylene SFR of 3.5 is drop-off for these tests CanmetENERGY r Leadership in ecolnnovation ■ 1 t | Natural Resources Canada Ressources naturelles Canada Canada