Reforming landfill gas to syngas Marco J. Castaldi Department of Earth & Environmental Engineering Henry Krumb School of Mines, Columbia University October 19, 2006 Waste-to-Energy Research and Technology Council October 19-20, 2006 New York, NY USA
Outline Background / Motivation Experimental Results Pre-reduced vs As-prepared Kinetic parameter estimation BET and EDX Future work Conclusions
Biogas: Landfill Gas (LFG) Facts about landfills and landfill gas. Landfill gas is created when waste in a landfill decomposes. 50% CH 4 & 50% CO 2, Oxygen, nitrogen, and hydrogen, and trace amounts of hazardous air pollutants (less than 0.2%). Largest anthropogenic (human-made) source of methane in the United States. Combustion of landfill gas in an engine, turbine, boiler significantly reduces odors and emissions of methane and hazardous air pollutants.
Background LFG = Landfill Gas - 50 % CO 2, 50 % CH 4 LFGE = Landfill Gas to Energy Electricity generation on site sell to the grid Cogeneration Thermal applications boiler, kiln, greenhouse, etc. Treatment of leachate Convert it to synthesis gas (H 2 and CO) which can be used: As a feed gas for a fuel cell Enhanced combustion processes Possible Chemical Synthesis (methanol, acetic acid, or hydrocarbons)
Background Steam Reforming CH 4 + H 2 O CO + 3 H 2 ΔH = + 226 kj/mol Partial Oxidation of Methane CH 4 + ½ O 2 CO + 2 H 2 ΔH = - 44 kj/mol Problem: At the stoichiometric CH 4 -to-o 2 ratio there is significant carbon formation on many catalysts Dry Reforming CH 4 + CO 2 2 CO + 2 H 2 ΔH = + 261 kj/mol
H 2 Spiking Data ~15% H 2 added to a flame Typical boiler combustor flame medium BTU gas (300 BTU/ft 3 ) Relative Concentration Reductions Reduces NO significantly 1.20 1.00 0.80 0.60 0.40 0.20 0.00 UHC CO Soot Precursors w/o H 2 with H 2 Reduced CH radical concentration Leaner flammability limit lower thermal NO
H 2 Spiking for Lower Emissions Lower NO x, CO and HC: NO x emissions as low as 0.11 g/hp-hr NO x reduction > 80% in a gasoline fueled SI engine Experiment w. ~15% H 2 added to a flame Flame represents process-heating combustor Low BTU refinery fuel gas used (~300 BTU/ft 3 ) Reduced GHG Eliminates LFG purging and flaring Relative Concentration Reductions 1.20 1.00 0.80 0.60 0.40 0.20 0.00 UHC CO Soot Precursors w/o H 2 with H 2 Advantages of H 2 in natural gas engines well documented: L.Bromberg, D.R.Cohn, A.Rabinovich and J.Heywood Emissions Reductions Using Hydrogen from Plasmatron Fuel Converters at Diesel Engine Emission Reduction Workshop, San Diego, CA, Aug 2000 Hydrogen Consultants, Inc. of Littleton, Colorado, has also demonstrated that methane blended with only 5% hydrogen (Hythane ) can result in NOx and CO reduction of up to 50%.
Emissions Benefits Reduction in CO and HC. Improvements in lean limit stability NO x emissions as low as 0.11 g/hp-hr over a range of equiv. ratios NO x reduction > 80% in a gasoline fueled SI engine Advantages of H 2 in natural gas engines well documented: R.Breashes, H.Cotrill and J.Rupe, Partial Hydrogen Injection into Internal Combustion Engines Effect on Emissions and Fuel Economy First Sympos. on Low Pollution Power Systems Development, Ann Arbor MI, 1973 J.Houseman and F.W.Hohn, A Two Charge Engine Concept: Hydrogen Enrichment, SAE 741169 (1974) S.R.Bell and M.Gupta, Extension of the Lean Operation Limit for Natural Gas Fueling of a Spark Ignited Engine Using Hydrogen Blending. Combust. Sci. and Tech., 123 pp. 23-48 1997 F.E Lynch and R.W.Marmaro; US Patent 5,139,002, 1992 L.Bromberg, D.R.Cohn, A.Rabinovich and J.Heywood Emissions Reductions Using Hydrogen from Plasmatron Fuel Converters at Diesel Engine Emission Reduction Workshop, San Diego, CA, August 20-24, 2000 Hydrogen Consultants, Inc. of Littleton, Colorado, has also demonstrated that methane blended with only 5% hydrogen (Hythane ) can result in NOx and CO reduction of up to 50%.
H 2 Generation Approach 2 reaction pathways Preferred: Undesirable: CH 4 + 1/2 O 2 CH 4 + 2O 2 Catalytic Partial Ox Deep Ox 2H 2 + CO CO 2 + 2H 2 O Novel catalytic reactor design identified extremely compact, efficient & lightweight Selective catalysts Low operating temp Non-exotic materials of construction Robust, durable, stable reactor Low temp gradients Net effect: Temp rise; Reactive partial ox. products (CO, H 2 )
Catalyst Attributes Activity Initiate or speed up reaction w/o being consumed. However, catalyst may undergo changes during process Energy E act no catalyst Reactants ΔH E act w. catalyst Products Reaction Path Catalyst A CO 2 + H 2 O Selectivity Ability to speed up desired rxn CH 4 + O 2 Catalyst B H 2 + CO Fuel to air ratio affects catalyst selectivity
Background Combustion benefits from H 2 Flame stability at low φ Ignition at low φ Low temperature operability Emission benefits Reduced GHG Lower NOx, CO, UHC LFG Reformer CO, H 2 Combustion Chamber H O, 2 CO 2
System Implementation Schematic Air LFG Bleed LFG Bleed Air CPOX reactor Secondary fuel H 2,CO Flow Splitter/mixer Burner IC Engine Gas Turbine Heat, CO 2, Steam, Energy More reactive system; Reduce/eliminate secondary fuel usage
Lower Ignition Temperature for Stability Assumes 47% CH 4 in LFG ( ~450 BTU/ft 3 ) Total flow = 250 ft 3 /sec ( 93 ft 3 /min CH 4 ) Secondary air ~ 80% of total flow (850 ft 3 /min) ~140 ft 3 /min LFG 100 ft 3 /min Air Air: ~850 ft 3 /min LFG : ~250 ft 3 /min Ignition at Air: ~850 ft 3 /min LFG : ~250 ft 3 /min Ignition at T> 900 o C Burner Burner IC Engine T> 700 o C IC Engine Gas Turbine Gas Turbine + secondary fuel Air: ~850 ft 3 /min LFG : ~250 ft 3 /min CPOX reactor Ignition at T~ 300 o C T adiabatic ~ 1280 o C 26% H 2 20% CO ΔT~ 675 o C Burner IC Engine Gas Turbine 10% H 2 7.7% CO 15% CH 4 Requires lower net energy & ignition temp. CH 4 fluctuation permissible
Combustion Enhancements Reformed LFG or biogas Better combustion performance Reduced emissions Little or no engine modifications CH 4 + ½ O 2 CO + H 2 Use the H 2 to enhance the combustion
Combustion Benefits from H 2 Measure of Turbulence Above the line - unstable below the line - stable H 2 addition increases reactivity, extending stability/blowout limits Calculation, GRI 2.1 mechanism for CH 4 @ 1 atm, 400 C inlet; Cong & Jackson (1999)
Combustion Benefits from H 2 Measure of Turbulence 10% H 2 yields stability at twice the turbulence levels (reduced ignition delay times)
CO 2 as a potential feedstock.
Equilibrium Calculation 0.14 CH 4 + CO 2 + N 2 H 2 + CO + N 2 + H 2 O(g) + C(s) + CO 2 + CH 4 0.12 H 2 Mole fraction 0.1 0.08 0.06 C(s) CO 0.04 0.02 0 0 200 400 600 800 1000 Temperature (Celsius)
Experimental Setup Heated tubing (120 o C) Micro-GC Const. Temperature Water circulation Calibrated RM Rotometers RM RM 80ml/min N 2 CO 2 CH 4 20ml/min Certified gases (pure and mixtures)
Pretreatment - Results Mass (%) 120 115 110 105 100 95 Abrupt change followed by increased rate of mass gain (carbon deposition) As-prepared Pre-reduced 0 50 100 150 200 Time (minutes) 1000 900 800 700 600 500 400 300 200 100 0 Temperature (degrees Celsius)
Pretreatment - Results H2 concentration (ppm) 14000 12000 10000 8000 6000 4000 2000 0 1000 900 800 700 600 Pre-reduced 500 As-prepared 400 300 200 100 0 0 50 100 150 200 Time (min) Temperature (degrees Celsius) The reduced catalyst shows a higher production of both H 2 and CO
Various Temperatures - Results 120 115 110 Mass (%) 105 100 900 degree C run 700 degree C run 500 degree C run 95 14000 90 0 50 100 150 200 12000 Time (min) Higher mass gain for 900 C Yet high activity H2 concentration (ppm) 10000 8000 6000 4000 2000 900 degree C run 700 degree C run 500 degree C run 0 0 50 100 150 200 Time (min)
14000 1000 H2 Concentration (ppm) 12000 10000 8000 6000 4000 2000 pre-reduced as-prepared Al 2 O 3 800 600 400 200 Temperature ( o C) 0 0 50 100 150 Time (min) 0
Ramp - Soak Results Dry Reforming With Pt at 500 o C, 700 o C, and 900 o C Consecutively 103 1000 102 800 mass (%) 101 100 99 600 400 temperature ( o C) 98 200 97 0 0 20 40 60 80 100 120 140 time (min)
Ramp - Soak Results H 2 Production at 500 o C, 700 o C, 900 o C Consecutively 6000 1000 H2 concentration (ppm) 5000 4000 3000 2000 1000 800 600 400 200 temperature ( o C) 0 0 0 50 100 150 time (min)
Various CO 2 Amounts 700 o C H2 production for dry reforming at 700 deg C with various CO2 concentrations 12000 800 concentration (ppm) 10000 8000 6000 4000 2000 1 x CO2 2 x CO2 700 600 500 400 300 200 100 Temperature (deg C) 0 0 50 100 150 200 0 time (minutes)
BET Analysis Sample Description Fresh Al 2 O 3 support Fresh Pt catalyst DR at 900 o C, As-prepared DR at 900 o C, reduction at 900 o C Surface Area (m 2 /g) 188 152 159 163 Active site (Pt metal) surface may be decreasing
EDX Results DR temp = 900, no reduction DR temp = 900, no reduction DR temp/red temp = 900 o C DR temp/red temp = 900 o C DR temp/red temp = 700 o C DR temp/red temp = 500 o C C C C C C C 19 % 24 % 28 % 22 % 10 % 4 % Questions Not reducing = less carbon deposition more carbon deposition at higher temperatures = thermal breakdown of CH 4 : CH 4 2 H 2 + C
Kinetic Analysis Heating rate = β =20 K/min mass (%) 116 114 112 110 108 106 104 102 100 98 96 Dry Reforming with Pt 0 50 100 150 200 1000 800 600 400 200 0 Isothermal temperature ( o C) Reactant Deposition time (min) Carbon Deposition
Kinetic Analysis: Basic Equations All kinetic models have same basic equations: dα/dt = k f(α) k = A exp (E/RT) dα/dt = k (1-α) n Conversion factor, α: decomposition: α = w i w / w i -w f deposition: α = w w i / w f w i w i = initial weight w f = final weight w = weight at some time t α = conversion factor A = pre-exponential factor E = apparent activation energy T = absolute temperature n = apparent order of reaction
Kinetic Analysis Coats-Redfern Method Heating rate, β: β = dt/dt β = dα/dt = A exp(-e/rt)(1-α) n Separation of variables, integrate, take log of each side: For n 1 log {(1-(1-α)1-n)/(T2(1-n))} = log AR/βE [1-2RT/E] - E/2.3RT For n =1 log {(-log(1-α))/t2} = log AR/βE [1 2RT/E] E/2.3RT Both are in the form y = mx + b: n 1: y = log {(1-(1-α)1-n)/(T2(1-n))} n = 1: y = log {-log(1-α)/t2} x = 1/T m = -E/2.3 R b = log AR/βE [1-2RT/E]
Kinetic Analysis Carbon Deposition Sample Description n E (kcal/mol) A (min -1 ) DR at 900 o C, reduction at 900 o C, double CO 2 2 7.96 0.25*10 12 DR at 700 o C, reduction at 700 o C, double CO 2 2 60.70 2.19*10 12
Conclusions Higher temperatures yields higher mass gains Higher temperatures yields more H 2 /CO No real significant change in BET Active sites may be changing EDX results show correct trends in terms of C Verified by subsequent burn-off experiments Kinetic parameters calculated Model development Estimation of Durability long term deactivation Flow reactor experimentation beginning Durability long term deactivation No appreciable performance on Au / Al 2 O 3
Acknowledgements National Science Foundation CTS SGER grant Tracy Jackson Noah Whitmore Thank You for listening
Various CO 2 Amounts 900 o C Mass (%) 125 120 115 110 105 100 95 90 4 x CO2 0 50 100 150 200 Time (min) 2 x CO2 1 x CO2 1000 900 800 700 600 500 400 300 200 100 0 Temperature (degrees Celsius)
Various CO 2 Amounts 900 o C H 2 Production for Dry Reforming at 900 o C With Various CO 2 Concentrations concentration (ppm) 20000 16000 12000 8000 4000 0 baseline 0 50 100 150 200 time (min) 4 x CO 2 1 x CO 2 2 x CO 2 1000 900 800 700 600 500 400 300 200 100 0 Temperature (deg C)