Real-World Emissions from Heavy Haulers in Alberta Oil Sands Mining

Transcription

1 Real-World Emissions from Heavy Haulers in Alberta Oil Sands Mining John G. Watson 1 (John.Watson@dri.edu) Judith Chow 1, Xiaoliang Wang 1, Steven D. Kohl 1, Barbara Zielinska 1, Allan H. Legge 2, and Kevin E. Percy 3 1 Desert Research Institute, Reno, NV, U.S.A. 2 Biosphere Solutions, Calgary, Canada 3 Wood Buffalo Environmental Association, Ft. McMurray, Canada Presented at North American Oil and Gas Conference Calgary, AB, Canada October 22, 2014

2 Study Objectives Integrate modern measurement systems to quantify multipollutant emissions with multiple effects Measure real-world emission factors for gaseous and particulate emittants Obtain VOC and PM 2.5 source profiles for emission inventory development and source apportionment

3 What do we mean by multipollutant and multiple effect? Future air quality management will address more than criteria contaminants Chow, J.C.; Watson, J.G. (2011). Air quality management of multiple pollutants and multiple effects. Air Quality and Climate Change Journal, 45(3):26-32.

4 What do we man by real-world emissions? Average emission rates measured in the laboratory for engine certification don t represent the full range of equipment, fuels, and operating conditions Emission Factor [gco/kgfuel] % CO PM 3% 0% 4% 0% 2% 1% 3% 1% 2% 2% 7% 3% 12% 5% 14% 12% 18% 78% 35% Emission Factor [gpm/kgfuel] Decile Mazzoleni et al., 2004, JAWMA, 54(6):

5 Engine certification tests are performed on engine dynamometers independent of the application Sample ISO 8178 Non-Road Steady- State Test Cycle Weighing Factors Mode Number Torque, % Speed Rated speed Intermediate speed Low idle Type C1 * Factors Engines behave differently within and outside of the vehicle Engines are not subject to same aging and maintenance conditions Fuels are different Operating cycles differ, especially for transient operations

6 Diesel engines are the world s workhorses Advantages: High fuel efficiency (40-50% as compared to 25-35% for gasoline engines), more powerful, and more durable due to: Higher compression ratio Fuel-lean combustion No throttling loss Disadvantages: Heavier and more expensive Emit more NO x and PM Ideal for off-road applications! Noisier and harder to start in cold weather than gasoline engines Four-stroke Diesel Cycle: Air intake Compression Fuel injection Exhaust Rudolf Diesel ( )

7 The AOSR heavy hauler fleet operates all the time and is growing rapidly Several hundred heavy haulers operate in the AOSR Each truck operates >6000 hr/yr and consumes > 1 million L/yr of diesel fuel Surface mining operations account for >75% of diesel fuel consumption M.J. Bradley & Associates (2008)

8 Caterpillar 797B heavy haulers constitute most of the fleet Parameter CAT 797B Introduction to Service 2002 Nominal Payload Capacity 345 tonnes Gross Operating Weight 624 tonnes Engine Power 3,370 hp (2,513 kw) Displacement Liter Top Speed (Loaded) 42 mph (68 km/h) Fuel Capacity 1,800 US gal (6,814 Liter) Emission Standards Tier 1

9 Four Caterpillar 797B heavy haulers were tested in three oil sands facilities Trucks CAT 797B-1 CAT 797B-2 CAT 797B-3 CAT 797B-4 Facility B A A C Study year # of trucks in the fleet

10 Continuous and integrated samples of ~300 separate pollutants were measured Exhaust Pipe Heavy Hauler Module 1: Sample Conditioning Insulated Transfer Line CaCl 2 Desiccant Elbow Connector Water Trap Muffler Engine Exhaust Activated Air Charcoal Compressor HEPA Filter HEPA Dryer Valve 0.8 L/min Dilutor Residence time is ~3 sec Teflon Filter Flowmeter 32 L/min 32.8 L/min Residence Thermocouple Chamber Omega TJ36-CASS-116U-6-SB For Exhaust T URG ENG Cyclone 7.1 µm Cut 3 35 times dilution ratio 3.17 L/min L/min 3.75 L/min PP Systems CO2 sensors Undiluted 1 L/min Valves Background 1 L/min Pump Diluted 1 L/min 0.16 L/min Testo 350 (CO, CO 2, NO, NO 2, SO 2, O 2,T, P) PID 102+ (Total VOCs) 1 L/min 1 L/min 0.01 L/min Valve PM 2.5 impactor Filter sampler Flow meter Pump 5 L/min 5 L/min Teflon + Citric acid Pump (mass, b abs, element, isotope, NH 3) DNPH 1 liter (Carbonyls) Canister (CH 4, C2-C12) 5 L/min Nuclepore (Lichen study) Quartz + K 2CO 3 (Ions, WSOC, carbohydrates, organic acids, HULIS, SO 2) 5 L/min 5.88 L/min Makeup Flow For Balance Quartz + AgNO 3 (EC/OC, markers, H 2S) 3 L/min Valve TSI DustTrak DRX (PM 1, PM 2.5, PM 4, PM 10, PM 15) 0.7 L/min Filter 0.05 L/min TSI CPC 3007 (Concentration µm) Magee AE51 (BC) Module 2: Continuous Gas Monitoring Module 3: Integrated Sampling Module 4: Continuous PM Monitoring Module 5: Power Supply

11 Measurement systems are located in interconnected modules Sample Conditioning Module Air Compressor Valve Flowmeter Carbon Filter Continuous Gas Module Stream to Background CO 2 HEPA Filter Dryer Teflon Filter Stream to Module 2 Stream to Module 3 Stream to Module 4 Residence Chamber Cyclone Elutriator Dilution Air/Sample Mixer Dilution Air Introduction Sample Introduction Stream for Undiluted CO 2 Testo 350 CO 2 Sensors PID Analyzer Integrated Sample Module Continuous PM Module Power Supply Battery Monitor Pumps Pump for Makeup Flow Flowmeters Canister Filter Packs Computer CPC DRX OPC Deep Cycle Marine Battery Each module measures = 80 cm L 52 cm W 32 cm H; Modules 1-4 weighs ~25 kg each, and power supply module with two battery weighs 80 kg. Voltage Regulator

12 Samples were drawn from the exhaust pipe (muffler outlet) of CAT 797B and diluted to ambient temperatures Flange connecting to the body Muffler Exhaust pipe Thermocouple Sampling port Sampling modules Sample transfer line Driver cabin Sampling platform

13 Samples were taken during real-world complete operating cycles Loading Traveling with load Dumping Traveling without load

14 Fuel-based Emission Factors (EF) are determined EF P CMF fuel C CO 2 M M C CO 2 CP C CO M M C CO EF P : Emission factor in g pollutant per g fuel CMF fuel : Carbon mass fraction of the fuel (86.2% for diesel) C P : Concentration of pollutant P in g/m 3 C CO2 : Concentrations of CO 2 in g/m 3 C CO : Concentrations of CO in g/m 3 M C : Atomic weight of carbon (12 g/mole) M CO2 : Molecular weight CO 2 (44 g/mole) M CO : Molecular weight CO (28 g/mole) This equation assumes all carbon in fuel is converted to CO 2 and CO. Moosmüller et al. (2003)

15 Emission Fctor (g/kg fuel) Differences between CAT 797B trucks are within experimental variations (Higher emissions for CO, NO x, and particle number at Facility C) 1000 CAT797B-1 (Facility B) CAT797B-2 (Facility A) CAT797B-3 (Facility A) CAT797B-4 (Facility C) x CO NMHC NO x Pollutant PM 2.5 Black Carbon Particle Number NO x is expressed as NO 2

16 Real-world operations emitted higher CO and NO x, lower NMHC, and comparable PM 2.5 to certification tests Emission Factor (g/kg fuel) CAT 797B-1 CAT 797B-2 CAT 797B-3 CAT 797B-4 Certification 0.1 CO NMHC NO x PM 2.5 Emittant Real-world/Certification Ratio CO: % NMHC: 12-66% NO x : % PM 2.5 :65-112% Note: NO x is expressed as NO 2

17 CAT 797B-3 (with fuel additive) reported 60 85% higher CO and BC, 70% lower NMHC, similar NO x, 27 56% less PM 2.5 and particle number than CAT 797B-2 (no additive) Emission Fctor (g/kg fuel) 1000 CAT797B-2 (Facility A, w/o Fuel Aditive) CAT797B-3 (Facility A, w/ fuel additive) x10 15 (#/kg fuel) 2.9x CO NMHC NO x Pollutant PM 2.5 Black Carbon Particle Number The fuel additive does not have significant impact on emissions in Facility A. The two trucks tested one year apart (CAT 797B-2 in 2009 and CAT 797B-3 in 2010) reported comparable emissions.

18 Emissions (tonnes/truck/yr) Different methods used to estimate emissions for the NPRI give different results Method 1: Emission Rate = EF (g/kg fuel) Fuel consumption (Facility A) Used 1985 EPA AP-42 EFs Method 2: Emission = EF Tier1 (g/kwh) Operating Hr Rated Vehicle Power Load Factor (Facility A) Load factor is taken conservatively (59% or 100% recommended by EPA; CAT estimated 20 50%; one facility estimated 35 48%) Method 3: Emission = max(ef Tier1, EF cert composite ) 1.1/BSFC Fuel consumption (Facility B) BSFC = brake-specific fuel consumption (kg fuel/kwh) Method 1 Method 2 Method 3 Real-World CO NMHC NOx PM Emittant

19 Emissions varied by operating conditions (time series) Total VOCs (ppm) CO (ppm) CO 2 (ppm) NO x (ppm) Number Concentration (cm -3 ) Black Carbon Concentration (mg/m 3 ) PM 2.5 Concentration (mg/m 3 ) Engine Speed (rpm) Ground Speed (mph) e+8 5e+8 4e+8 3e+8 2e+8 1e+8 0 2e+2 1e (a) PID analyzer (b) Emission analyzer (c) CO 2 analyzer (d) Emission analyzer (e) CPC (f) micro-aethalometer (g) DustTrak DRX (h) Truck (i) GPS i. Idling: Concentration stable and low. ii. Leaving parking lot: All concentrations increase. iii. Top of uphill: Spikes of concentrations. iv.leaving with load: high concentration spikes when accelerating. v. Leaving after dumping: concentration spikes when climbing uphill. vi.waiting for load: low concentration except when moving forward in line. Elevation (m) 0 (j) GPS 12:35 12:40 12:45 12:50 12:55 13:00 13:05 13:10 13:15 13:20 i. Idle ii. Leaving parking lot iii. Top of uphill road iv. Leaving with load Time (hh:mm) v. Leaving after dumping vi. Waiting for load Leaving with load

20 Traveling with load has the highest emission rates NO Emission (g/s) VOCs Emission (g/s) PM 2.5 Emission (g/s) CO Emission (g/s) Total VOCs Idle Traveling w/ load Traveling w/o load Truck Operation CO Idle Traveling w/ load Traveling w/o load Truck Operation NO PM Idle Traveling w/ load Traveling w/o load 0 Idle Traveling w/ load Traveling w/o load Truck Operation Truck Operation

21 Abundance Normalized to Sum of PAMS NMHCs consist mostly of alkenes and alkanes (~10-80%) Alkanes and Cylcoalkanes Alkenes Alkyne Aromatics NMHC Group Abundance is normalized to the sum of 56 photochemical assessment monitoring station (PAMS) compounds CAT 797B-1 CAT 797B-2 CAT 797B-3 CAT 797B-4 Compound % of PAMS Ethylene 28.3% Propylene 15.4% n-heptane 5.6% Acetylene 5.4% 1-Butene 4.9% n-decane 4.5% Isobutane 4.1% Isopentane 3.7% Toluene 3.6% Ethane 3.6% Benzene 3.4% Isobutylene 2.9% n-nonane 1.9% 2-Methyl-1-Pentene 1.8% 1-Pentene 1.7% n-octane 1.4% n-butane 1.3% 1-Heptene 1.3% Propane 1.3% m/p-xylene 1.3%

22 Chemical Abundance (%) Chemical Abundance (%) EC and OC are the most abundant (~68-88%) components in PM 2.5 source profiles; Elevated P, S, Ca, and Zn suggest the origin from lube oil. 100 OC EC CAT 797B-1 10 NO NO 2 - PO 4 SO 4 = + NH 4 Ca ++ Na+ Mg ++ Al Si P S Cl K Ca Sc Ti Cu Fe Zn Ga Sr Zr Mo Nb Ag Sn Sb Ba La Ce Sm Eu Ir Au Pb Ur Chemical Species Soluble ions 2% Elements 1% Unidentified 9% OC 21% Soluble ions 4% Elements 2% Unidentified 9% OC 36% Soluble ions 3% Unidentified 27% OC 14% Soluble ions 8% Elements 2% Unidentified 13% OC 21% CAT 797B-1 EC 67% EC 49% CAT 797B-2 Elements 1% CAT 797B-3 EC 55% CAT 797B-4 EC 56%

23 phenanthrene anthracene fluoranthene pyrene benzo[a]anthracene chrysene+ triphenylene benzo[b]fluoranthene benzo[j+k]fluoranthene benzo[a]fluoranthene benzo[e]pyrene benzo[a]pyrene perylene indeno[1,2,3-cd]pyrene dibenzo[a,h]anthracene benzo[ghi]perylene coronene dibenzo[a,e]pyrene 9-fluorenone dibenzothiophene 1 methyl phenanthrene 2 methyl phenanthrene 3,6 dimethyl phenanthrene methylfluoranthene retene benzo(ghi)fluoranthene benzo(c)phenanthrene benzo(b)naphtho[1,2- cyclopenta[cd]pyrene Mass Percentage of Organic Carbon (%) Particle-bound PAHs show two distinct abundance groups CAT 797B-1 CAT 797B-2 CAT 797B-3 CAT 797B PAHs

24 n-pentadecane (n-c15) n-hexadecane (n-c16) n-heptadecane (n-c17) n-octadecane (n-c18) n-nonadecane (n-c19) n-icosane (n-c20) n-heneicosane (n-c21) n-docosane (n-c22) n-tricosane (n-c23) n-tetracosane (n-c24) n-pentacosane (n-c25) n-hexacosane (n-c26) n-heptacosane (n-c27) n-octacosane (n-c28) n-nonacosane (n-c29) n-triacontane (n-c30) n-hentriacotane (n-c31) n-dotriacontane (n-c32) n-tritriactotane (n-c33) n-tetratriactoane (n-c34) n-pentatriacontane (n-c35) n-hexatriacontane (n-c36) n-heptatriacontane (n-c37) n-octatriacontane (n-c38) n-nonatriacontane (n-c39) n-tetracontane (n-c40) ** Mass Percentage of Organic Carbon (%) Particle-bound n-alkanes show abundances between C19 and C CAT 797B-1 CAT 797B-2 CAT 797B-3 CAT 797B n-alkanes

25 Summary Real-world emissions were characterized for four CAT 797Bs under normal operations. Real-world operations emitted higher CO and NO x, lower NMHC, and comparable PM 2.5 to certification tests. Emission estimation methods differ by facility and from the real-world measurements Truck emission source profiles showed 10 80% alkenes and alkanes in non-methane hydrocarbons, and 68 88% carbon (OC and EC) in PM 2.5.

26 How to use this information Criteria Contaminants Develop a more accurate and consistent estimate of fleet emissions, based on distributions of load factors and fuel consumption available from tracking systems Greenhouse Gases Estimate mine fleet Global Warming Potentials (GWP) for life-cycle assessments Toxics Inventory Revise emission estimates Source Apportionment Use chemical profiles to identify contributions to adverse ecosystem responses Anticipate future emission requirements Ultrafine particles and black carbon (EU and California are moving in this direction)

27 Future engines will substantially reduce emissions as they penetrate the fleet Emission Factor (g/kg fuel) 797B/T282B journal CO NMHC NOx PM 0.1 Base (<1988) T0 ( ) T1 ( ) T2 ( ) Truck Tiers (Model Year) T4 ( )

28 Acknowledgements Sponsor: Wood Buffalo Environmental Association (WBEA, Ft. McMurray, Alberta). The content and opinions expressed by the authors in this presentation do not necessarily reflect the views of WBEA or of the WBEA membership Special appreciation to: Drs. Ken Foster and Yu-Mei Hsu, Ms. Carna MacEachern, Ms. Veronica Chisholm, and environmental officers and staffs at each facility