Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
21 5.1 APEX1 The emissions of a CFM56-2C1 engine burning a range of fuels (base, high aromatic, high sulfur) were measured during the APEX1 campaign. The PM parameters measured at APEX1 were particle size distributions and emission indices for parti- cle number and mass (i.e., the number or mass, respectively, of particles produced per kilogram of fuel burned), including their volatile and non-volatile fractions. Emission indices were also measured for major combustion gases (carbon dioxide [CO2], carbon monoxide [CO], nitrogen oxides [NOx], total unburned hydrocarbons [UHC]) and trace combustion gases (specifically, sulfur dioxide [SO2], nitrous oxide [N2O], nitrous acid [HONO], and a number of volatile HC). All of the data recorded in the APEX1 database were acquired when the engine was declared to be stable for a given operating condition. The PM in an engine exhaust plume was found to vary in composition and physical and chemical properties as the plume ages, and the total PM detected had volatile and non- volatile (black carbon) components depending on the sampling location in the plume. In terms of non-volatile particle emis- sions, the following conclusions were drawn from APEX1: ⢠Non-volatile particles ranged in diameter from smaller than 10 to 300 nm (i.e. 0.01 to 0.3 μm); see Figure 1 for size comparisons. ⢠The number mean diameter of the particles increased with thrust ranging from ~15 nm at idle to ~40 nm at take off ⢠For the three fuels tested, the non-volatile PM parameters did not vary. ⢠Non-volatile PM parameters did not depend on plume age (sampling distance downstream of the exhaust nozzle), in- dicating that the black carbon component of the exhaust does not change as the plume ages. ⢠The number-based emission index was highest at takeoff thrust, with a smaller peak at idle thrust, and revealed a minimum at thrust levels corresponding to approach. ⢠The number-based emission index at low thrust levels was found to decrease during the first couple of hours of engine on time. Number-based emission index similarly decreased as the ambient air temperature increased. ⢠The mass-based emission index increased with thrust, ranging from <20 mg/kg of fuel burned at idle through cruise thrust levels and rising to >200 mg/kg of fuel burned at takeoff. Samples collected downstream of the exhaust nozzle often contained large numbers of volatile particles that contain materials that are gases at temperatures above 300°C (572°F). These volatile particles were not observed at the exhaust nozzle but were readily apparent downstream (~30 m [98 ft]). They evolve as the plume expands and mixes with the ambient air. In terms of volatile particles, the following conclusions were drawn from APEX1: ⢠Their number mean diameter ranged from smaller than 3 nm to 10 nm. ⢠At downstream sampling locations, the number-based volatile particle emission index was typically much higher than that of the black carbon and depended on fuel composition, thrust level, plume age, and ambient temperature. ⢠Compositional analysis of these volatile particles revealed that sulfur and HC species accounted for a significant fraction of the volatile mass, consistent with condensa- tion and nucleation of sulfuric acid/sulfate and HC in the cooling plume. In the case of the total PM, where no distinction is made between the volatile and non-volatile components, the fol- lowing characteristics were observed: ⢠At high thrust levels, particle mass emissions were domi- nated by black carbon at all sampling locations in the plume. C H A P T E R 5 Individual Reviews of Data from the Aircraft Field Measurement Campaigns
⢠At low thrust levels, the number-based emission index was substantially greater downstream (~30 m [98 ft]) than at the exhaust nozzle, indicating that significant gas- to-particle conversion occurred as the plume cooled and aged. ⢠The number mean diameter of the total particles increased linearly with thrust. From the gas-phase species measurements performed during APEX, the following conclusions were drawn: ⢠The emission indices for NOX, CO, and HC agreed with values archived in the ICAO Aircraft Engine Emissions DataBank. ⢠The NO2 fraction of NOX varied from ~0.7 at idle to ~0.09 at take off. ⢠Although substantial at idle, the HC emission index de- creased with increasing thrust and was below the minimum detection limit (roughly 0.01 to 0.05 g/kg-fuel depending on the species) above 15% rated thrust. ⢠The HC emission index depended strongly on ambient conditions such as temperature. A 20°C decrease in ambient temperature increased the emission index of HC species by a factor of 10. ⢠The emission index of SO2 was greater for the high sulfur fuel (1600 ppm sulfur) than for either the high aromatic or base fuels (400 ppm sulfur). ⢠Unburned hydrocarbons are emitted as a variety of com- pounds, including ethylene, formaldehyde, acetaldehyde, and benzene. Emissions of the various HC species rise and fall with one another, regardless of engine type or thrust setting. Even when the absolute magnitudes increase by a factor of 10 or more (as is the case for older engine tech- nology or for operation at low thrust condition or low ambient temperature), the ratio of one HC species to the next remains constant. ⢠Slight differences in formaldehyde emission index were observed between the various fuel types. 5.2 Delta Atlanta-Hartsfield Study The second of the APEX series of studies was carried out with the support of Delta Airlines at Hartsfield-Jackson Atlanta International Airport in September 2004. Mobile laboratories operated by Missouri University of Science and Technology (Missouri S&T), Aerodyne Research, Inc. (ARI), and NOAA were deployed to conduct both dedicated engine tests and runway studies (see Chapter 3 for details). The full LTO cycle for MD-88 and JT8D engines was studied during the Delta Atlanta-Hartsfield Study. Only thrust settings less than 60% full rated thrust were examined for larger engines such as CF6 and PW2037. The Delta Atlanta-Hartsfield Study yielded the following conclusions from the extractive sampling measurements: ⢠For the JT8D engines, number mean particle diameters in- creased with engine thrust. ⢠The number-based emission index was highest at takeoff, exhibited a smaller peak at idle, and revealed a minimum at thrust levels corresponding to approach. ⢠The mass-based emission index behaves similarly to the number-based emission index and is higher at idle, exhibits a minimum at approach, and then rapidly increases to a maximum at takeoff. ⢠The JT8D number and mass emissions trends are consis- tent with behavior of the CFM56-2C1 engine studied in the APEX campaign. ⢠The two JT8D engines in this study have greater black carbon emission indices than any other engine tested in the APEX studies. ⢠The CF6 and PW2037 have a greater number-based emis- sion index than the JT8D. At a given thrust condition and for a given fuel, the JT8D emits fewer but larger particles while engines designed to reduce smoke number certifica- tion measurements (those more recently developed) emit more numerous quantities of smaller particles. ⢠For thrust conditions near idle, the amount of volatile organic PM emitted by JT8D-219 engines decreases rapidly with increasing thrust, consistent with the thrust dependence of the emission index of UHC. This observation is consis- tent with nucleation/condensation of the least volatile UHC to form organic PM. The LIDAR analysis gave similar qualitative trends for all of the engines studied, but the reliability of the system employed in this study was judged to be low for quantitative measurements. Upon completion of the dedicated engine testing, the extractive sampling systems were positioned downwind of active runways. Exhaust plumes transported from source air- craft by the prevailing winds were continuously sampled and the source aircraft tail numbers logged. The tail numbers provided a unique method for correlating the plumes with spe- cific aircraft and, hence, specific engines. In excess of 500 taxi and takeoff events were sampled during a three-day period. The following general conclusions from the wind-blown plume analyses can be drawn: ⢠The combination of the PM and gas analysis of the trans- ported plume provides unique identification of the engine operating condition generating the plume (i.e., idle, spool- up, maximum thrust, etc.). ⢠Much more volatile material converts to the particle phase during plume transport across the runway than is observed during dedicated engine tests. 22
⢠The black carbon component of the PM emissions detected in the transported plume varies among engine types as is observed at the exhaust nozzle. ⢠For all plumes sampled, the number-based emission index ranged from 3 à 1016 to 2 à 1017 particles/kg fuel and the mass-based emission index ranged from 0.1 to 0.35 g/kg fuel. These averages, based on measurements of PM emissions from in-service aircraft during normal operating conditions, give credence to the rough averages reported in the Intergovernmental Panel on Climate Change (IPCC) report. (Penner et al. 1999) 5.3 JETS-APEX2 The objective of the next APEX study, JETS-APEX2, was to develop TOG and PM speciation profiles for engines used in newer Boeing 737-type commercial aircraft burning standard Jet A fuel. These aircraft were specifically chosen since they represent greater than 70% of the aircraft currently in opera- tion in the domestic commercial fleet. The JETS-APEX2 study aimed to produce a comprehen- sive data set of emission factors for TOG and PM for older (CFM56-3) and newer (CFM56-7) CFM56-class engines. The study was successful in producing the first state-of-the-art measurements for PM physical characterization of in-service CFM56-type engines. The major conclusions from the JETS-APEX2 study are as follows: ⢠The qualitative emissions trends observed are consistent with those measured for the CFM56-2C1 engine studied in the APEX1 campaign. ⢠At takeoff, the mass-based emission index for the -7B engines was significantly lower than that for the older tech- nology -3B and -2C1 engines. At takeoff (typically 85% rated thrust at OAK), the -7B mass-based emission index was found to be four times less than that of the -3B. ⢠NOX measurements were in good agreement with ICAO certification data, indicating that the engines were in good condition and lending credence to the assumption that the concomitant PM emissions are also representative of an engine in good condition. ⢠Most individual HC species decrease with increasing thrust in proportion to each other. As one of the most plentiful emitted hydrocarbons, formaldehyde is easily measured and provides a good standard for comparing the emissions of other less abundant trace hydrocarbons. ⢠The emission index of SO2 increases directly with fuel sul- fur content. ⢠Volatile particles are composed of both sulfate and organic materials, adding to the carbonaceous aerosol that is present already at the exhaust nozzle. ⢠The amount of sulfate emitted by the CFM56 engines increased with fuel sulfur content, as did the number of par- ticles formed in the diluting exhaust gas in the 30-m (98-ft) and 50-m (164-ft) samples. ⢠The CFM56 engine exhaust did not contain substantial quantities of lubrication oil in the exhaust, even at high thrust. Lubrication oil contributed at most about 3mg/kg to the overall quantity of emitted organic PM. ⢠A significant fraction of the organic PM contained in CFM56 exhaust appears to be UHC. Unburned hydrocar- bons constitute as much as 80% of the CFM56 organic PM emissions at idle and as much as 70% at climb-out/takeoff. ⢠The measured ratio of sulfate to organic PM was greater at high thrust (climb-out and takeoff) than at low thrust (approach and idle) by a factor of at least three. The observed thrust dependence is consistent with combustor efficiency (and, therefore, HC emissions) being more sensitive to thrust condition than conversion of fuel sulfur to condensible species (i.e., SO3). ⢠The data from APEX1, Delta Atlanta-Hartsfield, and JETS-APEX2 indicate that PM emissions depend on engine/airframe. Upon the completion of this study, the following recom- mendations for future aircraft emission characterization tests were made: ⢠Emissions studies of wider range of engines/airframes should now be performed (e.g., B747/CF6-80, etc.). ⢠The ideal testing conditions afforded by the GRE at OAK recommends the use of the GRE for future tests. ⢠Since the mix of transports routinely operating in and out of OAK will limit the range of engines/airframes that can be studied, for future studies where B747, B757, B767, and B777 and the larger Airbus transports A320, A340, etc. are anticipated test vehicles, it will be necessary to consider GREs located at other airports. ⢠In future tests, engine operating data (e.g., N1âfan rotor speed [rpm], N2âcore rotor speed [rpm], EGTâexhaust gas temperature, fuel flow rate) should be recorded to facilitate interpretation of emissions data. Ideally, engine operating data will be recorded at high-frequency and made available in real time. Recording of engine data may be difficult for older airframes, but straightforward for newer additions to the commercial fleet that digitally record engine operating conditions. ⢠Engine-to-engine variability is difficult to estimate when the engine sample size is small (in this study â¤4 engines per model). The value of accurately estimating this parameter warrants the consideration of a longer period of study, allowing more engines of a given model to be studied. 23
JETS-APEX 2 was similar to the Delta Atlanta-Hartsfield study in that it included a second series of experiments focus- ing on emissions sampled during normal airport operations from plumes transported downstream of the active taxiway and runway. The results of these downwind studies continue to be analyzed, and the analysis to date has been summarized in two conference proceedings (Whitefield et al. 2007, Herndon et al. 2007). The major conclusions reported are broadly con- sistent with those from the Delta Atlanta-Hartsfield down- wind studies. Specific conclusions include: ⢠As the plume expands and mixes with the ambient air, a large number of small particles are produced. These nucleation/growth mode particles are not present at the exhaust nozzle. ⢠The production of the small particles increases the number- based emission index by at least an order of magnitude relative to samples acquired at the exhaust nozzle. ⢠The nucleation/growth particles do not significantly con- tribute to the mass-dependent parameter values, and no significant changes in the mass-based emission index are observed. ⢠The -3B series takeoff mass-based emission indices were significantly greater than those for its taxi emissions and for both takeoff and taxi emissions for the -7B series. On average, the mass-based emission index for the -7B series, at both idle and takeoff is less than half that for the older technology -3B series. ⢠In some cases, because of the unique aircraft traffic pat- terns, sampling location, and prevailing wind direction at OAK, takeoff and taxi plumes for different aircraft are found to mix prior to sample extraction, greatly complicating data interpretation. The PM data from these mixed plumes can be de-convolved to yield single aircraft specific infor- mation and such analysis is currently underway. 5.4 APEX3 APEX3 is the most recent field study, and reporting on APEX3 data has not progressed as far as the earlier studies. Although most data are available through the FAA, it has not been interpreted and reported, as is the case for the previous APEX-type studies discussed in this document. Furthermore, the archived PM data have not been corrected for sample line loss, as was the case in the previous studies. Preliminary analyses were presented at the APEX3 conference held in November 2006 (Hagen et al. 2006) and these could be used to draw some qualitative results and intercomparisons. NASA may coordinate a final report, but that report was not available before publication of this ACRP document. For these rea- sons, any reference to APEX3 data in this report exclusively applies to Missouri S&T and ARI data available at press time. 24
