Emission measurements were obtained for a variety of military vehicles at Hill Air Force Base (Ogden, UT) in November 2000 as part of a Strategic Environmental Research and Development Program. Aircraft ground support equipment vehicles using gasoline, diesel, and JP8 fuels were tested using chassis dynamometers under predetermined load. The exhaust from the tested vehicle was passed to a dilution tunnel where it was diluted 30-40 times and collected using Micro-Orifice Uniform Deposit Impactor (MOUDI) fitted with aluminum substrates, an XAD-coated annular denuder, and a filter followed by a solid adsorbent. All MOUDI substrates were analyzed for mass and for organic and elemental (EC) carbon by the thermal/optical reflectance method and for polycyclic aromatic hydrocarbons (PAHs) by GC/MS. Black carbon was measured with a photoacoustic instrument. The denuder and filter/solid adsorbent samples were analyzed for semivolatile PAH. Overall, there is more mass and higher EC contribution when the vehicle is run under higher load in comparison with the low load. However, older vehicles generally show more mass and EC emissions than newer vehicles, and there is a shift toward smaller particle sizes for the low load, which is most pronounced for newer vehicles. The particle-associated semivolatile PAHs and nonvolatile four-through six-ring PAHs are present predominantly on the submicron particles collected on MOUDI stages 0.1-0.18, 0.18-0.32, and 0.32-0.56 microm. For the low-load runs, the distribution of PAHs seems to be shifted toward smaller size particles. The gas-particle phase distribution of semivolatile PAHs depends also on the engine loading. For idle, not only are the more volatile two- and three-ring PAHs, from naphthalene to dimethylphenanthrenes, retained on the denuder portion, but also less volatile four-ring PAHs, such as fluoranthene and pyrene, are retained by the denuder at the 80-90% range, which implies that they are present predominantly in the gas phase. In contrast, for engines under high loads, a much larger portion of three- and four-ring PAHs are partitioned to the particle phase.

A series of measurements have been performed at Hill Air Force Base to evaluate real-time instruments for measurements of black carbon aerosol and particle-bound PAHs emitted from spark and ignition compression vehicles. Vehicles were operated at idle or fast idle in one set of measurements and were placed under load on a dynamometer during the second series. Photoacoustic instruments were developed that operated at a wavelength of 1047 nm where gaseous interference is negligible, although sensitivity to black carbon is good. Compact, efficient, solid-state lasers with direct electronic modulation capabilities are used in these instruments. Black carbon measurements are compared with samples collected on quartz fiber filters that were evaluated using the thermal optical reflectance method. A measure of total particle-bound PAH was provided by photoelectric aerosol sensors (PAS) and is evaluated against a sum of PAH mass concentrations obtained with a filter-denuder combination. The PAS had to be operated with a dilution system held at approximately 150 degrees C for most of the source sampling to prevent spurious behavior, thus perhaps compromising detection of lighter PAHs. PA and PAS measurements were found to have a high degree of correlation, perhaps suggesting that the PAS can respond to the polycyclic nature of the black carbon aerosol. The PAS to PA ratio for ambient air in Fresno, CA is 3.7 times as large in winter than in summer months, suggesting that the PAS clearly does respond to compounds other than BC when the instrument is used without the heated inlet.

Diluted exhaust from a selection of Air Force ground support vehicles was subjected to gravimetric, carbon, and size distribution analyses in September 1999. The vehicles operated on diesel and JP-8 fuels. In most cases, the engines involved were similar to civilian counterparts. The tests involved "low" and "high" idle settings but no external loads were imposed. Particle size distribution data, obtained over the 10 to 352 nanometer diameter range using an SMPS instrument, showed that the relative number count of accumulation mode particles increased with respect to nucleation mode particles as the engine rpm increased. The SMPS distributions often explained the main variations in the integrated PM 2.5 gravimetric mass data. Particulate mass derived from the SMPS data and from cascade impactor measurements were well correlated (regression slope 1.02). Empirically determined "elemental" carbon (EC) and "organic" carbon (OC) were the main constituents of the PM 2.5 gravimetric mass (regression slope 0.89). EC contributed less, and OC contributed more to the PM 2.5 mass than was found in some recent studies of exhaust from vehicles operated under external loads. The observed particle nucleation modes were attenuated by coagulation with accumulation mode particles, but it does