DIESEL EXHAUST AEROSOL LEVELS IN UNDERGROUND COAL MINES

ABSTRACT

The University of Minnesota and the U.S. Bureau of Mines collaborated to develop and Geld test a personal diesel exhaust aerosol sampler (PDEAS). The PDEAS was field tested in five underground coal mines that use continuous miners and a variety of diesel vehicles, in­cluding diesel haulage and utility vehicles. One mine was surveyed a second time, with the haulage vehicles fitted with a low-temperature disposable diesel exhaust filter (DDEF). Aerosol samples were collected with a variety of instruments, including the PDEAS and the microorifice, uniform-deposit impactor (MOUDI). This paper presents the diesel exhaust aerosol (DEA) concentration data col­lected in these mines and assesses the impact of diesel face-haulage equipment, with and without exhaust filters, on underground mine air quality.

The average DEA concentration at the haulageway location for five mines, determined by the PDEAS, was 0.89 mg/m3, with a standard deviation of 0.44 mg/m3. DEA contributed 52 pct, of the respirable aerosol at the haulageway location. Use of the DDEF at one mine re­duced DEA by 95 pct, with a standard deviation of 6 pct, and filter life averaged 10 h. DEA contributed a large proportion of the respirable aerosol concentrations in mines with diesel equipment, and a substantial reduction was achieved with use of a DDEF.

SAMPLING AND ANALYSIS METHODS

The PDEAS, described elsewhere (8) and shown sche­ matically in figure 1, has three stages and employs inertial impaction for separating and collecting the diesel and mineral dust fractions of the sampled respirable aerosol. The first stage is an inertial preclassifier, a 10-mm Dorr- Oliver6 cyclone that separates and collects the larger, nonrespirable aerosol. The second stage is a four-nozzle impactor with a sharp 50 pct cut point of 0.8 pm aero­ dynamic diameter. Most aerosol particles larger than 0.8 pm are deposited on an impaction substrate in this stage. The third stage is a filter that collects the remain­ ing aerosol of less than 0.8 pm aerodynamic diameter. The sampler operates at a flow rate of 2 L/min, which is compatible with both the personal sampler pump and the 10-mm cyclone.
It was shown in the laboratory (72) and in underground mines (4-5, 7) that inertial impaction, followed by gravi­metric analysis, can be used to separate and sample diesel exhaust and mineral dust aerosol fractions and provide estimates of DEA concentrations. These preliminary evaluations of the sampling technique indicate that these are accurate to within 25 pct, 95 pct of the time, for concentration levels above the estimated limit of detection of 0.3 mg/m3. Below this level, indications are that the 95-pct confidence interval can exceed 60 pct because of interferences caused by submicrometer mineral dust and background atmospheric aerosol. During PDEAS field tests, aerosol samples were also collected using MOUDI samplers (13). Analyses of MOUDI-derived size distri­butions provided accurate concentrations of DEA and res­pirable coal mine dust aerosol which were used to meas­ure the performance of the PDEAS.

In each mine, 12 PDEAS’s were deployed at different locations in a continuous miner section utilizing diesel-powered shuttle cars. These locations included the intake entry, haulageway, return entry, shuttle car(s), and, in a few instances, on individuals. Figure 2 is a schematic diagram showing a typical room-and-pillar section with the location of the stationary sampling points used in the sampler field tests. Each of the mine sections surveyed used two to four diesel-powered shuttle cars to haul coal cut by an electric continuous miner. Additional samples were collected using a size selective sampler designed by MSHA. These data were reported elsewhere (74). Com­plete details of the mine layout, mining method, pro­duction tonnage, and ventilation for each mine are avail­ able from MSHA (75).

Mine Research Establishment (MRE) equivalent res­pirable coal mine dust concentrations are reported for PDEAS data. These data were calculated as (1.38 x (mass deposited on the impactor plate of the PDEAS) + mass collected on the after-filter) 4- the volume of sampled air (76). DEA concentrations were determined by dividing the mass collected on the filter by the volume of air sampled. No correction was made for mineral dust aero­ sol deposited on the submicrometer stages of the im­pactors or background aerosol entering the section through the intake airway. The 1.38 correction factor is intended to adjust collected sample mass for the difference in pene­tration efficiency between the 10-mm cyclone preclassifier and the Cassella elutriator preclassifier on which the MRE definition of respirable dust is based (76). Since the dif­ference in penetration efficiency affects only aerosol great­er than 1 nm in size, the correction factor is applied only to the coarse (greater than 0.8 /im) part of the PDEAS sample.

EVALUATION OF A DISPOSABLE DIESEL EXHAUST FILTER

The DDEF system was evaluated in a continuous miner section in mine M. MSHA safety standards require that vehicles used at the face be equipped with water-bath exhaust conditioners to control exhaust temperatures and to arrest flames and sparks emitted from diesel engines (77). Figure 3 illustrates the mounting of the DDEF sys­tem and provides a cross-sectional view of the DDEF. Ex­haust passes through the water scrubber before passing through the DDEF. The DDEF system was developed to take advantage of the low exhaust temperatures exiting the water scrubber and is described in detail elsewhere in this Information Circular. Filter life averaged about 10 h at this high-altitude mine.

The week-long field study to evaluate the performance of the DDEF’s was conducted at mine M. All the vehicle in the section, which included three diesel-powered shuttle cars and one scoop, were equipped with the DDEF system. During the first 4 days, the vehicles were operated with the DDEF installed, and on the last day the DDEF was removed. The sampling protocol and locations were similar to those previously described. The concentration of DEA in the mine environment was measured with the PDEAS and other aerosol instruments. Samples were collected during normal production shifts in the ventilation in the intake, in the haulageway, on the shuttle cars, and in the return. Coal production, measured by tonnage, varied from 270 to 475 kg/shift and ventilation varied from 1,600 to 2,300 m3/min during the test.

RESULTS

Table 2 summarizes the results obtained with the reported in table 3. Table 4 summarizes the results from PDEAS’s for each mine. Samples collected at mine M arc MOUDI samples collected at each mine.

Sampling times for the PDEAS and MOUDI varied with aerosol concentration and mining activity to avoid overloading the impaction substrates and to minimize particle bounce. The shortest sampling times were in the return airways, where aerosol concentrations were the highest. MOUDI and PDEAS data collected at the mine portal and section intake were almost always collected concurrently over a full, or nearly full, shift because aero­ sol concentrations were usually low. MOUDI sample col­lection times at the haulage and return sites averaged 96 min and ranged from 33 to 214 min. 

Two or three sets of MOUDI samples were collected per shift because long­ er sampling times result in an overload of the MOUDI substrates. Fewer MOUDI samples were collected in the return airway because of MSHA permissibility require­ments. These requirements prohibit the sampling pump from being in the return airway and prohibit the use of the MOUDI motor, forcing manual operation. Usually MOUDI return samples were collected only on the last day of sampling. PDEAS mean sampling times ranged be­ tween 193 and 309 min at the haulage and return loca­tions. All sampling was concurrent at the haulage location; therefore, PDEAS samples covered the same time periods as the multiple MOUDI samples, but the PDEAS substrates were not changed. PDEAS samples collected on shuttle cars, or on research personnel, are near full-shift data.

It is clear from MOUDI samples collected at the mine portals (table 4) that little respirable aerosol enters the mine environment from outside sources. Each of the mines is located in a sparsely populated region where air pollution is not a problem. Sources of respirable aerosol outside the mines are natural windblown dust, pollen, fugitive dust generated by vehicular traffic, storage piles, and belt conveyors.

Concentrations of respirable aerosol and DEA at the section intake, shown in tables 2 and 4, were very much dependent upon vehicular traffic in that area. Mines J, L, and O had the least traffic in the section intake. Mine L had only three diesel shuttle cars, which seldom passed the intake sampling points while the samplers were operating. Mines K and N used diesel equipment almost exclusively to move personnel, material, and coal; thus, intake DEA concentrations were higher.

The other sampling locations—haulageway, shuttle car, and scientist—were located near mining personnel and diesel activity. The return sampling site was selected to obtain a well-mixed aerosol sample, sufficiently far from the continuous miner to allow settling of large particles. The return sampling location was the area least likely to  have mine personnel on a full-shift basis. It was sampled only for the limited time that the continuous miner was operating.

The respirable and DEA concentrations are higher in the haulage and return entries of mine N and in the re­ turn entry of mine O. For mine N, this was due to low- ventilation airflow during the survey. Mine O was unique in that it used a combination of beet pulp and rock dust as fire retardants on mine ribs in the return airway and at the coal face. This caused unusually high dust concentrations during sampling, particularly in the returns, because it flaked and fragmented more than other mine materials and became entrained in mine air.

Results, summarized in table 3, show that in relative terms the DDEF reduced average DEA concentration by 95 pct, with a standard deviation of 6 pct. In absolute terms, DEA concentrations with the DDEF installed were less than 0.2 mg/m3 at all locations. The concentrations and standard deviations shown in table 3 are average values, uncorrected for intake air concentration or pro­duction and ventilation changes. However, the DEA re­ductions (A) shown in the table were calculated by includ­ing correction factors for these parameters. 

Detailed size distribution and indium fuel tracer data further confirm that submicrometer mine aerosol is comprised mainly of DEA, which was effectively removed when the DDEF was installed (11). Figure 4 is a lognormal probability plot that shows the cumulative frequency distribution of respirable aerosol (A) and DEA (B) concentrations obtained from the PDEAS samples collected in mines J, K, L, N, and O at the haul­ age, shuttle car, and return locations. The figure plots the cumulative percentage (y-axis) of samples with concen­trations less than the concentration shown on the x-axis. Table 5 provides an overall summary of the PDEAS and MOUDI data, by location, for mines J, K, L, N, and O and provides the summary statistics for PDEAS data plotted in figure 4.

Generally, haulage and shuttle car locations have similar distributions for DEA (table 5), with median concentrations of 0.75 and 0.65 mg/m3, respectively. The median MOUDI concentration obtained at the haulage site where sampling was concurrent was 0.78 mg/m3 (ta­ble 5). These data suggest that the concentration of die­sel exhaust is relatively uniform throughout the section. The median PDEAS concentration in the return location is higher at 1.06 mg/m3 (MOUDI median 1.16 mg/m3), which is expected. Respirable aerosol concentrations are also similar at the haulage and shuttle car locations, and again the return airway respirable aerosol concentrations are higher than concentrations in the haulageway and on the shuttle cars. DEA accounts for a large fraction of the respirable aerosol at every location with significant diesel activity.

