Airborne Crystalline Silica Concentrations at Coal-Fired Power Plants Associated with Coal Fly Ash

ABSTRACT

This study presents measurements of airborne concentrations of respirable crystalline silica in the breathing zone of workers who were anticipated to encounter coal fly ash. Six plants were studied; two were fired with lignite coal, and the remaining four plants used bituminous and subbituminous coals. A total of 108 personal breathing zone respirable dust air samples were collected. Bulk samples were also collected from each plant site and subjected to crystalline silica analysis. Airborne dust particle size analysis was measured where fly ash was routinely encountered. The results from bituminous and subbituminous fired plants revealed that the highest airborne fly ash concentrations are encountered during maintenance activities: 0.008 mg/m3 to 96 mg/m3 (mean of 1.8 mg/m3). This group exceeded the threshold limit values (TLV®) in 60% of the air samples. During normal production activities, airborne concentrations of crystalline silica ranged from nondetectable to 0.18 mg/m3 (mean value of 0.048 mg/m3). Air samples collected during these activities exceeded the current and proposed TLVs in approximately 54% and 65% of samples, respectively. Limited amounts of crystalline silica were detected in samples collected from lignite-fired plants, and approximately 20% of these air samples exceeded the current TLV. Particle size analysis in areas where breathing zone air samples were collected revealed mass median diameters typically between 3 μ m and 8 μ m. Bulk and air samples were analyzed for all of the common crystalline silica polymorphs, and only alpha quartz was detected. As compared with air samples, bulk samples from the same work areas consistently yielded lower relative amounts of quartz. Controls to limit coal fly ash exposures are indicated during some normal plant operations and during episodes of short term, but high concentrations of dust that may be encountered during maintenance activities, especially in areas where ash accumulations are present.

MATERIALS AND METHODS

Information was gathered about the common characteristics and typical coal types used at large coal-fired power plants in the United States. From this information, candidate power plants were identified, and six sites were selected for air sampling studies. Job descriptions and work activities were examined to identify likely tasks and workers who would encounter airborne fly ash. Air sampling was conducted during two visits to each power plant. Sampling was conducted during normal operations (normal production) and during scheduled and unscheduled maintenance outages.

Site Selection

The types, sizes, and fuel sources of coal-fired power plants are diverse. Many of these plants have common characteristics, some of which could affect the airborne concentrations of respirable crystalline silica. The primary criteria used in selecting power plants for this study included:

Type of Fly Ash Collection System

Approximately 90% of power plants use electrostatic precipitators for fly ash collection. Recent trends indicate increased use of fabric filters (i.e., baghouse).

Type of Fly Ash Handling and Disposal System

Approximately 70% of fly ash produced at electric generating plants is handled and disposed of dry, with the remainder handled and disposed of wet (i.e., as a slurry). Industry trends are moving toward dry handling due to water pollution and economic concerns. Airborne concentrations associated with wet handling jobs would likely be low or non-existent. All of the power plants selected for this study process and handle fly ash disposal in a dry state.

Type of Fuel Burned

In order of decreasing use, commonly used fuels are: Eastern bituminous coal, western subbituminous coal, interior bituminous coal, northern central lignite, and Gulf Coast lignite. Plants using each of these coal types were included in this study.

Other considerations included the size of the plant, the desire to study facilities with multiple boilers to enhance the opportunities to study both maintenance and normal production related activities, and the willingness of the plant to participate in this study.

Six power plants were selected for the field aspects of this study. A summary of relevant characteristics is presented in Table I.


Study Strategy

Interviews with plant personnel and preliminary site visits indicated that fly ash exposures occur at many locations and during various tasks routinely performed, which can be easily divided into maintenance and normal production activities. Normal production (or normal operations) activities are primarily associated with normal day-to-day operations, including collection, handling, and disposal of fly ash residue. During these normal production activities, the number of personnel working in areas where dry fly ash might be encountered is a small fraction of the work force engaged in the operation of the generating unit. There were approximately 5 to 10 workers per shift (less than 10% of the operations work force) for most of the plants studied.

Maintenance activities that result in exposures to airborne fly ash occur when the generating unit is operating, as well as when the unit is shut down and workers enter interior chambers of the boiler and related equipment. Major maintenance outages often engage hundreds of workers who clean interior surfaces during the first or second week of the outage. These cleaning and initial maintenance activities, which might include erection of scaffolding inside various compartments, often are associated with work around considerable quantities of accumulated fly ash in areas with limited ventilation. Once the initial cleaning of the accumulated ash is performed, airborne dust concentrations are considerably lower than the precleaning activities.

Air samples were collected from personnel engaged in various stages of maintenance work, including during normal minor maintenance around fly ash areas (e.g., ESP/baghouse hopper areas, ash silos, etc.), as well as during major scheduled maintenance outages when the operating unit is shut down and workers enter interior chambers. At two plants, site visits were conducted during unscheduled outages (termed emergency outages) that involved shutting down the boilers and entry into interior chambers due to equipment failure that required immediate repairs. Extensive cleaning activities are usually not performed during these short-term repairs. One plant site was also visited in the later stages of a major scheduled outage because it was anticipated that airborne fly ash concentrations would be lower because the interior chambers are free of accumulated ash at this stage.

Air Sampling and Analytical Methods

Crystalline silica air samples were collected and analyzed from workers' breathing zones during normal full-shifts according to National Institute for Safety and Health (NIOSH) method 7500. Battery-powered personal air sampling pumps (model 224; SKC, Eighty Four, Pa.) were attached to workers' belts. Air was passed through a preweighed 37 mm diameter 5.0 μ m pore size polyvinyl chloride membrane filter housed in a polystyrene cassette. The filter cassette was used with a preselection device (10-mm cyclone; Dorr-Oliver) that had size-selecting characteristics consistent with the American Conference of Governmental Industrial Hygienists (ACGIH®) guidelines for respirable dust fraction sampling. The sampling trains were calibrated at a flow rate of 1.7 L/min using a precision rotameter and a vacuum chamber. Each rotameter was calibrated against a primary standard calibration device (frictionless piston) prior to each day's use. Nylon cyclones were handwashed with a nonresidue soapy water solution after each use, rinsed in distilled water, and allowed to air dry in a clean environment.

The majority of the work shifts sampled were 8 hours, with some exceptions. In approximately 8% of all samples, there were extended work shifts (e.g., 10 to 12 hours). For all samples, the 8-hour time-weighted average was computed using the average airborne dust concentration and duration of the work shift in the numerator and 8 hours in the denominator, to compute the weighted average for comparison with the exposure limit. All time-weighted averages were compared with the current American Conference of Governmental Industrial Hygienists threshold limit value (TLV®) of 0.05 mg/m3, and the proposed TLV of 0.025 mg/m3.

Analysis of air and bulk samples was performed according to NIOSH method 7500 by an AIHA-accredited laboratory. Field blank samples were collected and analyzed following standard methods. All samples were analyzed for quartz, cristobalite, and tridymite. No crystalline silica was detected in any of the field blank samples. Bulk samples were collected in work areas where fly ash exposures were observed. In each of these areas, four samples of surface dust were collected and commingled.

For a limited number of heavily loaded samples (11 samples), the interior surfaces of the sampling cassette was rinsed with double distilled/deionized water to ensure that all of the dust was removed and placed in a metallic weighting container. The sample was then thoroughly desiccated and heated to remove any water residue prior to gravimetry.

Particle size analysis was conducted in the primary work locations, typically at stationary locations within 10 to 20 feet of workers, using a stationary cascade impactor positioned on a tripod at approximately 4 to 5 feet above the floor. The sampling device was an Andersen model SA235 sampler (Thermo Andersen Inc., Smyrna, Ga.), a 5-stage cascade impactor using slotted preweighted fiberglass filters and a high-volume vacuum pump. The flow rate through the impactor was calibrated at 20.0 standard ft3/min using a u-tube manometer according to the manufacturer's instructions. Airflows were calibrated at the beginning and end of each sampling period. Field blank filter samples were collected at a rate of approximately 10% of the total number of air samples. The cut points for this impactor at the established flow rate were 0.73 μm, 1.4 μm, 2.1 μm, 4.2 μm, and 10.2 μm. Analysis was performed by gravimetric methods at the laboratory.
RESULTS

A general description of the jobs, activities, and plant locations are presented in Table II.  For the majority of individuals sampled, fly ash appeared to be the only source of crystalline silica present; there was no coal dust or sandblasting grit present.

Results of air sampling are presented separately for bituminous/subbituminous-fired plants (Table III) and lignite-fired plants (Table IV) because the crystalline silica content is significantly different with these coal types. The data are grouped by general work classification (normal production compared with maintenance activities), job locations, and job description. Summary statistics are presented, when possible, for each job description, such as range, mean, standard deviation, geometric mean, and geometric standard deviation. Nondetects were given a value of one-half the detection limit when computing means and standard deviation values. The results are presented in terms of the computed 8-hour time-weighted average concentrations. All sampling times ranged from 304 min to 672 min (mean of 488 minutes).

During normal production operations, exposures were generally lower than during maintenance activities. For normal operations, mean exposure was 0.048 mg/m3; for maintenance operations mean exposure was 0.23 mg/m3. Consistently elevated airborne concentrations were encountered by workers engaged in dry vacuuming and scaffold setup and by mechanics/electricians at various plant locations, especially those within the boiler or other conveyance/ash collection equipment. One high value of 96 mg/m3 was considered suspect and was investigated further. The laboratory verified the analytical results and documented the sample handling practices to ensure all of the collected dust was quantitatively analyzed. Higher crystalline silica quantities were detected in air samples between the higher ranked coal-fired plants (bituminous/subbituminous) and compared with the lignite-fired sites. The mean crystalline silica content in the air samples from the bituminous/subbituminous sites was 7.5%, and the mean value from the lignite site was 1.7%.

Table IV presents the results of air sampling at the lignite-fired plants (Plants 5 and 6). At these plants, considerably lower airborne concentrations of respirable crystalline silica were encountered. At Plant 6, no crystalline silica was detected in the bulk or air samples. At Plant 5, airborne concentrations were above the TLV in approximately 20% of the samples (2 out of 11).

Table V presents the results of crystalline silica analysis of bulk dust and respirable air sample dust for crystalline silica content (the mean of all air samples in that location). On some occasions, industrial hygiene investigators assume the content of crystalline silica in bulk dust is similar to that found in air samples and may, therefore, limit crystalline silica analysis to bulk samples in a desire to limit the likelihood of nondetects from limited amounts of dust captured in air samples. Quartz was the only crystalline silica polymorph detected in any of the fly ash samples. The comparison indicates that in every circumstance, the content of respirable quartz in air samples was greater than the amount found in bulk samples collected in the same work area. In a relative sense, the quartz in bulk samples was approximately 35% to 66% of the quartz detected in air samples collected in the same location. This could be due to a concentration of quartz in the smaller, respirable-size particles.

A summary of particle size data from work areas where airborne fly ash was the predominate particulate present are presented in Table VI. These results were similar at all sites, and indicate that a considerable fraction of the airborne dust is in the respirable size range. 
