Control of Silica Exposure from Hand Tools in Construction: Grinding Concrete

ABSTRACT

When construction workers use handheld grinders to smooth poured concrete surfaces after forms are stripped, they risk overexposure to respirable dust and crystalline silica. This article examines the performance of a local exhaust ventilation system for handheld grinders. The system consisted of the grinder (equipped with a ventilated shroud), a length of flexible corrugated hose, and a portable electric vacuum cleaner that acted as the fan and dust collector for the ventilation system (see Figures 1 and 2). Personal breathing zone air samples for respirable dust and crystalline silica were collected during five days of grinding.


METHODS

Personal breathing zone samples for respirable particulates and crystalline silica were collected at a flow rate of 1.7 liters/minute using a battery-operated sampling pump connected via flexible tubing to a 10-millimeter (mm) nylon cyclone (Mine Safety Appliances Co., Pittsburgh, PA), and a preweighed, 37-mm-diameter, 5-micron (μm) pore size polyvinyl chloride filter supported by a backup pad in a two-piece filter cassette sealed with a cellulose shrink band, in accordance with NIOSH Methods 0600 and 7500. In addition to the personal samples, bulk samples of settled dust were collected in accordance with NIOSH Method 7500. Two personal breathing zone samples were collected each day.The pump was turned off during breaks, at lunch time, and when the worker left the area or performed a task other than grinding.

Gravimetric analysis for respirable particulate was carried out with the following modifications to NIOSH Method 0600: 1) The filters and backup pads were stored in an environmentally controlled room (21C ± 3C and 50% ± 5% relative humidity), and were subjected to the room conditions for at least two hours for stabilization prior to tare and gross weighing; and 2) two weighings of both the tare weight and gross weight were performed. The difference between the average gross weight and the average tare weight was the result of the analysis. The limit of detection for this method was 0.02 milligrams (mg).

Crystalline silica analysis was done using X-ray diffraction. NIOSH Method 7500 was used with the following modifications: 1) Filters were dissolved in tetrahydrofuran rather than being ashed in a furnace; and 2) standards and samples were run concurrently, and an external calibration curve was prepared from the integrated intensities rather than using the suggested normalization procedure. These samples were analyzed for two forms of crystalline silica: quartz and cristobalite. The limits of detection for quartz and cristobalite on filters were 0.01 and 0.02 mg, respectively. The limit of quantitation is 0.03 mg for both quartz and cristobalite on filters. Bulk samples were collected and analyzed qualitatively for quartz and cristobalite by X-ray diffraction to determine if any interference was present in the material.

The limits of detection in bulk samples were 0.8 percent for quartz and 1 percent for cristobalite. The limit of quantitation was 2 percent for both forms of crystalline silica in bulk samples.

Productivity was quantified by using a tape measure to mark off the area of concrete surface finished during each day of sampling. Vacuum cleaner performance was measured in terms of the capacity of the vacuum cleaner bag and the static pressure and air flow through the shroud-hose-vacuum cleaner system. Bag capacity was measured using a portable electronic scale to weigh the empty bag and to weigh the bag when the concrete finisher judged it to be full. The time was also noted each time the vacuum cleaner bag was changed in order to determine how long it took the vacuum cleaner to reach its capacity. Static pressure was measured by inserting a smooth-walled metal tube (of the same diameter as the vacuum cleaner hose) in line with the hose between the shroud and the vacuum cleaner, and by measuring the suction using a U-tube manometer placed on a hole drilled in the side of the metal tube while the shroud was held against a flat surface. Air flow was calculated by measuring the air velocity in the vacuum cleaner hose and multiplying this by the cross-sectional area of the hose. Pressure and air velocity measurements were made at a sufficient number of hose diameters from the tool and the vacuum cleaner to mitigate the effects of turbulence on the measurements. Centerline velocity was measured using a TSI Velocicalc Plus model 8386 multiparameter ventilation meter.

Criteria

Silicosis is an occupational respiratory disease caused by inhaling respirable crystalline silica dust. Silicosis is irreversible, often progressive (even after exposure has ceased), and potentially fatal. Exposure to silica dust occurs in many occupations, including construction. Because no effective treatment exists for silicosis, prevention through exposure control is essential. When proper practices are not followed, or controls are not maintained, silica exposures can exceed the NIOSH Recommended Exposure Limit (REL) or the Occupational Safety and Health Administration (OSHA) Permissible Exposure Limit (PEL).

The NIOSH REL for respirable crystalline silica is a 10-hour time-weighted average (TWA) level of 0.05 milligrams per cubic meter (mg/m3). NIOSH has classified crystalline silica as a potential occupational carcinogen. Therefore, NIOSH recommends that employers make efforts to reduce silica exposures to levels below the REL.

The OSHA PEL for respirable dust containing 1 percent or more quartz is expressed as the following equation:

Respirable PEL =  10/(% Silica) + 2

Thus, if the dust contains no quartz, the PEL is 5 mg/m3, and if the dust is 100 percent quartz, the PEL is 0.1 mg/ m3. For tridymite and cristobalite (other forms of crystalline silica), OSHA uses half the value calculated when using the formula for quartz.

RESULTS 

Industrial Hygiene Sampling

The results of personal breathing zone air sampling and analyses for respirable dust and quartz are presented in Table I. No cristobalite was detected in any of the air samples. Bulk sample results ranged from 7.4 to 15 percent quartz in samples of concrete dust. No cristobalite was detected in any of the bulk samples.

PELs for respirable dust containing >= 1 percent quartz for each day were calculated based on percentages of quartz in samples. PELs were tested and found to follow a lognormal distribution (in order to perform statistical tests based on the shape of the distribution) with mean PEL over the five days of 0.83 mg/m3. Table I shows that 8-hour TWA respirable dust results ranged from 0.55 mg/m3 to 1.2 mg/m3 , or from 0.45 to 1.5 times the PEL. These results exceeded the OSHA PEL for respirable dust containing >= 1 percent quartz on four of the five sampling days. Respirable dust data were tested and found to follow a lognormal distribution. Logarithms of the measured values were therefore used for further analyses. The arithmetic mean concentration was estimated to be 1.14 mg/m3, with geometric mean of 1.35 mg/m3 and geometric standard deviation of 1.43. An upper exact 95 percent confidence limit for exposure levels was found to be 1.83 mg/m3, indicating that we can be 95 percent confident that the true arithmetic mean exposure for respirable dust in this case is less than 1.83 mg/m3. We can be 95 percent confident that the true mean exposure is less than five times the (average) PEL of 0.83 mg/m3. Five times the PEL was chosen as an arbitrary limit because it is one-half the assigned protection factor of a half-facepiece air purifying respirator, and thus offers a large margin of safety for the use of that type of respirator.

Review of Table I also indicates that 8-hour TWA quartz exposures ranged from 0.036 mg/m3 to 0.13 mg/m3, or from 0.72 to 2.6 times the REL of 0.05 mg/m 3. These results exceeded the NIOSH REL on four of the five sampling days. The quartz data were also tested and found to follow a lognormal distribution. Logarithms of the measured values were therefore used for further analyses. The arithmetic mean concentration was estimated to be 0.16 mg/m 3 , with geometric mean of 0.14 mg/m 3 and geometric standard deviation of 1.7. An exact 95 percent confidence limit for exposure levels was found to be 0.23 mg/m3, indicating that we can be 95 percent confident that the true arithmetic mean exposure for quartz in this case is less than 0.23 mg/m3. We can be 95 percent confident that the true mean exposure is less than five times the REL of 0.05 mg/m3.

Productivity and Performance

Table II lists the square feet of concrete surface finished each day. These typically included a mixture of walls and columns. The vacuum cleaner bag was emptied twice on Day 1 and Day 2, once on Day 3 (a day that was devoted primarily to finishing six sets of paired columns), twice on Day 4, and once on Day 5 (a day that also involved grinding more columns than walls). The scale was used to weigh bags on Day 2. The two full bags weighed approximately 18 pounds each. It was not clear whether the decision to empty the bag was determined by the physical capacity of the bag, or by the concrete finisher’s need to change the bag at a manageable size and weight.

Air velocity measurements were made once each day on Days 1, 2, and 4. The average air flow calculated from these measurements was 96 cubic feet per minute (range: 86 to 106 cfm). Static pressure was measured on Day 4, and was found to be 21 inches of water (in wg). A graph in the manufacturer’s catalog rates the flow at approximately 100 cfm at a static pressure of approximately 21 in wg, indicating that performance in the field is in line with the manufacturer’s specifications. The manufacturer rates the maximum flow of this vacuum cleaner as 112 cfm, with a maximum negative pressure of 84 in wg.
