Purple Book (Guide for Controlling ACM in Buildings) Part 20 PDF

Appendix M. Detailed Specifications for Sampling and Analyzing Airborne Asbestos The following specifications are summarized from “Measuring Airborne Asbestos Following an Abatement Action” (USEPA 1985). M.1 Sampling M.1.1 Sampling Equipment Standard sampling equipment consists of a pump (operated a...

Appendix M. Detailed Specifications for Sampling and Analyzing Airborne Asbestos The following specifications are summarized from “Measuring Airborne Asbestos Following an Abatement Action” (USEPA 1985). M.1 Sampling M.1.1 Sampling Equipment Standard sampling equipment consists of a pump (operated at a 2 to 12 liter per minute flow rate), a filter in a cassette and associated tubing and supports. Three types of filters can be used: PCM — cellulose ester with 0.8 to 1.2 µm pore size; TEM — polycarbonate with 0.4 µm pore size (preferred); or cellulose ester with 0.8pm pore size. M.1.2 Number of Samples M.1.2.1 TEM A minimum of five samples inside and five outside the work site is recommended. When a negative air pressure ventilation system has been used during the abatement operation the “outside” samples should be collected outside the work site, but inside the building. This provides a comparison between the work site and the incoming air. If a negative air pressure ventilation system has not been used, the “outside” samples should be collected outdoors. These sample sizes are based on calculations of statistical reliability and on the following characteristics: ● ● ● The coefficient of variation for TEM measurements is between 100% and 150% based on data from EPA research studies, A false positive rate of .10 (i.e., based on the statistical test comparing inside and outside measurements, 10% of the “clean” work sites will fail and have to be recleaned). A false negative rate of at most .10 (i.e., the statistical test comparing inside and outside measurements will identify at least 90% of the sites that must be recleaned). M.1.2.2 PCM A minimum of five samples is recommended. A sample size of five controls the false negative error rate. At least 90% of the sites where the actual fiber concentration exceeds 0.01 f/cc will fail the test. If the actual concentration is 0.02 f/cc the probability of failure is 99%. M-1 M.1.3 Location of Samplers M.1.3.1 Indoors Indoor samplers should be placed so they are not influenced by unusual air circulation patterns. Avoid corners of rooms and obstructions (like furniture). Within the above constraints, samplers should be placed at random around the work site. For example, if the site is a single room of 1000 or more sq. ft., the five samplers should be distributed in an approximately uniform manner. If the site includes more than five rooms, the rooms to be sampled maybe selected randomly. The companion EPA document (USEPA 1985) describes this procedure in more detail. When TEM is used for the air test and a negative air pressure ventilation system has been employed during the abatement operation, the five “outside” samplers should be placed outside the work site but inside the building, and the negative air system left running during sampling. These outside samplers should be located to avoid any air that might escape through the containment barriers. Minimum recommendations are at least 50 ft. from the entry portal to the work site, or 25 ft. from the plastic containment barriers. M.1.3.2 Outdoors If TEM is to be used for the air test and a negative air pressure ventilation system has not been used during abatement, then five samplers should be placed outdoors. These should be placed at ground level (about 2 meters high), if possible, and away from obstructions that may influence wind patterns. If access to electricity and concerns about security dictate a roof-top site, do not place samplers near vents or other structures on the roof. M.1.4 Sampling Volumes M.1.4.1 TEM The required sampling volume for the TEM air testis calculated from the theoretical detection limit of the TEM analysis procedures, and from typical levels of asbestos against which measurements in the work site will be compared: Volume = Where: ● ● ● ● (1 f/10 grid squares) x (0.005 f/cc) (855 mm2 ) 2 (0 . 0 0 5 6 m m ) x (1 liter) = 3054 liters (1000 cc) 1 f/10 grid squares (the maximum recommended filter counting area) is the smallest number of fibers needed to make a non-zero measurement. (This is below the limit of reliable quantification.) 0.005 f/cc is a typical outdoor asbestos level in urban areas, as measured by TEM (Chatfield 1983). 855 mm2 is the collection area of a 37 mm diameter filter. 0.0056 mm2 is the area of each grid square (75 µm per side) in a 200 mesh electron microscope grid. This value will vary from 0.0056 to 0.0081 mm 2 for different grids. Larger grid squares will improve measurement accuracy for the same sampling volume. This equation is appropriate for TEM analysis using a direct sample transfer technique (see Section M.2.1). If an indirect technique is used, the required sampling volume is increased in proportion to the dilution used. For example, if the sample is diluted by a factor of 10, the required volume is 10 times larger. M-2 M.1.4.2 PCM The equivalent PCM limit of reliable quantification for a sampling volume of 3000 liters is: Quantification = Limit (10 f/100 fields) (3000 liters) x (855 mm2) x (0.003 mm 2 ) (1 liter) = 0.01 f/cc (1000 cc) Where: ● 10 f/100 fields is the limit of reliable quantification for the P&CAM 239 method. ● ● 855 mm2 is the collection area of a 37 mm diameter filter. 0.003 mm2 is the size of a typical field of view for a PCM microscope. This value will vary from 0.003 to 0.006 mm2 for different microscopes. Larger fields of view will improve (decrease) the limit of reliable quantification. By increasing the sampling volume, the PCM test criterion can be made proportionally more stringent: Volume Quantification Limit 0.01 f/cc 0.006 0.004 3000 liters 5000 7500 If the sampling scheme associated with the new NIOSH 7400 PCM method is used, the limit of reliable quantification will be lower for the same sampling volume. M.1.5 Aggressive Sampling Procedures for sampling aggressively are: ● Before starting the sampling pumps, direct the exhaust from forced air equipment (such as a 1 horsepower leaf blower) against all walls, ceilings, floors, ledges and other surfaces in the room. This should take at least 5 minutes per 1000 sq. ft. of floor. ● Place a 20-inch fan in the center of the room. (Use one fan per 10,000 cubic feet of room space.) Place the fan on slow speed and point it toward the ceiling. ● Start the sampling pumps and sample for the required time. ● Turn off the pump and then the fan(s) when sampling is complete. M.2 Analysis M.2.1 TEM Use the update to the EPA provisional method (Yamate 1984). The sample should be transferred directly from the polycarbonate filter to the electron microscope grid. If high levels of organic materials are suspected or found, cellulose ester filters and indirect transfer (involving ashing, sonicating, and refiltering the fibers) is recommended. However, levels of airborne organic particles should be low in a cleaned work site. M-3 M.2.2 PCM Use the NIOSH P&CAM 239 method (NIOSH 1979). The newer NIOSH 7400 methods can also be used, although OSHA has yet to replace P&CAM 239 with 7400 for workplace compliance monitoring. NIOSH reports that 7400 is at least as accurate as P&CAM 239 M.3 Interpretation of Results M.3.1 TEM Use student’s “t” test to compare inside and outside levels. ● Compute the natural logarithm of fiber concentration for each sample. ● Compute means of the log transformed data for inside samples and outside samples. ● Form the ratio Where: y1 = average of log concentrations inside the work site y2 = average of log concentrations outside the work site S = [( Σ ( y1j-y 1) 2 + Σ ( y2j-y 2) 2)/(n 1+ n2-2)] ½ n 1 = number of samples collected inside the work site n 2 = number of samples collected outside the work site Then compare T to the 95 percentile point of a “t” distribution with n 1 + n2 –2 degrees of freedom. (When 5 samples are collected inside and outside the 95 percentile point is 1.86.) If T exceeds the 95 percentile point, reclean. Otherwise, release the contractor. M-4 The following two examples illustrate the method: Example 1 Measurements inside the work site (f/cc) Measurements outside the work site (f/cc) 0.002 0.007 0.030 0.028 0.001 0.001 0.010 0.008 0.001 0.025 y2 = –5.39 y1 = –5,03 S = 1.49 T = 0.38 T is less than 1.86. The contractor is released. Example 2 Measurements inside the work site (f/cc) Measurements outside the work site (f/cc) 0.052 0.130 0.005 0.240 0.375 0.001 0.010 0.008 0.001 0.025 y1 = –2.54 y 2 = –5.39 s = 1.59 T = 2.84 T is greater than 1.86. The site must be recleaned. The test is based on the assumption that a homogeneous work site has been selected. If one sample has a much higher concentration than the others it is possible that the site is not homogeneous. Common sense should prevail in this case. Irrespective of the result of the “t” test, the high value should be investigated. The sample should be reanalyzed, additional samples collected, or the site recleaned and tested before the contractor is released. M.3.2 PCM The measured level of each sample is compared with the PCM limit of reliable quantification for the volume of air sampled (approximately 0.01 f/cc for 3000 liters). If any of the samples exceeds 0.01 f/cc, the work site must be re-cleaned. M-5