Document rBmy7qZm0bGMqXLKqO8xpkevE

t * PRO-TEKTM COLORIMETRIC AIR MONITORING BADGE SYSTEMS LABORATORY VALIDATION REPORT SULFUR DIOXIDE BADGE, TYPE C-20 Copyright 1981 by E. I. du Pont de Nemours & Company (Inc,) Wilmington, Delaware 19898 NOTICE The recommended procedures and product performance described herein have been developed by Du Pont. The systems are designed to be used under the supervision of knowledgeable technicians and the Du Pont Company does not offer any warranty expressed or implied for the products or their performances. Further, Du Pont does not assume liability and will not be responsible for any injuries or damage to persons or property which result from the improper use of this system. E-38245 2/12/81 R&S166328 ABSTRACT This laboratory validation report for the Du Pont PRO-TEKTM colorimetric air monitoring badge for sulfur dioxide (SC^) demonstrates that this system, when tested by rigorous and accepted standard laboratory practices, meets the NIOSH accuracy requirements for an analytical sampling method. Adequate storage stability of exposed badges has been shown. R&S166329 * PRO-TEKTM COLORIMETRIC AIR MONITORING BADGE SYSTEMS LABORATORY VALIDATION REPORT SULFUR DIOXIDE BADGE, TYPE C-20 TABLE OF CONTENTS Principle of the Method Test Protocol Analytical Method Sampling and Analysis Precision and Accuracy Environmental Variables Interferences Storage Stability Collection Efficiency Independent Method Generation of Test Atmospheres Summary of Statistical Terms and Formulae Page 1 1 2 2 2 3 3 3 4 4 A-l B-l 3eJ <oo2> CO CO o 9 PRO-TEKTM Colorimetric Air Monitoring Badge Systems LABORATORY VALIDATION REPORT Sulfur Dioxide Badge, Type C-20 SUBSTANCE: METHOD: PROCEDURE: DATE ISSUED: DATE REVISED: SAMPLING RANGE: PRECISION: OVERALL SYSTEM ACCURACY: Sulfur Dioxide, SO2 Du Pont Type C-20 Badge, PT-3 Colorimeter Colorimetric February 16, 1981 10-100 ppm-hours MCV = 0.017 + 12.6 percent PRINCIPLE OF THE METHOD The procedure for collection and analysis of air samples of sulfur dioxide using the Du Pont PRO-TEK^ Colorimetric Air Monitoring Badge System (C-20) is described in Attachment A. This method consists of collection of the sample in a badge that contains absorbing solution and color reagents similar to those described in NIOSH P&CAM, Method No. 160. This method has been tested for validity using similar criteria for validation outlined in Reference 1. Based on these criteria, the absolute total error (sampling and analysis) should be less than 25 percent at the OSHA standard level (equivalent to an eight-hour TWA exposure dose of 40 ppm-hours) 95 percent of the time. R&SI 66331 TEST PROTOCOL The protocol used for testing this method was to: (A) Analyze 22 samples (two each between 0.325X and 2.5X the OSHA standard) spiked with the appropriate amounts of sodium sulfite to represent exposure dose levels between 0 and 100 ppm-hours of sulfur dioxide. (B) Analyze 40 samples collected from dynamically generated test atmospheres (10 samples collected at each of 0.825X, 1.1X, 1.675X, and 2.15X the OSHA standard). 4 2- - (C) Test the storage stability (at 40F and 77F) of 11 collected samples. (D) Assess the overall system accuracy, bias, and precision of the method. (E) Environmental variables were considered. (F) Chemicals used for validation were: Sulfur dioxide in air mixtures from Scott Environmental Technology, Inc., Plumsteadville, Pennsylvania 19849. Accuracy + 2 percent (traceable to National Bureau of Standards). ANALYTICAL METHOD A description of the method of analysis is given in Reference 2. The results of the analytical method linearity tests are in Table 1. The precision of the analytical linearity method shown in Table 1 is MCV = 0.017. (The sensitivity of the method is 10 ppm-hours or 3.2 mg/m^ for eight hours.) SAMPLING AND ANALYSIS The system for generation of test atmospheres of sulfur dioxide is described in Attachment A. An electrochemical InterScan Model 1240SP analyzer was used to measure and verify the sulfur dioxide chamber concentrations. Test atmospheres were generated with certified mixed gas cylinders. Different dose levels (ppm-hours) were generated by varying the sample exposure time. These dose levels represent the measurement of concentrations between 0.825X and 2.15X the OSHA standard levels. The relative humidity in the chamber was less than 5 percent. Samples were collected as described in Reference 2 with PRO-TEKTM C-20 Colorimetric Badges. Ten samples were collected for each level of exposure for times varying from two to four hours. The results of the analyses of the samples are presented in Table 2. An additional 22 samples were collected at IX the OSHA level. These samples were used for the storage stability study. PRECISION AND ACCURACY The statistical procedures and a definition of the terms used are described in Reference 2 and summarized in Attachment B. R&S166332 3 The precision of the analytical method was assessed with the data in Table 1. The Mean Coefficient of Variation (MCV) for 11 sets of analytical samples was found to be 0.017. Precision and bias of the total sampling and analytical methods were evaluated using the data in Table 2 and the results obtained from storage stability tests (Tables 5 and 6). The MCV for the four sets of samples collected from test atmospheres is 0.058. The mean value for the concentration found by analysis with the badge at each level was compared with the value for the concentration taken with the sulfur dioxide electrochemical analyzer to obtain a measure of the bias of the method. The mean bias for all four levels was found to be - 0.010. The overall -accuracy of the system was found to be + 12.6 percent. ENVIRONMENTAL VARIABLES Additional badges were tested for exposure temperature effects at 2X TLV. The data in Table 3 show no significant temperature effects. Both graphical and statistical analyses of the data from a 24"^ factorial designed set of experiments show no significant effects from pressure, humidity, and face velocity on the Du Pont Ammonia Badge, Type C-10. It is assumed that the same would hold true for the sulfur dioxide badge because the physical dimensions of the diffusion controlling member are the same for both badges. (See Reference 2.) INTERFERENCES The major known interferences are oxides of nitrogen, ozone, and heavy metal salts. These interferences will not affect the PRO-TEKTM Sulfur Dioxide Colorimetric Badge because they have little or no affinity for the absorbing solution. One blister of the sulfur dioxide badge contains 0.6 percent sulfamic acid, which deactivates absorbed nitrogen dioxide. The effectiveness of sulfamic acid is shown in Table 4. STORAGE STABILITY A study was conducted to assess whether sulfur dioxide could be successfully stored as a tetrachlorosulfito mercury complex in exposed badges for one week refrigerated as well as at room temper ature. The results of the analyses are presented in Tables 5 and 6. R&Sl 66333 4 The criterion for acceptance was that the mean of the samples stored under refrigeration or room temperature for seven days should be within + 10 percent of the mean of the set analyzed at the beginning of the storage period. The two means of each set compare within 6 percent; thus, the storage stability was deemed adequate. COLLECTION EFFICIENCY Collection efficiency tests are necessary for dynamic test methods such as the pump/impinger method. Collection efficiencies for the PRO-TEKTM Colorimetric Air Monitoring Badge are constant through out the linear range for which the badge is used. INDEPENDENT METHOD The independent method used to determine the concentration of sulfur dioxide in the exposure chamber consisted of the continuous collec tion of samples at a point immediately after the exposure chamber with a calibrated InterScan Model 1240SP sulfur dioxide electro chemical analyzer. Mixed sulfur dioxide/dry air cylinders of approximately 20 ppm (52 mg/m^) were used. Different dose levels (ppm-hours) were obtained by varying the exposure time. These dose levels represent the measurement of concentrations between 0.8X and 2.2X the OSHA standard level. R&Sl 66334 Attachment A Generation of Test Atmospheres The basic dynamic contaminant generation system used (see Fig ure 1) was originally developed by Du Pont for the purpose of laboratory validation of sampling methods where an accurate measure of the true contaminant concentration could be determined. The system is constructed of glass and Teflon^ FEP-fluorocarbon resin tubing. Exposure chamber gas contaminant concentrations were estab lished with either certified mixed gas cylinders (Scott Environmental, Plumsteadville/ Pennsylvania) or the principle of permeation of an inorganic gas contaminant through a permeation tube. The generated concentration was verified with calibrated on-line instruments such as an infrared spectrophotometer (Perkin-Elmer Model 299) equipped with dual variable long path cells, a direct reading sulfur dioxide analyzer (InterScan Model 1240SP), a direct reading nitrogen dioxide analyzer (InterScan Model 1154SP), and a direct reading NO, N0X, and NH3 analyzer (Thermo Electron Model 10AR). These instruments were calibrated with certified gas cylinders (+ 2.0 percent) traceable to the National Bureau of Standards. All chamber exposures were obtained by integrating the concentration generated during the exposure period. Badges were placed in the chamber under zero concentration. Exposure was initiated by introducing the mixed contaminant in air gas into the chamber. On-line instrumentation and a strip-chart recorder were used to monitor the chamber concentration buildup to steady-state conditions. The resulting time-concentration plot was analyzed with Simpson's rule of approximation on a TI-59 electronic calculator that permits calculation of the integrated exposure limit. Water vapor was added to the system by passing part of the diluent air through a heated gas scrubber that contains water. Tlie actual humidity was determined with an electronic hygrom eter (YSI Model 91HC Dew Point Hygrometer with Model 9102 dual probe). A-l R&S166335 The badge exposure chamber used was a miniaturized wind tunnel made of glass rectangular tubing with a water jacket which allowed temperature control. Chamber temperature was con trolled by pumping an ethylene glycol-water mixture from a refrigerated circulator (Lauda Model RC-3T) through the chamber gi ass jacket. Temperatures were measured by thermocouples at the top and bottom of the chamber. The pressure in the chamber was changed by either increasing flow rates with line restric tion or using an on-line Du Pont Constant Flow Sampler Pump. The total flow rate of the system was measured at the end of the apparatus train with an electronic mass flow meter (Hastings Model ALK-5K, calibrated daily with a soap bubble meter). Flow rates could be varied between 0.1 and 2 liters per minute. PROCEDURES Solution and Reagent Standardization Absorbing solutions and color reagents used in the reagent pack generally are similar to those used in NIOSH published impinger methods. Initial badge color development consisted of: 1. Verification of the linear range of the NIOSH impinger test. 2. Reduction of absorbing solution and color reagent volumes to meet reagent pack dimensions. 3. Determination of absorbing solution and color reagent solution storage stability. All chemicals used were Fisher ACS Grade. All solutions were prepared with distilled water (conductivity, 0.5 /umhos/cm to 2 Mmhos/cm). Absorbance maxima of the color forming species were verified on an ultraviolet scanning spectrophotometer (Perkin-Elmer Model 552). The general analytical procedures followed for method development were similar to NIOSH recom mended procedures. Laboratory Exposure Testing and Analysis The badge exposure chamber can hold up to 12 badges. For PRO-TEK M badges, most tests involved using four to ten badges at standard conditions of 25C, 760 mm Hg, less than 5 percent relative humidity, and 10 ft/min face velocity. The chamber design permits easy exposure replication. Sampling times varied from two to eight hours. A-2 Attachment B Summary of Statistical Terms and Formulae The statistical analysis employed was similar to NIOSH methods (Reference 4) and is described in Reference 5. Some key terms, statistical formulae, acceptable limits, and statistical tests used in these reports are summarized below. Mean = Standard Deviation = CV MCV = Arithmetic mean or average, defined as the sum of all observations divided by the number of observations (n). The positive square root of the variance which is defined as the sum of squares of the deviations of the observations from the mean (x) divided by one less than the total number of observations (n-1). Std. Dev. n I] Ui - x)2 i=l n-1-------- Coefficient of Variation or Relative Standard Deviation, defined as the standard deviation divided by the mean. _ _ Std. Dev. mean Mean (pooled) coefficient of variation; the value is derived from the coefficients of variation obtained from the analysis of 10 samples tested over at least three concen tration levels between 0.5X and 2.5X the OSHA standard level. The mathematical equation is expressed as: MCV where: P (ni"l) CVi (nx-1) (CV!)2 + ... (nj-1) (CVj)2 E mi-1) i=l number of levels. degrees of freedom, equal to number of observations minus one, at the ith level. coefficient of variation of the observations at the ifc^ level. R&S166337 B-l Bias of the ten generated samples at any one concentration. The mathematical equation is expressed as: X^, aver, ppm (badge) - XQ, cert, cylinder concn. (ppm) XQ/ cert, cylinder concn. (ppm) b Mean (pooled) bias; the value is derived from the individual bias obtained from the analyses of ten samples tested over at least three concentration levels between 0.5X and 2.5X the OSHA standard level. The mathematical equation is expressed as: -- ^lnl ^2n2 + * bini OSA where: n = the number of badges exposed at each concentration level, p = the number of levels. Overall System Accuracy; defined as the percent difference between a measured concentration and the true concentration of a sample. The mathematical equation is expressed as: Test for Rejection of an Observation The Q test (2) was used to identify an outlier observation. The spread between the questionable observation and its nearest neighbor is divided by the spread separating the highest and lowest result in a set. This quotient (Q) is compared with the appropriate Q value in the table. The questionable observation may be rejected with 90 percent confidence if the calculated Q value exceeds that in the table below: Number of Observations Q 2 3 0.94 4 0.76 5 0.64 6 0.56 7 0.51 8 0.47 9 0.44 10 0.41 B-2 R&S166338 i REFERENCES 1. D. G. Taylor, R. E. Kupel, J. M. Bryant, Documentation of NIOSH Validation Tests, Publication No. 77-185. 2. E. V. Kring, Ph.D.; W. J. Lautenberger, Ph.D.; W. B. Baker; J. J. Douglas; and R. A. Hoffman, "A New Colorimetric Air Monitoring Badge System for Ammonia, Sulfur Dioxide and Nitrogen Dioxide," submitted for publication to AIHA. 3. D. G. Taylor, R. E. Kupel, J. M. Bryant, Documentation of NIOSH Validation Tests, Back-Up Data Report No. S 308. 4. NIOSH Manual of Analytical Methods, 2nd Ed. Part 1-5 DHEW (NIOSH) Pub. No. 77-157A, 5. R. B. Dean and W. J. Dixon, Anal. Chem. 23:636 (1951). I R&S166339 Table 1 Sulfur Dioxide Analysis Page 1 of 2 Level3 Mg SO3 Taken*3 Mg SO2 Taken Absorbance at 548 nM 00 0 0.325X 2.13 2.13 0.675X 4.13 4.13 0.975X 6.25 6.25 1.375X 8.25 8.25 1.575X 10.38 10.38 0 0 1.7 1.7 3.3 3.3 5.0 5.0 6.6 6.6 8.3 8.3 0 0 n= 2 m= 0 Std. Dev. = 0 cvx = 0 0.12 0.13 n=2 m = 0.125 Std. Dev. = 0.00707 cv2 = 0.0566 0.24 0.24 n=2 m = 0.24 Std. Dev. = 0.00 cv3 = 0.00 0.35 0.34 n=2 m = 0.345 Std. Dev. = 0.00707 CV4 S 0.0205 0.48 0.46 n=2 m = 0.470 Std. Dev. = 0.0141 cv5 = 0.03009 0.56 0.57 n2 m = 0.565 Std. Dev. = 0.00707 cv6 = 0.01252 Equivalent to an eight-hour Time Weighted Average (TWA) con centration for Permissible Exposure Limit (PEL) of 5 ppm (13 mg/m3). "Standardized solutions of sodium sulfite were used to prepare the spiked samples. R&Sl 66340 4 I Table 1 Sulfur Dioxide Analysis (Continued) Page 2 of 2 Levela u.q SOf Takenb 1.85X 12.50 12.50 /*g S02 Taken 10.0 10.0 2.025X 13.50 13.50 10.8 10.8 2.05X 14.5 14.5 2.30X 15.6 15.6 11.6 11.6 12.5 12.5 2.50X 16.6 16.6 13.3 13.3 MCV = 0.017 Absorbance at 548 nM Std. n m Dev. CV7 0.66 0.66 =2 = 0.66 = 0.00 = 0.00 Std. 0.70 0.69 n=2 m = 0.695 Dev. = 0.00707 CV8 = 0.01017 Std. n m Dev. CV9 0.73 0.72 =2 = 0.725 = 0.00707 = 0.00975 Std. n m Dev. CV1Q 0.82 0.85 =2 = 0.835 = 0.0212 = 0.0254 Std. 0.89 0.88 n=2 m = 0.885 Dev. = 0.00707 cvll = -00799 R&S166341 a Equivalent to an eight-hour Time Weighted Average (TWA) con centration for Permissible Exposure Limit (PEL) of 5 ppm (13 mg/m3)# Standardized solutions of sodium sulfite were used to prepare the spiked samples. Page 1 of 2 Table 2 Sulfur Dioxide Sampling and Analysis C Test Level 0.84X Found (020 Badge) (ppm-hrs) EETM mq/nr 34 4.25 11.1 34 4.25 11.1 33 4.13 10.7 33 4.13 10.7 35 4.38 11.4 33 4.13 10.7 31 3.88 10.1 34 4.25 11.1 34 4.25 11.1 29 3.63 9.4 Taken (S02 Analyzer) (ppm- h r s) ppm mg/n? 33.4 4.2 10.9 n = 10 Std. Dev. = 1.764 CV = 0.0534 Bias^ = -0.015 < 1.11X i 49 6.13 15.9 45 5.63 14.6 40 5.00 13.0 44 5.50 14.3 52 6.50 16.9 41 5.13 13.3 44 5.50 14.3 44 5.50 14.3 50 6.25 16.3 46 5.75 15.0 45.4 5.6 14.4 I n = 10 Std. Dev. = 3.837 CV2 - 0.084 Bias2 = +0.0247 I  Page 2 of 2 Table 2 Sulfur Dioxide Sampling and Analysis (Continued) Test Level (OSHA std.) (ppm-hrs) 1.68X 66 65 62 69 67 69 64 67 66 67 n = 10 Std. Dev. = 2.150 cv3 = 0.0325 Bias3 = -0.0149 2.17X 80 87 81 88 90 97 85 81 82 86 Found 22m 8.25 8.13 7.75 8.63 8.38 8.63 8.00 8.38 8.25 8.38 10.0 10.9 10.1 11.0 11.3 12.1 10.6 10.1 10.3 10.8 mg/m3 21.5 21.1 20.2 22.4 21.8 22.4 20.8 21.8 21.5 21.8 26.0 28.3 26.3 28.6 29.3 31.5 27.6 26.3 26.7 28.0 Taken (ppm-hrs) ppm 67.2 8.'4 mg/m3 21.8 82.8 10.4 26.9 n = 10 Std. Dev. = 5.208 cv4 = 0.0608 Bias4 = -0.0350 MCV = 0.058 Mean Bias = -0.010 Overall System Accuracy = + ^Absolute Mean Bias + 2 (MCV)J x 100 OSA = + |.01 + 2 (0.058) x 100 OSA 12.6% R&S166343 5 S' Table 3 Temperature Effect Study at 10.4 ppm (27 mq/m^) TWA for Eiqht Hours Temp. c OK 10.1 283.1 25 298 41.3 314 No. of Badges Tested 8 34 11 Average Absorbance 0.73 0.77 0.75 3pDo O2) 5 4 Table 4 Mg SO2 Added 38.4 38.4 38.4 38.4 38.4 38.4 38.4 38.4 38.4 38.4 Effect of NC>2 (Common Interference) on Analytical Method m1 NO2 Added Mg SO2 Found With Without Sulfamic Acid Sulfamic Acid 0 41.2 41.2 4 41.2 42.0 8 41.3 40.2 12 41.0 22.4 16 41.0 20.2 20 41.0 22.4 24 41.0 16.0 28 41.2 15.6 32 40.4 14.0 40 41.2 11.6 J3 Table 5 Storage Stability Study (Refrigerated) Samples Analyzed Immediately (mg/m3) 14.3 14.9 16.4 14.2 14.3 14.0 15.6 Mean = 14.8 Std. Dev . = 0.886 CV = 0.060 Samples Analyzed After One Week (mg/m3) 15.5 14.9 14.4 15.9 13.8 14.0 15.0 Mean = 14.8 Std. Dev. = 0.769 CV = 0.052 Table 6 Storage Stability Study (Room Temperature) Samples Analyzed Immediately (mg/m3) 14.6 14.6 13.0 13.8 Mean = 14.0 Std. Dev . = 0.766 CV = 0.055 Samples Analyzed After Seven Days at Room Temperature (mg/m3) 13.4 12.3 13.3 13.8 Mean 13.2 Std. Dev. = 0.637 CV 0.048 R&SI 66346 tr- %. FIGURE 1 LIQUID SORBENT LABORATORY VALIDATION APPARATUS Dryer Recorder Temp./Flow Hastings Flow Meter Vent Thermocouple Lead -- Chemiluminescence Analyzer for NH3 Thermoelectron Model 10 AR S02 Analyzer InterScan 1240 SP or No2 Analyzer InterScan 1154 SP Certified Mixed Gas Cylinder R&S166347