Appendix A--Reference Method for the Determination of Sulfur Dioxide in the Atmosphere (Pararosaniline Method) 1.0 Applicability. 1.1 This method provides a measurement of the concentration of sulfur dioxide (SO2) in ambient air for determining compliance with the primary and secondary national ambient air quality standards for sulfur oxides (sulfur dioxide) as specified in Sec. 50.4 and Sec. 50.5 of this chapter. The method is applicable to the measurement of ambient SO2 concentrations using sampling periods ranging from 30 minutes to 24 hours. Additional quality assurance procedures and guidance are provided in Part 58, Appendixes A and B, of this chapter and in references 1 and 2. 2.0 Principle. 2.1 A measured volume of air is bubbled through a solution of 0.04 M potassium tetrachloromercurate (TCM). The SO2 present in the air stream reacts with the TCM solution to form a stable monochlorosulfonatomercurate(3) complex. Once formed, this complex resists air oxidation(4, 5) and is stable in the presence of strong oxidants such as ozone and oxides of nitrogen. During subsequent analysis, the complex is reacted with acid-bleached pararosaniline dye and formaldehyde to form an intensely colored pararosaniline methyl sulfonic acid.(6) The optical density of this species is determined spectrophotometrically at 548 nm and is directly related to the amount of SO2 collected. The total volume of air sampled, corrected to EPA reference conditions (25 deg. C, 760 mm Hg [101 kPa]), is determined from the measured flow rate and the sampling time. The concentration of SO2 in the ambient air is computed and expressed in micrograms per standard cubic meter (micro-g/std m**3). 3.0 Range. 3.1 The lower limit of detection of SO2 in 10 mL of TCM is 0.75 micro-g (based on collaborative test results).(7) This represents a concentration of 25 micro-g SO2/m**3 (0.01 ppm) in an air sample of 30 standard liters (short-term sampling) and a concentration of 13 micro-g SO2/m3 (0.005 ppm) in an air sample of 288 standard liters (long-term sampling). Concentrations less than 25 micro-g SO2/m**3 can be measured by sampling larger volumes of ambient air; however, the collection efficiency falls off rapidly at low concentrations.(8, 9) Beer's law is adhered to up to 34 micro-g of SO2 in 25 mL of final solution. This upper limit of the analysis range represents a concentration of 1,130 micro-g SO2/m**3 (0.43 ppm) in an air sample of 30 standard liters and a concentration of 590 micro-g SO2/m**3 (0.23 ppm) in an air sample of 288 standard liters. Higher concentrations can be measured by collecting a smaller volume of air, by increasing the volume of absorbing solution, or by diluting a suitable portion of the collected sample with absorbing solution prior to analysis. 4.0 Interferences. 4.1 The effects of the principal potential interferences have been minimized or eliminated in the following manner: Nitrogen oxides by the addition of sulfamic acid,(10, 11) heavy metals by the addition of ethylenediamine tetracetic acid disodium salt (EDTA) and phosphoric acid,(10, 12) and ozone by time delay.(10) Up to 60 micro-g Fe (III), 22 micro-g V (V), 10 micro-g Cu (II), 10 micro-g Mn (II), and 10 micro-g Cr (III) in 10 mL absorbing reagent can be tolerated in the procedure.(10) No significant interference has been encountered with 2.3 micro-g NH3.(13) 5.0 Precision and Accuracy. 5.1 The precision of the analysis is 4.6 percent (at the 95 percent confidence level) based on the analysis of standard sulfite samples.(10) 5.2 Collaborative test results (14) based on the analysis of synthetic test atmospheres (SO2 in scrubbed air) using the 24-hour sampling procedure and the sulfite-TCM calibration procedure show that: * The replication error varies linearly with concentration from +/-2.5 micro-g/cubic m at concentrations of 100 micro- g/cubic m to +/-7 micro-g/cubic m at concentrations of 400 micro-g/cubic m. * The day-to-day variability within an individual laboratory (repeatability) varies linearly with concentration from +/-18.1 micro-g/m**3 at levels of 100 micro-g/m**3 to +/-50.9 micro-g/m**3 at levels of 400 micro-g/m**3. * The day-to-day variability between two or more laboratories (reproducibility) varies linearly with concentration from +/-36.9 micro-g/m**3 at levels of 100 micro-g/m**3 to +/-103.5 micro-g/m**3 at levels of 400 micro-g/m**3. * The method has a concentration-dependent bias, which becomes significant at the 95 percent confidence level at the high concentration level. Observed values tend to be lower than the expected SO2 concentration level. 6.0 Stability. 6.1 By sampling in a controlled temperature environment of 15 deg.+/-10 deg. C, greater than 98.9 percent of the SO2-TCM complex is retained at the completion of sampling. (15) If kept at 5 deg. C following the completion of sampling, the collected sample has been found to be stable for up to 30 days.(10) The presence of EDTA enhances the stability of SO2 in the TCM solution and the rate of decay is independent of the concentration of SO2.(16) 7.0 Apparatus. 7.1 Sampling. 7.1.1 Sample probe: A sample probe meeting the requirements of Section 7 of 40 CFR Part 58, Appendix E (Teflon(R) or glass with residence time less than 20 sec.) is used to transport ambient air to the sampling train location. The end of the probe should be designed or oriented to preclude the sampling of precipitation, large particles, etc. A suitable probe can be constructed from Teflon(R) tubing connected to an inverted funnel. 7.1.2 Absorber--short-term sampling: An all glass midget impinger having a solution capacity of 30 mL and a stem clearance of 4+/-1 mm from the bottom of the vessel is used for sampling periods of 30 minutes and 1 hour (or any period considerably less than 24 hours). Such an impinger is shown in Figure 1. These impingers are commercially available from distributors such as Ace Glass, Incorporated. 7.1.3 Absorber--24-hour sampling: A polypropylene tube 32 mm in diameter and 164 mm long (available from Bel Art Products, Pequammock, NJ) is used as the absorber. The cap of the absorber must be a polypropylene cap with two ports (rubber stoppers are unacceptable because the absorbing reagent can react with the stopper to yield erroneously high SO2 concentrations). A glass impinger stem, 6 mm in diameter and 158 mm long, is inserted into one port of the absorber cap. The tip of the stem is tapered to a small diameter orifice (0.4+/-0.1 mm) such that a No. 79 jeweler's drill bit will pass through the opening but a No. 78 drill bit will not. Clearance from the bottom of the absorber to the tip of the stem must be 6+/-2 mm. Glass stems can be fabricated by any reputable glass blower or can be obtained from a scientific supply firm. Upon receipt, the orifice test should be performed to verify the orifice size. The 50 mL volume level should be permanently marked on the absorber. The assembled absorber is shown in Figure 2. 7.1.4 Moisture trap: A moisture trap constructed of a glass trap as shown in Figure 1 or a polypropylene tube as shown in Figure 2 is placed between the absorber tube and flow control device to prevent entrained liquid from reaching the flow control device. The tube is packed with indicating silica gel as shown in Figure 2. Glass wool may be substituted for silica gel when collecting short-term samples (1 hour or less) as shown in Figure 1, or for long term (24 hour) samples if flow changes are not routinely encountered. 7.1.5 Cap seals: The absorber and moisture trap caps must seal securely to prevent leaks during use. Heat-shrink material as shown in Figure 2 can be used to retain the cap seals if there is any chance of the caps coming loose during sampling, shipment, or storage. [ ...Illustration appears here... ] Figure 1. Short term sampling train. [ ...Illustration appears here... ] Figure 2. 24-hour sampling system. 7.1.6 Flow control device: A calibrated rotameter and needle valve combination capable of maintaining and measuring air flow to within +/-2 percent is suitable for short-term sampling but may not be used for long-term sampling. A critical orifice can be used for regulating flow rate for both long-term and short-term sampling. A 22-gauge hypodermic needle 25 mm long may be used as a critical orifice to yield a flow rate of approximately 1 L/min for a 30-minute sampling period. When sampling for 1 hour, a 23-gauge hypodermic needle 16 mm in length will provide a flow rate of approximately 0.5 L/min. Flow control for a 24-hour sample may be provided by a 27-gauge hypodermic needle critical orifice that is 9.5 mm in length. The flow rate should be in the range of 0.18 to 0.22 L/min. 7.1.7 Flow measurement device: Device calibrated as specified in 9.4.1 and used to measure sample flow rate at the monitoring site. 7.1.8 Membrane particle filter: A membrane filter of 0.8 to 2 micro-m porosity is used to protect the flow controller from particles during long-term sampling. This item is optional for short-term sampling. 7.1.9 Vacuum pump: A vacuum pump equipped with a vacuum gauge and capable of maintaining at least 70 kPa (0.7 atm) vacuum differential across the flow control device at the specified flow rate is required for sampling. 7.1.10 Temperature control device: The temperature of the absorbing solution during sampling must be maintained at 15 deg. +/-10 deg. C. As soon as possible following sampling and until analysis, the temperature of the collected sample must be maintained at 5 deg. +/-5 deg. C. Where an extended period of time may elapse before the collected sample can be moved to the lower storage temperature, a collection temperature near the lower limit of the 15 +/- 10 deg. C range should be used to minimize losses during this period. Thermoelectric coolers specifically designed for this temperature control are available commercially and normally operate in the range of 5 deg. to 15 deg. C. Small refrigerators can be modified to provide the required temperature control; however, inlet lines must be insulated from the lower temperatures to prevent condensation when sampling under humid conditions. A small heating pad may be necessary when sampling at low temperatures (<7 deg. C) to prevent the absorbing solution from freezing.(17) 7.1.11 Sampling train container: The absorbing solution must be shielded from light during and after sampling. Most commercially available sampler trains are enclosed in a light- proof box. 7.1.12 Timer: A timer is recommended to initiate and to stop sampling for the 24-hour period. The timer is not a required piece of equipment; however, without the timer a technician would be required to start and stop the sampling manually. An elapsed time meter is also recommended to determine the duration of the sampling period. 7.2 Shipping. 7.2.1 Shipping container: A shipping container that can maintain a temperature of 5 deg. +/-5 deg. C is used for transporting the sample from the collection site to the analytical laboratory. Ice coolers or refrigerated shipping containers have been found to be satisfactory. The use of eutectic cold packs instead of ice will give a more stable temperature control. Such equipment is available from Cole-Parmer Company, 7425 North Oak Park Avenue, Chicago, IL 60648. 7.3 Analysis. 7.3.1 Spectrophotometer: A spectrophotometer suitable for measurement of absorbances at 548 nm with an effective spectral bandwidth of less than 15 nm is required for analysis. If the spectrophotometer reads out in transmittance, convert to absorbance as follows: A=log10(1/T) (1) where: A=absorbance, and T=transmittance (010 mL of absorbing solution was used, bring the absorber solution in each impinger to orginal volume with distilled H2O and pipet 10-mL portions from each impinger into a series of 25-mL volumetric flasks. If the color development steps are not to be started within 12 hours of sampling, store the solutions at 5 deg. +/- 5 deg. C. Calculate the total micrograms SO2 in each solution as follows: Ca x Qs x t x Va x 10-3 micro-g SO2 = ------------------------ V b where: Ca=concentration of SO2 in the standard atmosphere, micro-g/m**3; Os=sampling flow rate, std L/min; t=sampling time, min; Va=volume of absorbing solution used for color development (10mL); and Vb=volume of absorbing solution used for sampling, mL. Add the remaining reagents for color development in the same manner as in Section 10.2 for static solutions. Calculate a calibration equation and a calibration factor (Bg) according to Section 10.2, adhering to all the specified criteria. 10.3.2 Calibration Option 2B (24-hour samples): Generate a standard atmosphere containing approximately 1,050 micro-g SO2/m**3 and calculate the exact concentration according to equation 12. Set up a series of six absorbers according to Figure 2 and connect to a common manifold for sampling the standard atmosphere. Be sure that the total flow rate of the standard exceeds the flow demand at the sample manifold, with the excess flow vented at atmospheric pressure. The absorbers are then allowed to sample the atmosphere for varying time periods to yield solutions containing 0, 0.2, 0.6, 1.0, 1.4, 1.8, and 2.2 micro-g SO2/mL solution. The sampling times required to attain these solution concentrations are calculated as follows: Vb x Cs t = -------------- Ca x Qs x 10-3 where: t=sampling time, min; Vb=volume of absorbing solution used for sampling (50 mL); Cs=desired concentration of SO2 in the absorbing solution, micro- g/mL; Ca=concentration of the standard atmosphere calculated according to equation 12, micro-g/m**3 ; and Qs=sampling flow rate, std L/min. At the completion of sampling, bring the absorber solutions to original volume with distilled water. Pipet a 10-mL portion from each absorber into one of a series of 25-mL volumetric flasks. If the color development steps are not to be started within 12 hours of sampling, store the solutions at 5 deg. +/- 5 deg. C. Add the remaining reagents for color development in the same manner as in Section 10.2 for static solutions. Calculate the total micro-g SO2 in each standard as follows: Ca x Qs x t x Va x 10-3 micro-g SO2 = ------------------------ V b where: Va=volume of absorbing solution used for color development (10 mL). All other parameters are defined in equation 14. Calculate a calibration equation and a calibration factor (Bt) according to Section 10.2 adhering to all the specified criteria. 11.0 Sample Preparation and Analysis. 11.1 Sample Preparation. Remove the samples from the shipping container. If the shipment period exceeded 12 hours from the completion of sampling, verify that the temperature is below 10 deg. C. Also, compare the solution level to the temporary level mark on the absorber. If either the temperature is above 10 deg. C or there was significant loss (more than 10 mL) of the sample during shipping, make an appropriate notation in the record and invalidate the sample. Prepare the samples for analysis as follows: 1. For 30-minute or 1-hour samples: Quantitatively transfer the entire 10 mL amount of absorbing solution to a 25-mL volumetric flask and rinse with a small amount (<5 mL) of distilled water. 2. For 24-hour samples: If the volume of the sample is less than the original 50-mL volume (permanent mark on the absorber), adjust the volume back to the original volume with distilled water to compensate for water lost to evaporation during sampling. If the final volume is greater than the original volume, the volume must be measured using a graduated cylinder. To analyze, pipet 10 mL of the solution into a 25-mL volumetric flask. 11.2 Sample Analysis. For each set of determinations, prepare a reagent blank by adding 10 mL TCM absorbing solution to a 25-mL volumetric flask, and two control standards containing approximately 5 and 15 micro-g SO2, respectively. The control standards are prepared according to Section 10.2 or 10.3. The analysis is carried out as follows: 1. Allow the sample to stand 20 minutes after the completion of sampling to allow any ozone to decompose (if applicable). 2. To each 25-mL volumetric flask containing reagent blank, sample, or control standard, add 1 mL of 0.6% sulfamic acid (Section 8.2.1) and allow to react for 10 min. 3. Accurately pipet 2 mL of 0.2% formaldehyde solution (Section 8.2.2) and then 5 mL of pararosaniline solution (Section 8.2.13) into each flask. Start a laboratory timer set at 30 minutes. 4. Bring each flask to volume with recently boiled and cooled distilled water and mix thoroughly. 5. During the 30 minutes, the solutions must be in a temperature controlled environment in the range of 20 deg. to 30 deg. C maintained to +/- 1 deg. C. This temperature must also be within 1 deg. C of that used during calibration. 6. After 30 minutes and before 60 minutes, determine the corrected absorbances (equation 10) of each solution at 548 nm using 1-cm optical path length cells against a distilled water reference (Section 10.1). (Distilled water is used as a reference instead of the reagent blank because of the sensitivity of the reagent blank to temperature.) 7. Do not allow the colored solution to stand in the cells because a film may be deposited. Clean the cells with isopropyl alcohol after use. 8. The reagent blank must be within 0.03 absorbance units of the intercept of the calibration equation determined in Section 10. 11.3 Absorbance range. If the absorbance of the sample solution ranges between 1.0 and 2.0, the sample can be diluted 1:1 with a portion of the reagent blank and the absorbance redetermined within 5 minutes. Solutions with higher absorbances can be diluted up to sixfold with the reagent blank in order to obtain scale readings of less than 1.0 absorbance unit. However, it is recommended that a smaller portion (<10 mL) of the original sample be reanalyzed (if possible) if the sample requires a dilution greater than 1:1. 11.4 Reaqent disposal. All reagents containing mercury compounds must be stored and disposed of using one of the procedures contained in Section 13. Until disposal, the discarded solutions can be stored in closed glass containers and should be left in a fume hood. 12.0 Calculations. 12.1 Calibration Slope, Intercept, and Correlation Coefficient. The method of least squares is used to calculate a calibration equation in the form of: y=mx+b (16) where: y=corrected absorbance, m=slope, absorbance unit/micro-g SO2, x=micrograms of SO2, b=y intercept (absorbance units). The slope (m), intercept (b), and correlation coefficient (r) are calculated as follows: n S xy - ( S x) ( S y) m = ------------------ n S x2 - (S x)2 S y-mS x b = ---------- n [ ...Illustration appears here... ] where n is the number of calibration points. A data form (Figure 5) is supplied for easily organizing calibration data when the slope, intercept, and correlation coefficient are calculated by hand. 12.2 Total Sample Volume. Determine the sampling volume at standard conditions as follows: Qi+Qf Vstd = -------- x t (20) 2 where: Vstd=sampling volume in std L, Qi=standard flow rate determined at the initiation of sampling in std L/min, Qf=standard flow rate determined at the completion of sampling is std L/min, and t=total sampling time, min. 12.3 Sulfur Dioxide Concentration. Calculate and report the concentration of each sample as follows: (A-Ao)(Bx)(103) Vb micro-g SO2/m3 = --------------- x -- Vstd Va (21) where: A=corrected absorbance of the sample solution, from equation (10); Ao=corrected absorbance of the reagent blank, using equation (10); Bx=calibration factor equal to Bs, Bg, or Bt depending on the calibration procedure used, the reciprocal of the slope of the calibration equation; Va=volume of absorber solution analyzed, mL; Vb=total volume of solution in absorber (see 11.1-2), mL; and Vstd=standard air volume sampled, std L (from Section 12.2). Data Form [For hand calculations] Calibration Micrograms Absorbance point no. So2 units (x) (y) x2 xy y2 1 2 3 4 5 6 Sx=YYY Sy=YYY Sx**2=YYY SxyYYY Sy**2YYY n=YYY (number of pairs of coordinates.) -------------------------------------------------------------------- Figure 5. Data form for hand calculations. 12.4 Control Standards. Calculate the analyzed micrograms of SO2 in each control standard as follows: Cq=(A-Ao)xBx (22) where: Cq=analyzed micro-g SO2 in each control standard, A=corrected absorbance of the control standard, and Ao=corrected absorbance of the reagent blank. The difference between the true and analyzed values of the control standards must not be greater than 1 micro-g. If the difference is greater than 1 micro-g, the source of the discrepancy must be identified and corrected. 12.5 Conversion of micro-g/m**3to ppm (v/v). If desired, the concentration of sulfur dioxide at reference conditions can be converted to ppm SO2 (v/v) as follows: micro-g SO2 ppm SO2 = --------- x 3.82 x 10-4 m3 (23) 13.0 The TCM absorbing solution and any reagents containing mercury compounds must be treated and disposed of by one of the methods discussed below. Both methods remove greater than 99.99 percent of the mercury. 13.1 Disposal of Mercury-Containing Solutions. 13.2 Method for Forming an Amalgam. 1. Place the waste solution in an uncapped vessel in a hood. 2. For each liter of waste solution, add approximately 10 g of sodium carbonate until neutralization has occurred (NaOH may have to be used). 3. Following neutralization, add 10 g of granular zinc or magnesium. 4. Stir the solution in a hood for 24 hours. Caution must be exercised as hydrogen gas is evolved by this treatment process. 5. After 24 hours, allow the solution to stand without stirring to allow the mercury amalgam (solid black material) to settle to the bottom of the waste receptacle. 6. Upon settling, decant and discard the supernatant liquid. 7. Quantitatively transfer the solid material to a container and allow to dry. 8. The solid material can be sent to a mercury reclaiming plant. It must not be discarded. 13.3 Method Using Aluminum Foil Strips. 1. Place the waste solution in an uncapped vessel in a hood. 2. For each liter of waste solution, add approximately 10 g of aluminum foil strips. If all the aluminum is consumed and no gas is evolved, add an additional 10 g of foil. Repeat until the foil is no longer consumed and allow the gas to evolve for 24 hours. 3. Decant the supernatant liquid and discard. 4. Transfer the elemental mercury that has settled to the bottom of the vessel to a storage container. 5. The mercury can be sent to a mercury reclaiming plant. It must not be discarded. 14.0 References for SO2 Method. 1. Quality Assurance Handbook for Air Pollution Measurement Systems, Volume I, Principles. EPA-600/9-76-005, U.S. Environmental Protection Agency, Research Triangle Park, North Carolina 27711, 1976. 2. Quality Assurance Handbook for Air Pollution Measurement Systems, Volume II, Ambient Air Specific Methods. EPA-600/4-77-027a, U.S. Environmental Protection Agency, Research Triangle Park, North Carolina 27711, 1977. 3. Dasqupta, P. K., and K. B. DeCesare. Stability of Sulfur Dioxide in Formaldehyde and Its Anomalous Behavior in Tetrachloromercurate (II). Submitted for publication in Atmospheric Environment, 1982. 4. West, P. W., and G. C. Gaeke. Fixation of Sulfur Dioxide as Disulfitomercurate (II) and Subsequent Colorimetric Estimation. Anal. Chem., 28:1816, 1956. 5. Ephraim, F. Inorganic Chemistry. P. C. L. Thorne and E. R. Roberts, Eds., 5th Edition, Interscience, 1948, p. 562. 6. Lyles, G. R., F. B. Dowling, and V. J. Blanchard. Quantitative Determination of Formaldehyde in the Parts Per Hundred Million Concentration Level. J. Air. Poll. Cont. Assoc., Vol. 15(106), 1965. 7. McKee, H. C., R. E. Childers, and O. Saenz, Jr. Collaborative Study of Reference Method for Determination of Sulfur Dioxide in the Atmosphere (Pararosaniline Method). EPA- APTD-0903, U.S. Environmental Protection Agency, Research Triangle Park, North Carolina 27711, September 1971. 8. Urone, P., J. B. Evans, and C. M. Noyes. Tracer Techniques in Sulfur--Air Pollution Studies Apparatus and Studies of Sulfur Dioxide Colorimetric and Conductometric Methods. Anal. Chem., 37: 1104, 1965. 9. Bostrom, C. E. The Absorption of Sulfur Dioxide at Low Concentrations (pphm) Studied by an Isotopic Tracer Method. Intern. J. Air Water Poll., 9:333, 1965. 10. Scaringelli, F. P., B. E. Saltzman, and S. A. Frey. Spectrophotometric Determination of Atmospheric Sulfur Dioxide. Anal. Chem., 39: 1709, 1967. 11. Pate, J. B., B. E. Ammons, G.A. Swanson, and J. P. Lodge, Jr. Nitrite Interference in Spectrophotometric Determination of Atmospheric Sulfur Dioxide. Anal. Chem., 37:942, 1965. 12. Zurlo, N., and A. M. Griffini. Measurement of the Sulfur Dioxide Content of the Air in the Presence of Oxides of Nitrogen and Heavy Metals. Medicina Lavoro, 53:330, 1962. 13. Rehme, K. A., and F. P. Scaringelli. Effect of Ammonia on the Spectrophotometric Determination of Atmospheric Concentrations of Sulfur Dioxide. Anal. Chem., 47:2474, 1975. 14. McCoy, R. A., D. E. Camann, and H. C. McKee. Collaborative Study of Reference Method for Determination of Sulfur Dioxide in the Atmosphere (Pararosaniline Method) (24-Hour Sampling). EPA-650/4-74-027, U.S. Environmental Protection Agency, Research Triangle Park, North Carolina 27711, December 1973. 15. Fuerst, R. G. Improved Temperature Stability of Sulfur Dioxide Samples Collected by the Federal Reference Method. EPA-600/4-78-018, U.S. Environmental Protection Agency, Research Triangle Park, North Carolina 27711, April 1978. 16. Scaringelli, F. P., L. Elfers, D. Norris, and S. Hochheiser. Enhanced Stability of Sulfur Dioxide in Solution. Anal. Chem., 42:1818, 1970. 17. Martin, B. E. Sulfur Dioxide Bubbler Temperature Study. EPA-600/4-77-040, U.S. Environmental Protection Agency, Research Triangle Park, North Carolina 27711, August 1977. 18. American Society for Testing and Materials. ASTM Standards, Water; Atmospheric Analysis. Part 23. Philadelphia, Pennsylvania, October 1968, p. 226. 19. O'Keeffe, A. E., and G. C. Ortman. Primary Standards for Trace Gas Analysis. Anal. Chem., 38:760, 1966. 20. Scaringelli, F. P., S. A. Frey, and B. E. Saltzman. Evaluation of Teflon Permeation Tubes for Use with Sulfur Dioxide. Amer. Ind. Hygiene Assoc. J., 28:260, 1967. 21. Scaringelli, F. P., A. E. O'Keeffe, E. Rosenberg, and J. P. Bell, Preparation of Known Concentrations of Gases and Vapors With Permeation Devices Calibrated Gravimetrically. Anal. Chem., 42:871, 1970. 22. A Procedure for Establishing Traceability of Gas Mixtures to Certain National Bureau of Standards Standard Reference Materials. EPA-600/7-81-010, U.S. Environmental Protection Agency, Environmental Monitoring Systems Laboratory (MD-77), Research Triangle Park, North Carolina 27711, January 1981. [47 FR 54899, Dec. 6, 1982; 48 FR 17355, Apr. 22, 1983]