Back CRL QAFP
Section: Appendix 6.9.1
Revision No: 0
Date: 3 March 88
Page :12
STANDARD OPERATING PROCEDURE
FOR THE
DETERMINATION OF ARSENIC IN SUSPENDED PARTICULATE MATTER
COLLECTED FROM AMBIENT AIR
AS MEASURED BY GRAPHITE FURNACE ATOMIC ABSORPTION
United States Environmental Protection Agency
Region 5 Central Regional Lbaoratory
536 South Clark St. (5SCRL)
Chicago, Il 60605
Date:
CONCURRENCES
TEAM LEADER
Signed John V. Morris 3/3/88 SECTION CHIEF
------------------------ QUALITY CONTROL COORDINATOR
Signed David A. Payne 3/3/88 LABORATORY DIRECTOR
Signed
STANDARD OPERATING PROCEDURE
For the
Determination of Arsenic in Suspended Particulate Matter
Collected from Ambient Air
TABLE OF CONTENTS
SECTION
REV DATE 1. SCOPE AND APPLICATION
0 2/88 2. SAFETY AND WASTE HANDLING
0 2/88 3. SUMMARY OF METHOD
0 2/88 4. SAMPLE HANDLING AND PRESERVATION
0 2/88 5. INTERFERENCES
0 2/88 6. APPARATUS
0 2/88 7. REAGENTS
0 2/88 8. INSTRUMENT CALIBRATION
0 2/88 9. ANALYTICAL PROCEDURE
0 2/88 10.CALCULATIONS
0 2/88 11.QUALITY CONTROL
0 2/88 12. INSTRUMENT MAINTENANCE
0 2/88 REFERENCES
0 2/88
STANDARD OPERATING PROCEDURE
for the
Determination of Arsenic in Suspended Particulate Matter
Collected from Ambient Air
as Measured by Graphite Furnace Atomic Absorption
1. SCOPE AND APPLICATION
1.1 This method is applicable to the determination of arsenic in suspended particulate matter collected from ambient air by means of a Hi-Vol sampler, on glass fiber filters or quartz fiber filters.
The detection limit is dependent on the volume of air used in the collection of the particulates and the sensitivity of the instrument used for the determination. The concentration of an element due to the filter (filter blank) must be determined for each lot of filters. If this is higher than the normal detection limit the higher value must be used.
1.2 This method assumes total suspended particulate (TSP) has been requested in the same survey. Filters will be preweighed by CRL, and a sufficient number from the same lot must be available to CRL for filter blank analyses prior to taring off filters. Blank filters must also be provided with the survey.
1.3 Requestor must provide information with the survey to indicate whether distribution of the analyte on the filter is expected to be uniform, or if triplicate filter strips are to be analyzed from each filter.
2. SAFETY AND WASTE HANDLING
2.1 Arsenic compounds are toxic and should be handled with care.All unused solutions shall be disposed of in the red labeled waste containers.
2.2 Nitric acid is a strong oxidizer. Avoid skin contact by wearing gloves. All waste nitric acid solutions should be poured into the red labeled waste container, or neutralized with sodium bicarbonate before disposing of them in the sink. Nitric acid fumes can be toxic, and are a strong
irritant to eyes and nasal membranes even in small doses. Perform all operations with nitric acid in a hood.
3. SUMMARY OF METHOD
3.1 Particulate matter suspended in ambient air is collected for 24 hours on a glass fiber filter using a high volume air sampler. After the filter is weighed for TSP determination, a portion (one ninth) of the filter is extracted with nitric acid, and the extract is analyzed by graphite furnace absorption spectrophotometry, using the method of standard additions.
3.2 The concentration of arsenic in the air sample is calculated from the concentration determined in the extract, the fraction of the filter taken, and the volume of air passed through the Hi-Vol sampler.
4. SAMPLE HANDLING
4.1 Glass fiber filters to be analyzed for metals must first be weighed for total suspended particulates (TSP).
4.2 Filters and strips should always be handled with forceps to avoid contamination. Care should be taken to prevent loss of particulate matter from the filter. Filters are normally folded in half for this purpose.
4.3 There is no holding time established for this analysis.
5. INTERFERENCES
5.1 Chemical interferences are not known to be a problem with the extraction, alt'nough arsenic species can be volatile. This extraction was checked with solutions of cacodylic acid, to verify that arsenic was not lost.
5.2 Spectral interferences due to line overlap are well known and characterized. Zeeman background correction is recommended for all arsenic analyses using graphite furnace atomic absorption.
5.3 Nickel nitrate modifier is used with all arsenic graphite furnace analyses. The method of standard
additions is to be used until spiking studies have shown that standard curve is appropriate.
6. APPARATUS
6.1 Perkin-Elmer Zeeman 5000 Atomic Absorption Spectrophotometer
6.1.1 HGA 500 Graphite Furnace and Programmer
6.1.2 AS40-Automatic Sampler
6.1.3 Perkin-Elmer Model 3600 Data Station
6.1.4 PRS10 Printer Sequencer
6.2 Lamps and lamp accessories
6.2.1 Hollow-Cathode Lamp (HCL)
6.2.2 Electrodeless Discharge Lamp (EDL)
6.2.3 EDL power supplies, single and dual
6.3 Graphite Sample Tubes and Contact Rings
6.3.1 Non-pyrolytic graphite tubes; Part #0290-1633
6.3.2 Pyrolytically-coated graphite tubes; Part #091-504
6.3.3 Graphite contact rings (cones); Part #0290-1783
6.4 Glassware and Accessories
6.4.1 Eppendorf pipets and tips
6.4.2 100 mL volumetri,l flasks
6.4.3 30 mL polystyrene sample cups
6.4.4 2 mL polystyrene or Teflon sample cups
6.5 Volumetric flasks for graphite furnace atomic absorption are kept separate from flasks used elsewhere in the laboratory. These flasks are not washed with the rest of the Metals glassware, but rather are rinsed thoroughly with both tap and deionized distilled water and stored with approximately 5% HN03 until ready for use. The acid should touch the bottom of the cap, and the flask should so stand for at least 2 or 3 days before reuse. The acid is discarded and the flask again rinsed with tap and deionized distilled water before use.
6.6 For most applications, it is best to keep a supply of sample cups in Zip-loc plastic bags until use to protect them from room dust. Similarly, the Eppendorf pipet tips should be kept in their bag with a hole only large enough to reach a pipet into the bag to retrieve a tip. If the tips come in boxes, close the box after each tip is removed.
6.7 Microwave Digestion Apparatus
6.7.2 Teflon lined laboratory microwave oven from CEM, MDS 81D
6.7.3 Vented Teflon pressure vessels, top and bottom numbered to match. Use only matching pieces. Be sure insert in top is smooth side toward the vessel.
6.7.4 Constant torque closure apparatus for the pressure vessels
6.7.5 Twelve place circular sample holder with vent connections for the pressure vessels
6.7.6 Centrifuge tubes, about 30mm X 95mm, Oak Ridge style, polyalloy, with polyethylene screw caps, 30 ml size
6.8 T square with a one inch wide arm for cutting or cutting board with appropriate markings for even cutting
6.9 Single-edged razor blade or pizza cutter
6.10 Linear polyethylene bottles, 125 mL, and polypropylene caps
6.11 Forceps and protective disposable gloves for handling the filters and pieces
6.12 Disposable particle mask to be worn while cutting filters
6.13 Glass fiber filters: Before a new lot of glass fiber filters is preweighed for the sampling personnel, a representative sample of the blank filters should be analyzed to determine metals content. Use the method described in 40 CFR, Part 50, Appendix G, Section 6.1.1.2 to select filters for analysis.
6.14 Centrifuge
7. REAGENTS
7.1 Distilled Water: The laboratory distilled water is passed through an ultrapure mixed-bed resin before use. All water, unless otherwise specified, has been passed through this resin (Super Q).
7.2 Nitric Acid: Baker Instra-Analyzed may be used as received or as diluted (1:1) for ICP digestions
7.3 Hydrochloric Acid: Baker Instra-Analyzed may be used as received or as diluted (1:1) for ICP digestions
7.4 Mix 6.4 mL conc. nitric acid and 182 mL conc. hydrochloric acid dilute to 1 liter.
7.5 Atomic Absorption Standard: 1000 mg As/L. May use Fisher, Ventron, or Spex brands have been found acceptable. Note that Ventron or Spex standards may deviate from the 1000 mg/L by a few percent, but the true concentration is stated on the label.
7.6 Matrix Modifiers
7.6.1 Nickel Nitrate: Dissolve 24.77 g of Aesar Ni(NO3)2.6 H20 in water and dilute to lOOmL.
7.6.2 Hydrogen Peroxide: Commercial 30% dydrogen peroxide may be used as received. Keep refrigerated.
7.7 Stock Standard: Dilute 100 uL (0.1 mi) of the 1000 mg/L As solution (7.5) in a 100 mL volumetric flask, to which as been added 9.5 mL concentrated nitric acid and about 500 mL of water. Dilute to volume with water. This solution may be kept for a week.
7.8 Working Standards: Fill four 100 mL volumetric flasks about halfway with water. Add 0.5 mL of concentrated nitric acid to each flask, and wash down the inside of the neck with water. Add 2 mL of nickel nitrate modifier (7.6.1), and 2 mL of hydrogen peroxide (7.6.2) to each flask. Add to three of the flasks 1, 2, and 3 mL of stock (7.7). Dilute all four flasks to volume. The four standards are now ready: 0, 10, 20, and 30 ug As/L.
7.9 AQC: the AQC is prepared by diluting a concentrated solution from an ampule with 300 mL of the acid mixture (7.4) to 1 L. The current AQC is made from WP284 concentrate #2, by diluting 1 mL into 1 L.
8. INSTRUMENT CALIBRATION
8.1 For detailed intructions on the calibration of the Zeeman 5000 atomic absorption spectrophotometwer, see the Instrument Operating Manual.
9. ANALYTICAL PROCEDURE
9.1 Sample preparation
Cut a 1" x 8" strip from the exposed filter using the template and the razor blade. The strip should be taken away from the edge of the filter, preferably near the center. If samplling information indicates that the distribution of metals on the filter may be variable, such as sampling near a roadway, a minimum of three strips shall be taken from the filter. Care should be taken not to tear the filter during cutting. Cut slowly and support the filter with the T square or a ruler.
9.1.2 Fold the strip in half twice or coil it and place it in the bottom of a centrifuge tube (6.3.6).
9.1.3 Add 15 mL of acid mixture (7.4) to cover the filter and loosely cap.
9.1.4 Place the loosely capped centrifuge tube in the Teflon microwave vessel. Tighten the cap on the microwave digestion vessel in the capping station.
9.1.5 Place the digestion vessel in the circular holder and connect the venting tubes. Place the circular holder in the microwave. Following manufacturer's instructions digest the samples at 100% power for 5 minutes.
9.1.6 To insure that the same flux of microwave energy each sample, all 12 turntable positions must be used. Fill the unused positions with vessels containing blank water.
9.1.7 Cool the digestion vessels for 30 minutes before opening. Open and remove the centrifuge tube. Add 10 mL of water to the centrifuge tube, tighten the cap and shake well.
9.1.8 Centrifuge the digested filter for 10 minutes. Pour off the supernatant liquid into a 50 mL volumetric flask.
9.1.9 Add 20 mL of water, cap tightly and shake well. Centrifuge the water, rinse and pour off into the 50 mL volumetric flask.
9.1.10 Bring the water in the 50 mL volumetric flask up to the marked volume and pour into a labeled polyethylene bottle and cap.
9.1.11 For each batch of 20 or fewer filters, the following quality control samples are prepared: 1 reagent blank; 1 filter blank; 1 duplicate filter strip (unless triplicate strips are run: cf. section 9.1.1). These audits are defined in section 11.2.
9.2 Sample analysis
9.2.1 Analyze each solution for arsenic, using the procedures described in CRL Method 206.2 DNS, "Standard Operating Procedure for the Analysis of Arsenic in Water." The same instrument quality control used in that method is required for this SOP.
9.2.2 Further instructions on the operation of the Zeeman 5000 atomic absorption spectrophotometer are contained in the Zeeman 5000 Operating Manual and the Manufacturer's Manuals.
10. CALCULATIONS
10.1 The calculations to calculate the concentrations of arsenic in the extracts can be performed off-line with an IBM PC. All calculations shall be performed as in the method referenced above (CRL Method 206.2 DNS).
10.2 If there is a significant blank filter correction, it must be made with each extract.
10.3 The concentration of arsenic in the ambient air is calculated as follows:
C = (S x 0.05 L/strip x 9 strips/filter) - Fb
v
where
C = concentration of As in air, ug/m3
S = concentration of As in the sample extract, ug/L
Fb= amount of As in the blank filter, ug, as determined in section 6.13
V = volume of air sampled by Hi -Vol , m3
10.4 The detection limit for graphite furnace atomic absorption analysis of As is 2 ug As/L. If the blank filter analysis, Fb, corresponds to an extract concentration of 2 ug As/L or less, the detection
limit in ug As/m3 is calculated for that filter by inserting that value for S in the above equation. If Fb is greater than this value, the reporting detection limit will correspond to twice the standard deviation
determined in the lot analysis, plus the instrument detection limit, inserted in the above equation.
10.5 Results at or near the detection limit are reported with one significant figure. Higher results are reported with two significant figures.
11. QUALITY CONTROL
11.1 The instrument quality control consists of a blank and a quality assurance check sample. The
instrument quality control is run as specified in the above referenced method (CRL Method 206.2
DNS).
11.2 Extraction quality control includes reagent blanks, filter blanks, and duplicate filter strips. These
audits should be included at a frequency of one set for each 20 samples or fraction thereof. Corrective action for failure of any of these audits is repreparation of the extracts and reanalysis. If
triplicate extractions have been performed be certain that sufficient filter remains for all other analysis before attempting repreparation.
11.2.1 A reagent blank is the extraction conducted without a filter strip. Any contamination from the extraction process should be seen in this audit. The acceptable
variation is the detection limit.
11.2.2 A filter blank is a filter strip from an unexposed filter of the same lot as the exposed filters. This is in addition to the blanks which were checked in the
original determination before the lot was weighed and exposed. This audit shall not exceed the average plus two standard deviation of the previous lot analysis.
11.2.3 A duplicate filter strip is extracted for each group of 20 or less samples. If variation in the coating of the filter is suspected and triplicate samples are taken no duplicates need to be run. The limit for the duplicate pair is the detection limit as a difference, or 10% relative percent difference. Relative percent difference is calculated as follows: The absolute difference of the two values divided by the average of the two values times 100%.
11.3 Analysis of a sample from the National Bureau of Standards or a reference standard from RTP should be run when the method is verified, and each month when samples are scheduled. The reference values should be obtained within the stated windows. Because there is little experience with this method, the limits cannot be set at this time.
12. INSTRUMENT MAINTENANCE
12.1 See the Instrument Operating Manual.
REFERENCES
1. Standard Operating Procedure for the Analysis of Arsenic in Water, Method 206.2 DNS (AA, Furnace, Standard Addition) USEPA, CRL, 1987
2. Instrument Operating Manual, Zeeman 5000, Furnace Mode, USEPA, CRL, 1987
3. Operation and Service Manual, Microwave Digestion System Model MSD-81D, CEM Corp., Indian Trail, NC, 1985
PARAMETER:
ARSENIC MATRIX:
AIR FILTER METHOD:
METHOD (ATOMIC ABSORPTION.FURNACE) WORKING RANGE
WITHOUT DILUTION:
0.5 ng As/m3 - 7 ng As/m3 assuming 2000 m3 air METHOD DETECTION LIMIT
not yet determined: estimated 0.5 ng/m3 for 2000 m3- air through filter. SAMPLE HANDLING:
Container air filter
Preservation none
Holding Time none REGULATORY LIMIT:
LABORATORY CONTROLS:
In addition to Field Controls. the tollowing will be analyzed within each analytical run:
1. An instrument blank which will be run at the beginning and end of the run to check for stability of the baseline.
Control limit is O+/ -2 micrograms/liter.
2. An undigested control standard at approximately 23.5 ug/'L will be analyzed at the beginning and end of the run to evaluate instrument performance.
Control limits are +/- 23.5 ug/L from the true value.
3. An extraction reagent blank is run once per 2c) filters to check the extraction for contamination.
Control limit is 0 +/- 2 ug/L
4. Filter blanks are analyzed as a lot analysis ot the filters before their release to the sampling entity. This result is subtracted from the subsequent analysis ot the exposed
filters. In addition. an unexposed filter strip from the same lot is passed through the extraction tor each 20 tilters analvzpd.
Control limit is the average plus two standard deviations or
the previous lot analysis.
5. A duplicate filter strip is extracted for each group of 20 filters. It variation in the distribution of analyte on the filter is suspected. triplicate filter strips are analyzed
and averaged. and the duplicate is not necessary.
Control limit is the absolute difference is less than or equal to the detection limit. or +/- 10% relative percent difference.
6. Each month when filter samples are scheduled. a Standard Reference Material filter strip from the National Bureau of Standards. or a reterence tilter strip from RTP will be
analyzed.
Control limits are within the stated windows tor the reterence material.
REFERENCES: