2. 120-mL (Nalgene or equivalent) plastic
9.2 INITIAL DEMONSTRATION OF PERFOR-
bottle with cap (one for working standard).
MANCE: An initial calibration curve for the meth-
3. Volumetric pipets: 10, 8, 5, 2 mL.
od should be developed as directed in CALIBRA-
4. Volumetric flasks: 100-mL (one), 1 L (one).
TION AND STANDARDIZATION. The calibra-
5. 2-L graduated cylinder.
tion curve should be linear, with a zero inter-
cept and a slope (response factor) of 280 75.
6. 10-mL graduated cylinder (two).
The linear range of the method is 0.0200 g/L.
7. Disposable eyedropper (for sulfuric acid).
8. Glass fiber filters, 47 mm (for sediment-
9.3 ASSESSING PERFORMANCE: A method
laden samples).
detection limit (MDL) analysis should be per-
9. Glass cuvettes, 1-cm path length (two).
formed. A 20-L sample of deionized water is spiked
10. Vacuum filtration apparatus, 1-L vacuum
with 20 mL of the 10-mg/L working standard to
produce a picric acid concentration of 10 g/L.
flask with 47-mm filter holder and funnel.
11. 24- 200-mm glass test tubes with screw-
The water is stirred as completely as possible.
caps (one per sample).
Seven 2-L subsamples are processed according to
12. Empore Anion Extraction Disks, 47-mm
the method. The concentration of picric acid in the
(Varian).
subsamples is determined using the initial cali-
13. Pasteur pipets, 9 inch (one per sample).
bration curve. The standard deviation (SD) of the
14. 500-mL squirt bottles (two).
seven determinations is calculated. The MDL =
3.14 SD and should be about 3.6 g/L.
The recovery of the spiked picric acid should
7.0 REAGENT AND STANDARDS
be between 75 and 125%.
9.4 ASSESSING ANALYTE RECOVERY AND
1.
Picric acid (SARM or reagent grade).
DATA QUALITY: Blank and spiked water samples
2.
Methanol, reagent grade.
should be analyzed daily as directed in CALIBRA-
3.
Concentrated sulfuric acid, reagent grade.
TION AND STANDARDIZATION and records
4.
Water, deionized.
kept. The background-corrected absorbance of a
150-g/L water spike should be above 80% of the
predicted absorbance based on the initial calibra-
8.0 SAMPLE COLLECTION, PRESERVATION
tion curve. The blank water should produce a back-
AND STORAGE
ground-corrected absorbance less than 0.01 absor-
bance units. The relative standard deviation of
This method may be used with well water or
daily matrix spikes should be less than 27%.
surface water.
A water sample is collected in two 1-L amber
glass bottles which are capped and kept cold (4C)
10. CALIBRATION AND
in the dark until extraction. Samples should be
STANDARDIZATION
analyzed the same day they are collected.
10.1 PREPARATION OF STANDARDS: About
0.2 g of solid picric acid (SARM or reagent grade)
9.0 QUALITY CONTROL
is dried to constant weight in a vacuum desicca-
9.1 GENERAL: The accuracy and precision of
tor in the dark. About 0.125 g is weighed out to
this method are subject to the common errors
the nearest 0.1 mg, transferred to a 100-mL volu-
associated with poor-quality measurements of
metric flask and diluted to volume with deion-
weights and volumes. There is a source of error
ized water. A stir bar is added and the solution
specific to this method. The removal of residual
stirred overnight in the dark. The concentration
water from the spectrophotometer cuvettes
of picric acid in this stock is about 1250 mg/L. A
(PROCEDURE Step 17) is critical for successful
working standard is made by diluting 8 mL of
application of the background correction calcu-
the stock with deionized water to 1 L in a volu-
lation. The manufacturer's instructions for the
metric flask. The concentration of the working
extraction disks call for a conditioning rinse. This
standard is about 10 mg/L.
step is not necessary. The disks may be used
Calibration solutions are prepared in a 2-L
directly from the package.
graduated cylinder by diluting the working stan-
19
Previous Page
