Application
News
Analysis of unknown sample
The performance of the analytical system for each analytical batch
shall be verified when analyzing unknown samples. First, it shall
be verified that the recovery rates are within 80–120% and the
retention times are within the retention time window calculated
in the “verification of initial performance” step (Calibration
Verification) through quantitative determination of the
calibration standard solution using the calibration curve created
in the “Creation and evaluation of calibration curve”. The method
blank shall be prepared using the same procedure described
above and it shall be verified that the quantified concentration is
less than 4.0 µg/L (Method Blank). Furthermore, it shall be verified
that the recovery rate for PFHxS standard solution of known
concentration is within the range of 70–130% (Ongoing Precision
and Recovery, OPR). Furthermore, when analyzing unknown
samples, a matrix-spiked sample (Matrix Spike, MS) and its
duplicate (Matrix Spike Duplicate, MSD) shall be analyzed to verify
that the retention times are within the retention time window
and within ±0.2 min of the retention time during the verification
of initial performance, that the spiked recovery rate is within 50–
150%, and that the Relative Percent Difference (RPD) is within
30%. The analytical procedure is shown below.
(1) Boat Blank Measurement (at least 2)
(2) Calibration Verification (Low-level)
(3) Method Blank Measurement
(4) OPR Measurement
(5) Samples (10 or fewer)
(6) Matrix Spike Measurement
(7) Calibration Verification (Mid-level)
(8) Samples (10 or fewer)
(9) Calibration Verification (Mid-level)
Result Criterion
RSD for AOF concentration(%) 2.07 ≦20
Average Recovery (%) 94.8 80-120
AOF(µg/L)=C
MB
×
V
abs
V
MB
0.0 5.0 10.0 15.0 20.0 25.0 min
-5.0
0.0
5.0
10.0
2.0 3.0 4.0 min
-0.1
0.0
0.1
F
µS/cm
µS/cm
F
Result Criterion
Method Blank (µg/L) 0.57 < 4.0
MDL
b
=ｘ＋ t
(n-1,1-α=0.99)
S
b
MDL
S
=t
(n-1,1-α=0.99)
S
S
MDL
b
MDL
S
MDL ML
0.74 1.16 1.16 5
ML=MDLｘ3.18
On the C-IC Solution analysis screen (Fig. 5), the fluoride ion
concentration (IC concentration) measured by ion
chromatograph is automatically converted to the concentration
in the sample. Both the IC concentration and the concentration in
the sample are simultaneously confirmed on the same screen.
The quantitative results (not shown in the table) were calculated
by converting them to AOF concentrations (µg/L) in the solution
before the activated carbon treatment in accordance with the
procedure described in EPA 1621. The conversion formula is
shown below, and Table 3 presents the repeatability and recovery
rate of AOF concentration (µg/L) from four consecutive analyses of
the PFHxS standard solution.
MDL and ML
To verify the presence of fluoride ions originating from reagents
and equipment used in the pretreatment process, 100 mL of
ultrapure water was passed through activated carbon and
subjected to the same pretreatment as the sample (method
blank). Fig. 6 shows the chromatogram of the method blank, and
Table 4 presents its quantitative result. The EPA 1621 standard
specifies that the quantitative result for the method blank must
not exceed 4.0 µg/L, and it was confirmed that the sample met
this criterion.
Furthermore, it is stated that the ML calculated using the
following equation shall be reported after rounding off to make a
form as 1, 2, or 5 × 10n (where n is 0 or integer). In this article, the
ML is reported as 5 since the calculation resulted in 3.7. The
calculated
results for the MDL and the ML are shown in Table 5.
AOF = Concentration of adsorbable organic fluorine in μg F-/L
C
MB
= Method blank measured concentration (sum of top and bottom GAC)
reported without volume corrections
V
abs
= Total volume of absorption solution in mL of the sample prior to
injection to IC
V
MB
= Volume in mL of method blank adsorbed
Fig. 5 Data analysis screen of C-IC Solution
Table 3 IPR calculation result
Table 4 Determination result of method blank
Fig. 6 Chromatogram of method blank
In EPA 1621, MDL was calculated in accordance with the
description of method procedure and 40 CFR Appendix B to Part
136
2)
. In this method, MDL is defined as the greater of MDL
b
,
which is calculated from the result of consecutive analyses of the
method blank and MDL
S
, which is calculated from the result of
consecutive analyses of 25 µg/L of PFHxS standard solution. The
formula used to calculate MDLb and MDLS is shown below.
Table 5 MDL and ML (µg/L)
MDL
b
= the MDL based on method blanks
X = mean of the method blank results
t
(n−1, 1α= 0.99)
= the Student's t-value appropriate for the single-tailed 99th
percentile t statistic and a standard deviation estimate with n−1 degrees of
freedom.
S
b
= sample standard deviation of the replicate method blank sample
analyses.
MDL
s
= the method detection limit based on spiked samples
t
(n-1, 1−α= 0.99)
= the Student's t-value appropriate for a single-tailed 99th
percentile t statistic and a standard deviation estimate with n-1 degrees of
freedom.
S
s
= sample standard deviation of the replicate spiked sample analyses.
C-IC Solution contains QC evaluation function that determines
whether the analytical results meet the criteria specified in EPA
1621 or not, enabling support for analyses that are in accordance
with testing requirements. Both MS and MSD met the retention
time criterion, verifying that the retention time window set during
the verification of initial performance was appropriate (Fig. 7).
Fig. 7 QC examination result screen of C-IC Solution
