Basic HTML Version
Appendix B
5/20/2009
11
instrument performance. If the determined concentrations are not
within ± 10% of the certified values, performance of the determinative
step of the method is unacceptable. The source of the problem must be
identified and corrected before proceeding with the initial determination
of MDLs.
9.2.3.
Method Detection Limits (MDLs)
– MDLs should be established for
PC and PN using a low level estuarine water sample, typically three to
five times higher than the estimated MDL. The same procedure should
be followed for sediments or other weighed samples. To determine the
MDL values, analyze seven replicate aliquots of water or sediment and
process through the entire analytical procedure. Perform all calculations
defined in the procedure (Section 12) and report the concentration
values in the appropriate units. Calculate the MDL as follows:
MDL = 3 x S
Where,
S = Standard deviation of the replicate analyses.
9.2.4.
MDLs should be determined annually, whenever there is a significant
change in instrumental response, change of operator, or a new matrix is
encountered.
9.3.
Assessing Laboratory Performance
9.3.1.
Laboratory Reagent Blank (LRB)
– The laboratory must analyze at
least one LRB (Section 3.40) with each batch of samples. For sediment
samples the LRB consists of the ladle, sample sleeve and sample
capsule, as there are no reagents involved in this procedure. For
aqueous samples the LRB is a precombusted filter of the same type and
size used for samples. LRB data are used to assess contamination from
the laboratory environment. For sediment samples, the blank value for
carbon should not exceed 150 uv and the blank value for nitrogen
should not exceed 50 uv. For aqueous samples, the blank value for
carbon should not exceed 375 uv and the blank value for nitrogen
should not exceed 50 uv.
9.3.1.1.
If the nitrogen blank during a BLANK analysis is in
excess of 2000% the nitrogen blank in memory the
“COPPER APPEARS SPENT” is printed. If the nitrogen
blank increased over 100 uv over BN in memory and the first
STANDARD KC/KN is more than any following
STANDARD KC/KN by 0.2 uv/ug, then a “COPPER
APPEARS SPENT” warning will be printed either during a
BLANK analysis or a STANDARD analysis.
9.3.2.
Quality Control Sample (QCS)/ Standard Reference Material
(SRM)
- When using this procedure, a quality control sample is
required to be analyzed at the beginning and end of the run, to verify
data quality and acceptable instrument performance. If the determined
concentrations are not within ± 3σ of the certified values, performance
of the determinative step of the method is unacceptable. The source of
the problem must be identified and corrected before either proceeding
with the initial determination of MDLs or continuing with the analyses.