Site index
Created by:
9.1.1
Minimum Field Quality Control Requirements
Field Quality Control activities shall meet or exceed the following
requirements.
9.1.1.1 Quality
Control Checks
a.
Field Quality Control Blanks
1.
General Considerations:
a.
All equipment blanks shall be collected and analyzed for the same
parameters as the associated samples.
b.
All blanks shall be preserved, transported, documented and handled as if
they were samples. Once collected,
they must remain with the sample set until they have been received by the
laboratory.
c.
All equipment blanks are prepared by rinsing the sampling equipment with
analyte-free water and collecting the rinsate in appropriate sample containers
(see a and b above).
2.
The following types of blanks shall be collected as specified:
a.
Precleaned Equipment Blank: These
blanks shall be collected from sampling equipment that has been brought to the
site precleaned and ready for use. At
least one equipment blank shall be collected for each water and solid matrix
analytical group. These blanks
shall be collected AT THE BEGINNING of the sampling episode.
b.
Field Cleaned Equipment Blank: These blanks shall be collected from
sampling equipment AFTER the equipment has been cleaned in the field (i.e.,
between sampling points).
c.
Trip Blank: These blanks are required only if samples are to be analyzed
for VOCs. They shall be prepared by the organization that is providing
the VOC vials, and shall be prepared by filling vials with analyte-free water.
The vials shall be placed in the same transport containers as the empty
VOC vials. They must remain with
the VOC vials during the sampling episode and shall be transported to the
analyzing laboratory in the same shipping or transport container(s) as the VOC
samples. The trip blanks
shall remain UNOPENED for the entire sampling episode.
A trip blank must be submitted for each cooler that transports empty or
full VOC vials.
b.
Field Duplicates
1.
Field duplicates shall be collected and analyzed for the same parameters
as the associated samples.
2.
They shall be preserved, transported and documented in the same manner as
the samples. THESE SAMPLES ARE NOT
CONSIDERED LABORATORY DUPLICATES.
3.
Duplicates are collected to measure the variability inherent in the
sampling process. They shall be
obtained by DUPLICATING (simultaneously or in rapid succession) the entire
sample acquisition technique that was used to obtain the first sample.
a.
Duplicates for water are collected by sampling from successively
collected volumes (i.e., samples from the next bailer of sample water).
b.
Duplicates for soils are collected from the same sample source (i.e.,
soil is obtained from the same soil sampling device).
c.
Frequency
1.
The frequency with which the above quality control samples are collected
is summarized below:
|
Number
of |
Precleaned |
Field
cleaned |
Trip
BLK |
Duplicate |
|
Samples |
Equipment
BLK |
Equipment
BLK |
(VOCs) |
|
|
10+ |
minimum
of one then 5% |
minimum
of one then 5% |
one
per cooler |
minimum
one then 10% |
|
5-9 |
one* |
one* |
NR |
one |
|
<
5 |
one* |
one* |
NR |
NR |
NR=Not Required
BLK=Blank
*
Note: For 9 or
fewer samples, a precleaned equipment blank or a field cleaned equipment blank
is required. A field cleaned
equipment blank must be collected if equipment is cleaned in the field.
2.
Clarification and Explanation
a.
The number and type of equipment blanks (EQB) are dependent upon two
factors above and beyond the number of samples taken at a given site.
1.
The first factor is the number of precleaned pieces of sampling equipment
(bailer, Nansen bottle, Kemmerer, etc.) brought into the field for use.
2.
The second factor is the total number of field cleanings performed on
this equipment.
b.
To calculate the number of each kind of EQB (precleaned and field
cleaned) the following sequence should be used:
1.
Determine the total number of samples taken for each matrix.
2.
Determine the total number of pieces of equipment by type to be used for
a given matrix which will be brought into the field precleaned.
3.
Determine the total number of field decontaminations that will be
necessary for the sampling event by subtracting #2 from #1 above.
4.
Once these have been determined, the requirements on the following table
and the generalities listed below may be used to determine the total number of
EQBs that must be collected and analyzed:
a.
For 1 to 9 samples, at least one EQB must be taken, either Precleaned OR
Field cleaned. If field cleaning is performed, then the EQB must be a Field
cleaned EQB.
b.
For 10 to 20 samples, one Precleaned EQB AND one Field cleaned EQB must
be taken unless all equipment used is precleaned.
If no field cleaning is performed, then only one Precleaned EQB is
required.
c.
For greater than 20 samples, the 5% (1 in 20) requirement must be met for
BOTH the Precleaned (number 2 above) and Field cleaned (number 3 above)
equipment.
Examples
for sample sets greater than 20:
|
Total
Samples Taken |
Total
Pre-Clean Equip. Used |
Required
Field Cleanings |
Precleaned
EQB # Required |
Field
Cleaned EQB# Required |
|
25 |
1 |
24 |
1 |
2 |
|
|
5 |
20 |
1 |
1 |
|
|
21 |
4 |
2 |
1 |
|
|
25 |
0 |
2 |
0 |
|
|
|
|
|
|
|
30 |
1 |
29 |
1 |
2 |
|
|
10 |
20 |
1 |
1 |
|
|
21 |
9 |
2 |
1 |
|
|
30 |
0 |
2 |
0 |
|
|
|
|
|
|
|
35 |
1 |
34 |
1 |
2 |
|
|
15 |
20 |
1 |
1 |
|
|
21 |
14 |
2 |
1 |
|
|
35 |
0 |
2 |
0 |
Note:
To reduce the number of submitted EQBs, bring sufficient precleaned
equipment into the field such that 20 or fewer field cleanings are performed.
9.1.1.2 Split
Samples
a.
DER or the client may require split samples as a means of determining
compliance or as an added measure of quality control.
These type of samples are intended to measure the variability between
laboratories and should be obtained as subsamples from the same parent sample.
b.
A true split sample of soil, sediment or sludge is almost impossible to
accomplish under field conditions.
c.
Split samples shall be collected, preserved, transported and documented
using the same protocols as the related samples.
In addition, an attempt should be made to use the same preservatives (if
required).
d.
Split samples for water shall be collected in one of two ways:
1.
Mix the sample in a large, appropriately precleaned, intermediate vessel
and pour aliquots of the mixed sample into the appropriate sample containers.
This method shall not be used if VOCs are of interest.
2.
Fill the sample containers from consecutive sample volumes FROM THE SAME
SAMPLING DEVICE (i.e., from the same bailer).
If the sampling device does not hold enough sample to fill the sample
containers, the following protocol shall be used:
a.
Fill the first container with half of the sample, and pour the remaining
sample into the second container.
b.
Obtain additional sample, and pour the first half into the SECOND
container. The remaining portion
shall be poured into to first container.
c.
Continue with steps 1 and 2 until both containers are filled.
9.1.1.3 Quality
Control on Field Measurements
a.
All field instruments must be initially calibrated at the beginning of
each working day.
b.
A continuing calibration check shall be analyzed at intervals of no more
than 4 hours and at the end of the sampling day to determine if the instrument
has maintained calibration.
c.
The instrument shall be recalibrated if the continuing calibration checks
fail to meet acceptance criteria.
d.
All quality control data shall be recorded in the daily field notes.
The laboratory shall follow the minimum
quality control requirements specified by each method.
If no quality control requirements are listed in the method, or if the
method quality control requirements are less stringent than those listed below,
the laboratory shall follow the guidelines listed below:
9.1.2.1
Chemistry QC Checks
a.
Method reagent blanks - shall be prepared and analyzed at a rate of one
per sample set (see definitions in Appendix A).
b.
Matrix Spikes - At least one sample in a sample set (or 5%, whichever is
greater) with similar matrices shall be prepared and analyzed by the specified
method. If a set contains samples
of different matrices, matrix spikes should be prepared and analyzed for each
matrix type. For work submitted to
DER, matrix spikes must be included as routine protocol.
c.
Reagent water or reagent matrix spikes - Reagent water or reagent matrix
spikes may be used as additional QC checks to monitor the effectiveness of the
method. If used, these must be analyzed at a frequency of 5%.
d.
Quality control check samples - shall be analyzed in duplicate
semiannually. Such samples shall be
analyzed as blind samples (i.e.,
the component concentrations in these samples shall not be provided to the
analyst until after analysis). If
the data are not acceptable, the analytical results must be reported in a QA
report to DER (see Section 13).
e.
Quality control check standards - shall be analyzed at a continuing
frequency equivalent to 5% of the samples in the analytical set (i.e. one every
20 samples) or shall be analyzed at the beginning of each run to verify the
standard curve.
f.
Duplicate samples or matrix spike duplicates - at least one or 5% of all
samples in a sample set with a similar matrix shall be selected and analyzed in
duplicate. If a sample set contains samples from different matrices
(e.g., effluent and drinking water), then duplicates or matrix spike duplicates
should be analyzed for each matrix.
g.
Continuing calibration standards shall be analyzed at a frequency
equivalent to 5% of the samples in an analytical set.
Alternatively, quality control check standards may be used (see e.
above). At least one of these checks shall be a standard at a
concentration of 1 - 2 times the laboratory stated PQL.
h.
Additional quality control checks may be included and shall be used if
specified by the approved method:
a.
Reagent purity checks
b.
Internal standards
c.
Surrogate spikes
9.1.2.2
Microbiology QC Checks
a.
Blanks
1.
Membrane Filter Analysis: For
each set of samples, a control blank shall be run at the beginning (dilution
water blank), every tenth sample (sample carry over blank), and at the end of
the set.
2.
MPN Analysis: A single tube of LTB broth media shall be inoculated with 10
milliters of sterile phosphate buffered dilution water (dilution blank control).
b.
Duplicates - At least 10% of the known positive samples that have been
processed shall be analyzed in duplicate or a minimum of one duplicate analysis
per month for MF and MPN analyses.
c.
Positive/Negative Controls - Microorganisms obtained from the American
Type Culture Collection (ATCC) or equivalent sources shall be used to confirm
the morphological and biochemical responses to test media.
Positive and negative controls shall be run with each new lot of media,
and monthly thereafter.
d.
Water Quality Indicators:
1.
Water source shall be tested monthly for chlorine residual, conductivity
and standard plate count.
2.
The concentration of metals in the water source shall be determined
annually.
3.
Water Suitability Test shall be conducted annually.
e.
5% of all positive environmental samples analyzed by membrane filter
shall be verified per method requirements.
f.
When using the MPN test, the MPN test shall be completed on 10% of all
positive confirmed samples.
9.1.2.3
Laboratory QC Checks (Bioassays)
a.
At least one set of controls (dilution water and hardness or salinity, if
appropriate) shall be run with each test.
b.
Analytical equipment shall follow the chemistry laboratory quality
control checks listed above.
9.1.2.4
Laboratory QC Checks (Species Identification)
a.
Should maintain or have access to a type specimen collection.
b.
Must, at a specified frequency use outside experts to corroborate species
identification.
9.2
ROUTINE METHODS USED TO ASSESS PRECISION AND ACCURACY
9.2.1 Precision and accuracy targets listed
in the tables of Methods, Matrices and QA Targets (Section 5.0 of a CompQAP)
must be generated from matrix spikes and matrix spike duplicates or duplicates
of environmental samples. The
laboratory must maintain a list of QC checks, as presented in Section 9.1.2,
which identifies applicable analytical methods and the concentrations to be used
to make the determination in terms of low, mid or high levels:
1. Low
level is defined as concentrations from the minimum detection limit to a level 5
times the MDL.
2. Mid
level is defined as the mean level between the minimum detection level and the
upper end of the linear range.
3. High
level is defined as concentrations at the upper and of the linear range.
9.2.3. The laboratories shall use the following formulas
for calculating the precision and accuracy of test measurements and the
associated acceptance ranges:
9.2.3.1 The
precision of replicate pairs shall be calculated using one of the following
formula:
a.
The precision of duplicate pairs shall be calculated using
1.
Percent Relative Standard Deviation (% RSD)
% RSD
=
s
X 100
0
here:
0
= Mean (average) of the data points
s = Standard deviation calculated as:
n
3
x-0
i=1
n-1
In the case of pairs (duplicates) this formula becomes:
% RSD = A-B
X 2 X 100
A+B X 2
2.
Relative Percent Difference (RPD)
RPD =
A-B X 200
A+B
Where:
A = concentration in sample A
B = concentration in sample B
3.
Industrial Statistic (I)
I =
A-B
A+B
Where:
A = concentration in sample A
B = concentration in sample B
9.2.3.3
The accuracy of a measurement shall be determined by the recovery of a
known amount of analyte in a real sample as:
% R =
Cs - Cu (100)
s
Where:
Cs = concentration of spiked sample
Cu = concentration in unspiked sample
S = expected concentration
of spike in sample
%R = percent recovery
9.2.3.4
Upper and Lower Warning and Control Limits to be used as acceptance
criteria shall be calculated as follows:
n
3
x-0
i=1
n-1
Where:
s = Standard Deviation
n = Number of points or data pairs to be included in the calculation
x = Sample Percent Recovery or precision of replicates
0
= Mean (average) of the data points
CL = Pav "
3 S
Where:
CL = Control limit (upper and/or lower)
Pav = Mean of P (average percent recovery or average % RSD)
WL = Pav "
2 S
Where:
WL = Warning limit (upper and/or lower)
9.2.3.5
Microbiological quality control acceptance criteria shall be calculated
per formulae specified in Standard Methods for the Examination of Water and
Wastewater, 17th Edition, Method Number 9020, Section 4.b (pp. 9-17 and
9-18).
9.2.4 Quality Control charts must be prepared or
other easily followed system instituted to track results of QC checks.
9.2.5 Quality control charts or tabulation systems
must be updated every 20 data points or annually at a minimum.
9.3
METHOD DETECTION LIMITS AND PRACTICAL QUANTITATION LIMITS
9.3.1
Method Detection Limits (MDLs)
MDLs shall be determined by one of
three protocols:
1. EPA -
"Definition and Procedure for the Determination of the Method Detection
Limit - Revision 1.11", 40 CFR Part 136, Appendix B;
2. IUPAC
- "Nomenclature, symbols, units and their usage in Spectrochemical analysis
II," Spectrochim Acta B 1978, 33B, 242 and "Limit of Detection, A
Closer Look at the IUPAC Definition," Analytical Chemistry 1983, V. 55, No.
7 page 712A;
3.
USATHAMA - "Decision and detection limits for linear calibration
curves," Hubaux, A. and Vox, G., Analytical Chemistry 42, 849 - 855 (1970).
IN ALL CASES, THE METHOD DETECTION LIMIT IS DEFINED TO BE THREE TIMES THE
STANDARD DEVIATION DERIVED FROM THE STUDY.
9.3.2
Practical Quantitation Limit (PQLs)
The PQL is defined as 12 times the
standard deviation that is derived from the procedures used to determine MDL.
Note:
EPA is currently considering changing the names BUT NOT THE DEFINITIONS
of the above-mentioned terms:
MDL
is proposed to be Method Detection Level (currently method detection limit); and
RDL
is proposed to be Reliable Quantitation Level (currently practical quantitation
limit).
MDLs
or PQLs shall be verified on an continuous basis (see 9.1.2.1.g).
See
the Custody Section for Field QC checks.
1. Records
which document sample/standard preparation, source and concentration (this
includes protocols for preparation and certification, if applicable) must be
maintained. All required records specified in the Calibration SOP must be
maintained.
2.
Identification of analyses set that the applicable QC sample is
associated with.
3.
Calculations performed to determine QC results.
4. Control
limits used to evaluate analysis results and how these were determined.
1.
Documentation for the MDL studies must be conducted according to all
other SOPs regarding sample and standard handling procedures, calibrations, QC
checks and analyses.
2.
Documentation must include:
a.
Date of studies and calendar period for which the studies are applicable
b.
Analytical Method
c.
Identification of analyst responsible for analysis
d.
Compounds studied