ASTM D5847-22

This document is not an ASTM standard and is intended only to provide the user of an ASTM standard an indication of what changes have been made to the previous version. Because
it may not be technically possible to adequately depict all changes accurately, ASTM recommends that users consult prior editions as appropriate. In all cases only the current version
of the standard as published by ASTM is to be considered the official document.
Designation: D5847 − 02 (Reapproved 2020) D5847 − 22
Standard Practice for
Writing Quality Control Specifications for Standard Test
Methods for Water Analysis
This standard is issued under the fixed designation D5847; the number immediately following the designation indicates the year of
original adoption or, in the case of revision, the year of last revision. A number in parentheses indicates the year of last reapproval. A
superscript epsilon (´) indicates an editorial change since the last revision or reapproval.
1. Scope Scope*
1.1 This practice provides specific, mandatory requirements for incorporating quality control (QC) procedures into all test methods
under the jurisdiction of Committee D19.
1.2 ASTM International has adopted the following:
This practice is under the jurisdiction of ASTM Committee D19 on Water and is the direct responsibility of Subcommittee D19.02 on Quality Systems, Specification,
and Statistics.
Current edition approved May 1, 2020May 1, 2022. Published May 2020July 2022. Originally approved in 1999. Last previous edition approved in 20122020 as
D5847 – 02 (2012).(2020). DOI: 10.1520/D5847-02R20.10.1520/D5847-22.
*A Summary of Changes section appears at the end of this standard
Copyright © ASTM International, 100 Barr Harbor Drive, PO Box C700, West Conshohocken, PA 19428-2959. United States
D5847 − 22
Policy on implementation of requirements for a quality control
section in standard test methods generated by Committee D19
on Water.
GENERAL—By July 29, 1998, or at the next reapproval or revision,
whichever is later, every D19 Standard Test Method shall contain a
QC section that is in full compliance with the requirements of this
practice.
NEW COLLABORATIVE TESTING—As of July 29, 1998, each
collaborative study design shall include a QC section as part of the
method to be tested. Prior to approval of the study design, the
Results Advisor shall ascertain the appropriateness of the QC
section in meeting the requirements of this practice and Practice
D2777, and shall advise the designer of the study of any changes
needed to fulfill the requirements of these practices. Before a collab-
orative study may be conducted, approval of the study design by the
Results Advisor must be obtained.
NEW COLLABORATIVE TESTING—As of July 29, 1998, each col-
laborative study design shall include a QC section as part of the
method to be tested. Prior to approval of the study design, the Re-
sults Advisor or equivalent shall ascertain the appropriateness of the
QC section in meeting the requirements of this practice and Practice
D2777, and shall advise the designer of the study of any changes
needed to fulfill the requirements of these practices. Before a collab-
orative study may be conducted, approval of the study design by the
Results Advisor or equivalent shall be obtained.
OLDER VALIDATED METHODS—Standard test methods that were
validated using Practices D2777 – 77, D2777 – 86, or D2777 – 94,
when balloted for reapproval or revision, shall contain a QC section
based upon the best information from the historical record. Where
appropriate, information derived from the record of the collaborative
study shall be utilized for this purpose. The introduction of the QC
section into these standard test methods shall not be construed as a
requirement for a new collaborative study, though the Subcommittee
may opt for such a study. Any information available regarding QC or
precision/bias testing shall be included in the appropriate sections of
the published test method.
1.3 Required QC sections in all applicable test methods are intended to achieve two goals. First, users of Committee D19 test
methods will be able to demonstrate a minimum competency in the performance of these test methods by comparison with
collaborative study data. Second, all users of test methods will be required to perform a minimum level of QC as part of proper
implementation of these test methods to ensure ongoing competency.
1.4 This practice contains the primary requirements for QC of a specific test method. In many cases, it may be desirable to
implement additional QC requirements to assure the desired quality of data.
1.5 The specific requirements in this practice may not be applicable to all test methods. These requirements may vary depending
on the type of test method used as well as the analyte being determined and the sample matrix being analyzed. See Explanation
1 in Appendix X1.
1.5.1 If there are compelling reasons why any of the specific QC requirements listed in this practice are not applicable to a specific
test method, these reasons mustshall be documented in the QC section of the test method.
1.5.2 With the approval of Committee D19 on the recommendation of the D19 Results Advisor or equivalent and the Technical
Operations section of the Executive Subcommittee, a statement giving the compelling reasons why compliance with all or specific
points of this practice cannot be achieved will meet the requirements of both ASTM and this practice.
1.5.3 Test methods developed prior to the approval of this practice with a QC section that meet the requirements of Practice D5789
are considered in compliance with this practice.
1.6 This practice is for use with quantitative test methods and may not be applicable to qualitative test methods.
1.7 Presently, this practice is applicable primarily to chemical test methods. It is intended that, in future revisions, the practice will
be expanded to include other test methods such as microbiological test methods.
D5847 − 22
1.8 Units—The values stated in SI units are to be regarded as standard. No other units of measurement are included in this
standard.
1.9 This international standard was developed in accordance with internationally recognized principles on standardization
established in the Decision on Principles for the Development of International Standards, Guides and Recommendations issued
by the World Trade Organization Technical Barriers to Trade (TBT) Committee.
2. Referenced Documents
2.1 ASTM Standards:
D1129 Terminology Relating to Water
D1141 Practice for Preparation of Substitute Ocean Water
D1193 Specification for Reagent Water
D2777 Practice for Determination of Precision and Bias of Applicable Test Methods of Committee D19 on Water
D3648 Practices for the Measurement of Radioactivity
D3856 Guide for Management Systems in Laboratories Engaged in Analysis of Water
D4375 Practice for Basic Statistics in Committee D19 on Water (Withdrawn 2018)
D5789 Practice for Writing Quality Control Specifications for Standard Test Methods for Organic Constituents (Withdrawn
2002)
D5810 Guide for Spiking into Aqueous Samples
D6362 Practice for Certificates of Reference Materials for Water Analysis
3. Terminology
3.1 Definitions:
3.1.1 For definitions of terms used in this standard,practice, refer to Terminology D1129 and Practice D4375.
3.2 Definitions of Terms Specific to This Standard:
3.2.1 batch, n—a set (group) of samples analyzed such that results of analysis of the QC samples (laboratory control sample,
method blank, matrix spike, and duplicate or matrix spike duplicate) analyzed with the batch are indicative of the quality of the
results of analysis of samples in the batch. The number of samples in the batch is defined by the task group responsible for the
test method. See 6.4 and Explanation 2 in Appendix X1.
3.2.1.1 Discussion—
When The number of samples in the batch is defined by the task group responsible for the test method. See 6.4 and Explanation
2 in Appendix X1.When results from tests of any of the QC samples associated with batch the batch fail to meet the performance
criteria, the test method should define the appropriate corrective action. To make such a response valid, the batch mustshall be
constructed in such a way as to assure that all variables affecting the batch will affect all samples in the batch in a statistically
equivalent manner.
3.2.2 calibration standard, n—a solution containing the analyte of interest at a known concentration either purchased from an
external source or prepared in-house from materials of known purity or concentration, or both, and used to calibrate the
measurement system.
3.2.3 detection limit, n—the minimum concentration or amount of a substance that can be detected with a known degree of
confidence.
3.2.3 independentcertified reference material (IRM),(CRM), n—a material of known purity and concentration reference material,
accompanied by a certificate, one or more of whose property values are certified by a procedure that established its traceability
to an accurate realization of the unit in which the property values are expressed and for which each certified value is accompanied
by an uncertainty at a stated level of confidence obtained either from the National Institute of Standards and Technology (NIST)
or other reputable supplier. The IRM shall be obtained from a different lot of material than is used for calibration.national institute
or ISO 17034 accredited supplier.
For referenced ASTM standards, visit the ASTM website, www.astm.org, or contact ASTM Customer Service at service@astm.org. For Annual Book of ASTM Standards
volume information, refer to the standard’s Document Summary page on the ASTM website.
D5847 − 22
3.2.3.1 Discussion—
The CRM shall be obtained from a different lot of material than is used for calibration, if possible. There is significant variation
in the overall quality of commercially available Certified Reference Materials and caution should be used when choosing Certified
Reference Materials. Use Practice D6362 to provide guidance as to what information needs to be included on certificate of a
certified reference material.
3.2.4 detection limit, n—minimum concentration or amount of a substance that can be discriminated from a blank with a known
degree of confidence.
3.2.5 laboratory control sample (LCS),sample, LCS, n—a sample of known concentration and composition that is taken through
the entire test method to determine whether the analytical system is in control. The LCS must be prepared in the appropriate
ASTM-grade water from a material that sufficiently challenges the test. See Explanation 3 in Appendix X1. The LCS can be an
IRM obtained from an outside source or prepared in-house from materials of known purity and concentration. Alternatively, the
LCS may be a real sample of the matrix that is typically analyzed and which has been fully characterized.
3.2.5.1 Discussion—
The LCS may also be commonly known as a “quality control sample” or an “ongoing precision and recovery sample” (OPR).
3.2.6 matrix spike (MS),spike, MS, n—addition of a known concentration of analyte to a routine sample representing a specific
matrix for the purpose of evaluating interference from matrix components. (See Guide D5810.)
3.2.7 method blank (blank), n—reagent water (see Specification D1193) either known to be free of the constituent(s) of interest
or containing only a low, known concentration of the constituent(s) of interest not exceeding five times the estimated detection at
the quantitation limit.
3.2.7.1 Discussion—
The purpose of analysis of the method blank is to confirm that the reagents or analytical system, or both, do not contribute a
measurable amount of the constituent(s) of interest during analysis of routine samples or, if they do, to determine what the
contribution is.
3.2.8 quantitation limit, n—the minimum concentration or amount of a substance that can be measured with a known degree of
confidence.
3.2.9 sample pretreatment (pretreatment), n—any handling, manipulation or treatment of a sample prior to subjecting the sample
to the analysis. Examples are filtration, digestion, dilution, pH adjustment and extraction.
4. Summary of Practice
4.1 This practice provides the writer of a test method in Committee D19 specific steps to be included in the QC section of the test
method. A QC section is required in all applicable standard test methods that mandates use of the following QC measures:
4.1.1 Periodic calibration or verification of calibration of the measurement system,
4.1.2 Initial demonstration of laboratoryanalyst capability,
4.1.3 Analysis of at least one blank per batch,
4.1.4 Analysis of at least one LCS per batch,
4.1.5 Analysis of at least one MS per batch, where applicable, and
4.1.6 Periodic analysis of an IRM.a CRM.
4.2 Duplicate analysis of at least one sample per batch is suggested. The duplicate analysis may be of a sample or of a matrix spike
(matrix spike duplicate; MSD). See Explanation 4 in Appendix X1.
4.3 If there are valid reasons why any of the above QC requirements are inapplicable to a specific test method (see Section 1),
these reasons mustshall be documented in the QC section of the test method. See 1.5 and Explanation 1 in Appendix X1.
D5847 − 22
5. Significance and Use
5.1 In order to be certain that the end user of analytical results obtained from using an ASTM Committee D19 test method can
be confident that the values have been obtained through a competent application of the test method, a demonstration of the
proficiency of the analytical system mustshall be performed. Appropriate proficiency is demonstrated by achievement of
performance criteria derived from results of the test method collaborative study. The QC measures specified in this practice
mustshall be included in each ASTM test method, as applicable, to ensure the quality of measurements.
5.2 In order for users of D19 test methods to achieve consistently valid results, a minimum level of QC mustshall be performed.
This minimum level of QC is stipulated in this practice and by the task groups developing D19 test methods. If the specific
requirements outlined in this practice are not applicable to the test method, alternative QC mustshall be defined in the test method.
6. Requirements for QC Specifications in Test Methods
6.1 Every test method mustshall have a quality control (QC) QC section. Listed below are requirements applicable to nearly all
chemical test methods and that mustshall be followed to ensure that the test method is in control and to validate the accuracy of
data generated for a specific matrix.
6.1.1 The measures that mustshall be specified in the QC section of test methods and the reasons for these measures are as follows:
6.1.1.1 Calibration and calibration verification are necessary to ensure that the analytical system is properly calibrated during the
period that the analysis is performed.
6.1.1.2 An initial demonstration of laboratory capability is necessary to prevent errors as a result of unfamiliarity with the test.
6.1.1.3 Analysis of a blank with each batch may indicate that analytes in a test sample are the result of contamination.
6.1.1.4 An LCS is run with each batch to determine that the measurement system is in control at the time samples are being
analyzed.
6.1.1.5 An MS (recovery check) provides information on the bias of the test method in a specific matrix.
6.1.1.6 A duplicate analysis (Dup) or duplicate of the MS (matrix spike duplicate; MSD) indicates the repeatability of the method
for a specific matrix.
6.1.1.7 An IRMA CRM is analyzed periodically to validate the accuracy of the test system and standards used for calibration.
6.1.2 In addition to the QC measures required above, each test method should contain a detection limit and a quantitation limit
so that there is an indication of the lowest level at which the substance(s) determined by the test method can be detected and
measured.
6.1.3 Statistical tests should be done at a significance level of α ≤ 0.01, that is, ≥ 99 % confidence level. If other levels are
specified, the reason for deviation should be delineated in the test method.
6.1.4 The operational principles and characteristics of detectors used for radioactivity measurements are somewhat different from
those of instruments used for measurements of chemical and physical properties. Therefore, authors of ASTM International test
methods for radioactivity measurements should provide specific guidance within each test method, practice, or guide relative to
applicable QC program requirements. Guidance on the preparation and use of instrument tolerance and control charts can be found
in Practices D3648, Guide D3856, and ASTM MNL 7.
6.2 Calibration and Calibration Verification—For test methods requiring calibration of instrumentation, an appropriate number of
calibration standards mustshall be analyzed during the day that an analysis is performed to confirm that the instrument is properly
set up and required sensitivity is being obtained. The actual number of standards required will depend on the requirements of the
test method. For tests run infrequently, analysis of a single calibration standard to verify an existing calibration curve may suffice.
For tests run frequently, it may be necessary to intersperse verification standards with test samples. frequently without an internal
ASTM Manual on Presentation of Data and Control Chart Analysis, ASTM MNL 7.
D5847 − 22
standard, a verification standard should be analyzed at the beginning, at a given time or sample interval for a large batch and at
the end. Under these circumstances, it is recommended that a different standard concentration be used each time the calibration
is verified. Raw data (absorbance, intensity, etc.) and so forth) should be compared to data generated in the past under the same
conditions and should fall within three standard deviations of the mean value found in the past based on the pooled single operator
precision. Alternatively, data should be compared to the calibration limits stated in the test method or should be developed from
collaborative study data. Refer to Guide D3856 and Practices D3648 for further information on calibration checks.
6.2.1 For titrimetric test methods, titrants mustshall be standardized on a scheduled basis against a standard solution of known
concentration in duplicate or triplicate. The average normality/molarity is then used for calculation. The frequency of
standardization is left to the judgment of the writer of the test method and should be based on the stability of the titrant.
6.2.2 An alternate calibration procedure, such as an internal standard, external standard, or single-point calibration procedure,
mustshall be specified in the test method.
6.2.3 The test method mustshall establish the frequency of calibration and calibration verification.
6.3 Initial Demonstration of LaboratoryAnalyst Capability—A test mustshall be included in the test method to confirm that the
laboratoryanalyst is capable of running the test method and generating acceptable data. This test of laboratoryanalyst capability
will vary depending on the test method. Whenever appropriate, a precision and bias (as recovery) test is performed. For most test
methods this can be done by analyzing at least sevenfour replicates of a standard solution prepared from a reference material
containing the analyte at one of the concentration levels used in the collaborative study. The matrix and chemistry of the solution
should be such that, when spiked, results statistically equivalent to results produced in the collaborative study should be produced.
Each of the replicates should be presented to the operator as unknowns and should be interspersed with other samples following
the procedures used in the collaborative study. For some test methods, fewer replicates may be used, however, the statistical power
of the test is dependent on the number of replicates, and the meaningfulness of the study is reduced when fewer than seven
replicates are used. (For the examples in this practice, fewer than seven replicates are used for convenience.) Each replicate
mustreplicate shall be taken through the complete analytical test method including any pretreatment. The mean and standard
deviation of these results are then calculated as described in Practice D4375and compared to the single operator precision and
recovery found in the collaborative study.
NOTE 1—Initial Demonstration of LaboratoryAnalyst Capability—The type of test designed to assess the capability of a laboratory or operator an analyst
is at the discretion of the test method writer. It can be designed any way the test method writer believes is appropriate for the test method so long as it
provides meaningful data to ensure that the laboratory or operator analyst is capable of generating results that are valid and accurate within the confidence
limits defined in the precision and bias statement of the test method.
6.3.1 To establish that results produced by a laboratoryan analyst will be acceptable, the test method writer mustshall prepare a
table containing aan upper limit for acceptable precision and a range for acceptable recovery for the analytes determined by the
test method. The limit for acceptable precision is established by carrying out a one-sided F test at the α = 0.01 significance level,
and the range for acceptable recovery is established by carrying out a two-sided Student’s t test. Instructions for performing these
calculations are provided in 6.3.1.1 and 6.3.1.2. An example is given as Example 1 in Appendix X2.
6.3.1.1 The single-sided F test for a limit on precision is carried out using the square of the standard deviation found by the
operator, S , and the square of the expected pooled single operator standard deviation reported in the collaborative study, S , at
A O
the concentration level at which the precision study was carried out, and dividing the square of S by the square of S . The resulting
A O
value mustshall be less than or equal to the F value at the 0.01 significance level (99 % confidence level) for the number of degrees
of freedom in the operator’s study and the number of degrees of freedom in the collaborative study. The following formula is used:
Eq 1:
S
~ !
A
# F at df , df (1)
~ !
2 0.99 S S
A O
S
~ !
O
where:
S = standard deviation found by operator,
A
S = single operator standard deviation reported in collaborative study,
O
F = F value at 99 % confidence level,
0.99
df = degrees of freedom in laboratory’s study (usually 6 because 7 replicates are usually run), and
S
A
df = degrees of freedom for the single operator standard deviation estimate from the collaborative study.
S
O
If S < S , S /S is inverted to S /S in Eq 1Formula 1. . See Example 1 in Appendix X2.
A O A O O A
D5847 − 22
6.3.1.2 The two-sided Student’s t test for a recovery range is carried out using Eq 2:
¯ ¯
X 2 X
A
# t at df (2)
0.99
~n 2 1! ~S !
O
* *
S 2
Œ~ !
T
n
where:
X¯ = mean value found by laboratory,
A
X¯ = mean value found in collaborative study,
S = overall standard deviation found in collaborative study, and
T
S = single operator standard deviation found in collaborative study.
O
NOTE 2—If S > S from the collaborative study, let S = S :
O T O T
n = number of replicates used in laboratory’s precision study (usually 7),
t = Student’s t value at 99 % confidence level, and
0.99
df = degrees of freedom for the overall standard deviation estimate from the collaborative study (one less than the number of laboratories that
provided usable data at the concentration being tested.)
See Example 1 in Appendix X2.
6.3.2 The test method shall contain the requirement that the initial demonstration mustshall be repeated until the results fall within
these criteria.
6.4 Batch QC—The QC for routine operation is governed by a batch. A batch consists of a set of samples accompanied by QC
samples. The QC samples are an LCS, blank, MS, and optionally, a Dup or MSD. The result obtained for the QC samples that
accompany each batch mustshall meet performance criteria developed from collaborative study data using the procedures in this
practice or such as those found in Practice practice. D5789. The control limits are included in each test method. The task group
mustshall specify in the test method the consequence of a result for a QC sample that fails to meet a performance criterion.
6.4.1 The size and frequency of the batch is determined by identifying the key variables affecting the batch and selecting a batch
size and frequency so that these variables do not vary – are controlled – during analysis of the batch. The task group may specify
any batch size or frequency, or both, so long as the results of analysis of the LCS, blank, MS, and Dup or MSD can be assured
to be indicative of the variables affecting the remaining samples in the batch; that is, all samples in the batch are subject only to
the same set of random variables. If the risk or consequence of failure of a QC sample is high, the batch size should be small; if
the risk is low, the batch size may be large. The task group mustshall establish a maximum time between QC samples or the
maximum number of samples in the batch, or both, or instruct the test method user of the risk. See Explanation 2 in Appendix X1.
6.4.2 Method Blank (Blank)—Each test method shall require that, where applicable, a blank mustshall be analyzed with each batch,
as appropriate to the test method. The blank is taken through all the steps of the test method including any preservation and
pretreatment that may be necessary for samples. The value found for the blank should be below the detectionquantitation limit of
the test method or significantly below the confidence limits of the known concentration of the analyte in the associated test sample.
6.4.3 Laboratory Control Sample (LCS)—LCS—Each test method shall require that, where applicable, an LCS mustshall be run
with each batch, preferably at both the beginning and end of the batch, to determine if the measurement system is in control.
6.4.3.1 The LCS mustshall be prepared in the appropriate ASTM-grade water from a material that sufficiently challenges the test
method (see Explanation 3 in Appendix X1). The LCS must be can be a CRM obtained from an outside source or prepared inhouse
from materials of known purity and concentration. The LCS shall be taken through all steps of the test method. The concentration
of the LCS mustshall be known within a specified range of error. It is recommended that an independent reference material be used
as the LCS, where possible.
6.4.3.2 Selecting an analyte concentration for the LCS other than the one employed in the collaborative study will require, for
purposes of comparison, using a mean and standard deviation obtained from the collaborative test regression expressions at the
selected true concentration. In this instance, a procedure different from that in Example 1 in Appendix X2 mustshall be used to
determine the degrees of freedom for the Student’s t value for the two-sided test.
6.4.4 Matrix Spike (MS)—The MS tests the bias of the test method in the matrix being analyzed. A portion of at least one sample
D5847 − 22
from each batch is spiked with a known concentration of the analyte and the sample is taken through the test method including
any sample pretreatment that may be required. Guidance on spiking can be found in Guide D5810. The concentration of the analyte
in the spiked sample should be at least double, but not over five times, the concentration of the analyte in the unspiked sample.
For multi-analyte methods, such as gas chromatography (GC) or inductively coupled plasma (ICP) methods, it may be complicated
to spike all analytes at a concentration in the range of 2two to 5five times the concentration of the analytes in the unspiked sample.
For this condition, the analytes may be spiked at a fixed concentration or groups of analytes may be spiked at a few concentrations.
The spike concentration plus the concentration found in the unspiked sample mustshall fall within the demonstrated working range
for the test method.
6.4.4.1 Selecting an analyte concentration for the MS other than the one employed in the collaborative study will require, for
purposes of comparison, using a mean and standard deviation obtained from the collaborative test regression expressions at the
selected true concentration. In this instance, a procedure different from that in Example 1 in Appendix X2 mustshall be used to
determine the degrees of freedom for the Student’s t value for the two-sided test.
6.4.4.2 Two choices are available for development of performance criteria for MS recovery when multiple matrices are evaluated:
(1) develop overall performance criteria by pooling data across all matrices. These criteria will reflect the performance of the test
method across all matrices but will be broader than criteria developed for a specific sample matrix; (2) develop performance criteria
for each matrix and include a table of matrices and their respective performance criteria in the test method. Use the test data from
each matrix to develop the performance criteria for that matrix.
6.4.4.3 If, after the test method is balloted and approved, the test method will be applied to a matrix considerably different from
those used to create the performance criteria included in the test method, it may be appropriate for the task group to develop
additional performance criteria and add these criteria to the test method. Also, if the test method will be applied to a matrix
considerably different from that used in the collaborative study, the task group may stipulate in the test method that the method
user may develop performance criteria as specified in Guide D3856. In this event, the task group mustshall also stipulate that if
the performance criteria developed by the test method user are less stringent than those specified in the test method, the client or
data user mustshall be informed that less stringent performance criteria are being used. See Explanation 5 in Appendix X1.
6.4.4.4 The following procedure is used for development of performance criteria for recovery. An example is given as Example
2 in Appendix X2.
6.4.4.5 Include a test for percent recovery (P) of the spike using Eq 3:
A~V 1V! 2 BV
? s s?
P 5 100 (3)
CV
where:
A = estimated concentration obtained from analysis of the spiked sample,
B = estimated concentration obtained from analysis of the unspiked sample,
C = known concentration of analyte in the spiking solution,
V = volume of sample used, and
s
V = volume of spiking solution added.
Because both A and B are experimentally determined, the mean percent spike recovery (P¯) mustshall be estimated as follows:
¯
P 5 100/CV x¯ V 1V (4)
~ !~ ~ !!
T s
where:
x¯ = the expected mean of analytical results at concentration T, when T = CV / (V + V)
T s
x¯ = expected mean of analytical results at concentration T, when T = CV / (V + V) and
T s
and the standard Standard deviation of such percent spike recoveries (s ) is estimated as follows:as:
P
2 2 2 2 1/2
s 5 ~100/CV!~s ~V 1V! 1s ~V ! ! (5)
P A s B s
where:
s = the expected standard deviation of analytical results at measured concentration A, and
A
s = the expected standard deviation of analytical results at measured concentration B.
B
D5847 − 22
s = expected standard deviation of analytical results at measured concentration A, and
A
s = expected standard deviation of analytical results at measured concentration B.
B
6.4.4.6 A specific P value is acceptable if it is in the following interval developed from the collaborative test:
¯ ¯
~P 2 3 s ! # P #~P13 s ! (6)
~ ! ~ !
P P
6.4.4.7 A specific P value is acceptable if it is in the following interval developed from the collaborative test:
¯ ¯
~ ! ~ !
P 2 3~s ! # P # P13~s ! (6)
P P
If P does not fall within these limits, a matrix interference may be present in the sample selected for spiking. Under these
circumstan
...