Get documents about "Guidelines for Single Laboratory Validation (SLV) of Chemical
Document Sample
1 Best Practices for Single Laboratory Validation (SLV) of Chemical Methods for Trace Elements in
2 Foods
1 1
3 Cory J. Murphy , James D. MacNeil
1
4 Canadian Food Inspection Agency, Dartmouth Laboratory, 1992 Agency Drive, Dartmouth, Nova Scotia,
5 B3B 1Y9, Canada
6
7
8 Introduction
9
10 The use of analytical methods within a regulatory analysis or accredited laboratory framework
11 imposes certain requirements on both the analyst and laboratory. It is expected that regulatory analyses
12 will be conducted according to what may generally be described as “best practices” to ensure the
13 reliability of findings leading to regulatory action. In some situations, such analyses and the sampling
14 associated with them must also be conducted in a manner that meets requirements for legal proceedings,
15 including presentation as evidence in court. Under the International Organization for Standardization’s
16 (ISO) and International Electrotechnical Commission (IEC) general requirements, accredited laboratories
17 are expected to demonstrate both “fitness for purpose” of the methods for which they are accredited and
1
18 competency of their assigned analysts in performance of the methods . There is, therefore, activity in
19 many areas of regulatory analysis to develop consensus on best practices associated with particular
20 types of analyses.
21
22 In 1997 (amended in 2006), the Codex Alimentarius Commission (CAC) issued a general
23 guideline for analytical laboratories involved in the import and export testing of foods which contains four
2
24 principles :
25
26 The laboratory should have in place internal quality control procedures which meet the
3
27 requirements of the Harmonised Guidelines for Internal Quality Control in Analytical Chemistry ;
28 The laboratory should participate regularly in any available proficiency testing schemes,
29 appropriate to their area of testing, which have been designed and conducted as per the
30 requirements of the International Harmonized Protocol for Proficiency Testing of (Chemical)
4
31 Analytical Laboratories ;
32 The laboratory should become accredited according to ISO/IEC-17025:1999 General
33 requirements for the competence of calibration and testing laboratories (now ISO/IEC-
1
34 17025:2005 ) for tests routinely performed; and
35 The laboratory should use methods which have been validated according to the principles laid
36 down by the Codex Alimentarius Commission whenever such methods are available.
37
1
1 General requirements for validation of analytical methods according to principles laid down by the
2 Codex Alimentarius Commission are provided in the Codex Manual of Procedures, including provision for
5
3 “single laboratory” validation of analytical methods . Additional guidance is provided through a number of
4 general guidelines issued by a consensus process in international scientific organizations and
6,7,8,9,10
5 subsequently adopted as CAC guidelines . The CAC has also issued guidelines related to the
11
6 validation of methods used for the analysis of pesticide residues , mass spectrometric analysis of
12 13
7 pesticide residues , the estimation of uncertainty of measurements and the analysis of veterinary drug
14
8 residues in foods . A recent CAC guideline on the settlement of disputes over analytical test results also
15
9 makes reference to method validation requirements . However, there remains considerable
10 misunderstanding among analysts and laboratory managers as to precisely what is meant and what is
11 required to demonstrate “method validation”. Furthermore, no specific guidance on the validation of
12 methods used for the determination of elemental composition or element speciation is provided within
13 Codex documents to supplement the general guidance provided in other documents or contained in
14 guidance from independent international scientific organizations. Additional guidance on method
15 validation for future inclusion in the CAC Manual of Procedures and CAC guidelines is currently under
16 discussion in the Codex Committee on Methods of Analysis and Sampling (CCMAS) and other Codex
16
17 Alimentarius committees, but does not relate to this specific issue .
18
19 A new project was established by the Analytical Chemistry Division of the International Union of
20 Pure and Applied Chemistry (IUPAC) in 2009 to provide guidance on experimental designs suitable for
17
21 use in method validation , supplementing the general guidance previously provided by IUPAC on single
18
22 laboratory validation requirements . It may reasonably be anticipated that any such guidance will also be
23 adopted by the CAC. While compliance with CAC standards and guidelines is voluntary for member
24 states, subject to World Trade Organization (WTO) agreements, they do reflect international scientific
25 consensus on issues related to the analysis of foods. These guidelines can therefore be informative for
26 the development of guidance documents to be used within AOAC International for issues such as single
27 laboratory validation of analytical methods for trace elements, whether in foods or in other matrices.
28
29 Validation was defined by ISO in 1994 as “confirmation by examination and provision of objective
19
30 evidence that the particular requirements for a specified intended use are fulfilled ” . In analytical
31 chemistry, method validation was defined by Eurachem in 1998 as a process of “establishing the
32 performance characteristics and limitations of a method and the identification of the influences which may
33 change these characteristics and to what extent” and thereby “verifying that a method is fit for purpose,
20 21
34 i.e., for use for solving a particular analytical problem.” A recent guideline issued by CAC defines a
35 validated method as an “accepted test method for which validation studies have been completed to
22
36 determine the accuracy and reliability of this method for a specific purpose ” and validation as
23
37 “verification, where the specified requirements are adequate for an intended use” . The process includes
2
1 identification of the method scope and method performance characteristics. The scope defines the
2 analytes and the matrices in which they can be determined, the concentration range and any known
3 effects from interferences, while the expected performance characteristics are usually stated in terms of
4 precision and accuracy. The IUPAC Harmonized Guidelines for Single Laboratory Validation of Methods
5 of Analysis state that “strictly speaking, validation should refer to an ‘analytical system’ rather than an
6 ‘analytical method’, the analytical system comprising a defined method protocol, a defined concentration
1818
7 range for the analyte, and a specified type of test material.” An AOAC International guidance Formatted: Font: (Default) Times New
Roman, 12 pt
8 document defines validation as “the process of demonstrating or confirming the performance
24
9 characteristics of a method of analysis.” Similarly, the International Conference on Harmonisation of
10 Technical Requirements for Registration of Pharmaceuticals for Human Use (ICH) guidance states that
11 the “objective of validation of an analytical procedure is to demonstrate that it is suitable for its intended
25
12 purpose.”
13
14 Method validation can therefore be practically defined as a set of experiments which confirm that
15 an analytical method is suitable for its intended purpose when conducted using specific instrumentation
16 and within a specific laboratory environment in which the set of experiments have been conducted. An
17 inter-laboratory collaborative study is considered to provide a more reliable indicator of statistical
18 performance characteristics of the method because it requires testing of the method in multiple
19 laboratories, by different analysts using different reagents, supplies and equipment and working in
26
20 different laboratory environments . Validation of a method, even through collaborative study, does not,
21 however, provide a guarantee of method performance in any laboratory performing the method. This is
18
22 where a second term, verification, is sometimes used . In this context, verification may be defined as a
23 set of experiments conducted by a different analyst or laboratory on a previously validated method to
24 demonstrate that in their hands, the performance standards established from the original validation are
27
25 attained. Verification has been described as part of internal quality control (QC) procedures . That is, the
26 verification experiments demonstrate that the performance achieved meets requirements for attributes
27 such as scope (analytes/matrices), analytical range, freedom from interferences, precision and accuracy
28 that have been identified for suitable application of the method to the intended use in the initial method
29 validation.
30
31 The guidelines for conduct of an inter-laboratory collaborative study stress the importance that
32 the performance of the method should first be well-characterized in the developing laboratory (or
33 laboratory sponsoring the study) before the method is tested in multiple laboratories in the collaborative
28,29
34 trial . Current guidance from AOAC International for conduct of a collaborative study stresses the
35 importance of optimizing the performance of the method (usually demonstrated through completion and
30
36 reporting of a “single laboratory validation”) before attempting the collaborative study . Thus, following a
37 recognized approach based on scientific consensus to method validation within a single laboratory is
3
1 important not only to demonstrate “fitness for purpose” as required by accrediting bodies, but also to lay
2 the proper base when methods are proposed to be tested in an inter-laboratory method trial.
3
4 In contrast, method development is the series of experiments conducted to develop and optimize
5 a specific analytical method for an analyte or group of analytes. This can involve investigations into
6 detection/extraction of the analyte, stability of the analyte, analytical range, selectivity, ruggedness, etc. It
7 is important to note that method validation experiments will always take place after method development
8 is complete; that is, validation studies are intended to confirm method performance parameters which
9 were demonstrated during method development. Validation should not begin until method development,
10 including ruggedness testing, has been completed. A ruggedness design should identify if small changes
11 at certain steps of the analytical method, which might occur when other analysts use the method, affect
12 method results. A common approach is to vary seven factors simultaneously and measure these
31
13 changes to determine how they may affect method performance . Once method development and
14 ruggedness experiments are complete, the method should not be further modified or changed during the
15 validation process.
16
17 When validating a method for elements in food products, many factors should be considered
18 during the planning phase of the validation experimental design. For example, it should be determined if
19 the method is to be used in a regulatory environment, and if the analyte(s) of interest have a maximum
20 level (ML) which is to be assessed for compliance. In some cases, such as the analytes for which no safe
21 limits have been established, the purpose of the method may be to achieve the lowest possible detection
22 limit. The method may be intended for use in the determination of a single element in a particular matrix,
23 or it may require capability for multi-analyte analyses in various matrices. The availability of an authentic
24 blank matrix to be used as the analytical sample for method characterization should be considered. For
25 example, many elements are naturally present in some intended test matrices (such as arsenic or
26 cadmium in shellfish tissue). The inability to obtain authentic blank test sample material can therefore
27 cause many validation challenges when assessing parameters such as matrix effects and limits of
28 detection and quantification, particularly when attempting to use the signal of the “blank” as a basis for the
29 latter determinations.
30
31 Although food testing programs frequently include testing for a range of elements (predominantly
32 metals), there are actually few formally established MLs or other action limits for these analytes. The
33 Codex Alimentarius Commission has established limits for arsenic (total), cadmium and lead in a variety
32
34 of foods, total mercury in mineral waters and salt, methylmercury in fish and tin in canned goods .
35 Similarly, the European Union (EU) has established regulatory limits for cadmium, lead, mercury and tin in
33
36 a variety of foods . Requirements for analytical methods to enforce EU standards for lead, cadmium and
34
37 mercury in foodstuffs are the subject of another EU regulation . Canada has established maximum limits
4
35
1 for arsenic, lead and tin in various foods and standards for mercury in seafood have been set by both
36 37
2 Canada and the United States .
3
4 Table 1: Regulated Toxic Elements of Codex Alimentarius Commission and Various Countries
Organization/Country Regulated Element
32
Codex Alimentarius Commission As, Cd, Pb, Hg, methyl mercury in a variety of
foods
33
EU and member states Hg, Cd, Pb Sn in some foods
3536
Canada Hg in fish, Cd, Pb, Sn in some foods
37
USA Hg in fish
Japan Hg and methyl mercury in some fish
5
6 The aim of this paper on single laboratory validation (SLV) is to provide guidance for the scientist
7 when validating a method for trace elements in food as “fit-for-purpose” for an element or a group of
8 elements in those products. Definitions for common analytical chemistry terms used in food analysis are
9 taken from contemporary references and the procedures proposed for method validation are based on
10 available technical guidelines and recommended approaches. An example of a SLV experimental plan to
11 implement the proposed approach for methods used in elemental analysis in foods samples is provided.
12 The proposed approach is intended to address any specific requirements that are currently provided in
13 Codex Alimentarius guidance documents or in regulations or guidelines for the analysis of trace elements
14 in foods set by national or regional authorities, so is intended to be generally applicable for a variety or
15 potential users.
16
17 Definitions
18
19 In general, it is recommended that definitions included in the Codex Alimentarius Commission
21
20 “Guidelines on Analytical Terminology” should be used as a primary source for methods used in the
21 analysis of foods as these have been adopted after extensive international consultation and are taken
22 from authoritative sources, such as the Joint Committee for Guides in Metrology (JCGM), ISO, IUPAC
23 and AOAC International. Definitions of key terms used in method validation recommendations contained
24 in this document are contained in Appendix I, with a reference to the source. Adherence to these
25 definitions when reporting the validation of an analytical method will provide transparency to the process
26 and should eliminate the misunderstandings that can occur when different laboratories use different
27 definitions for the same analytical terminology. When definitions are available from multiple sources and
28 there are differences in the wording, accredited laboratories should use definitions contained in the
23
29 International Vocabulary of Metrology (VIM) as the primary source of definitions for analytical terms, as
5
1 national bodies performing laboratory accreditation under ISO/IEC-17025 refer to this source. The VIM is
2 the source of many of the definitions cited by the CAC.
3
4 For terms related to “sample” the analyst should use the nomenclature recommended by the Comment [M1]: I added this based on Steve’s
advice. I will fix the refs later, just needed to capture
5 International Union of Pure and Applied Chemistry (IUPAC) [Reference: Horwitz, H. (1990) Nomenclature for now.
6 for Sampling in Analytical Chemistry Pure & Appl. Chem. 62, 1193-1208.], for analytical chemistry, based
7 upon the International Organization for Standardization (ISO) recommendations. The terminology is also
8 supported by AOAC International [Reference: Official Methods of Analysis of AOAC INTERNATIONAL
9 (2005) AOAC INTERNATIONAL, Gaithersburg, MD, USA, Definition of Terms and Explanatory Notes,
10 Sample (23). OMA Online http://www.eoma.aoac.org/; accessed August 2, 2010. Terms most frequently
11 applicable to element analysis of foods are the following and will be used throughout this document:
12 Laboratory sample—sample or subsample sent to or received by the laboratory
13 Analytical (or test) sample—sample, prepared from the laboratory sample (by homogenization,
14 grinding, blending, etc.), from which analytical portions are removed for analysis.
15 Analytical (or test) portion—quantity of material removed from the analytical sample for analysis.
16 Analytical (or test) solution—solution prepared by dissolving (with or without reaction) of an
17 analytical portion in a liquid.
18
19 Concern has been expressed that the limit of detection (LOD) and the limit of quantification (LOQ)
20 should not always be used as mandatory fixed performance limits for validated methods, due to the
21 inherent variability which may be observed in the determination of these limits by different analysts using
22 different instruments. For example, an expert meeting on the validation of analytical methods noted in its
23 report that:
24 “LOD and LOQ are estimates of variable parameters, the values of which depend on various
25 factors, including the conditions of measurement and the experience of the analyst. The use of
26 these estimates in client reports can be misleading. In view of this, it was requested that the
27 FAO/IAEA expert consultation following the Workshop would consider that the lowest calibrated
28 level of the analysis be recommended to be used in client reports as an alternative to the LOD
38
29 and LOQ.”
30 The report of the subsequent expert consultation defined two terms to reflect the performance
31 characteristics which may be required of analytical methods used in a regulatory setting, the accepted
27
32 limit and the lowest calibrated level (See Appendix I) . More recently, the IUPAC Guidelines for Single
33 Laboratory Validation of Methods of Analysis advised that “the detection limit need not be part of
34 validation” when the actual concentration range measured by the method “does not include or approach”
35 this limit and also, regarding the limit of quantification, recommended that the measurement uncertainty
36 “as a function of concentration” should be assessed with regard to fitness for purpose, rather than using a
18
37 “fixed multiple” of the detection limit to establish a limit of quantification .
6
1
2 It also is important to note that while many analytical chemistry texts and older papers in scientific
3 journals use the term “specificity” for “selectivity”, the term “selectivity” is now recommended and use of
21
4 the term specificity is discouraged . It is considered that a method is either “specific” or it is “non-
5 specific”, while the term selectivity implies that there may be varying degrees of “selectivity”.
6
7 Performance Criteria
8
9 The Codex Committee on Methods of Analysis and Sampling (CCMAS) has recommended new
10 guidance on method performance with respect to implementation of the criteria approach for analytical
th 5
11 methods which has been included in the 19 Edition of the Codex Manual of Procedures . This guidance
12 is based on accepted approaches to the establishment of performance criteria for analytical
39,40,41
13 methods and was subject to extensive consultation by representatives of major international
14 organizations and national regulatory authorities prior to acceptance and implementation. It therefore is
15 recommended that these recommendations should be followed, particularly with regard to acceptable
16 performance for recovery and precision expected at various concentrations of analyte(s), which may be
17 found in Table 1, tiltled “Guidelines for establishing numeric values for the criteria” on page 53 of the CAC
5
18 Manual . Comment [M2]: Still wondering if we should
include the Table in this document. What about
19 Thompson , 2000. I also found a reference for
Pocklington, 1990. He also presents performance
20 Performance Characteristics criteria.
21
22 In order for a method to be considered “fit-for-purpose” certain performance requirements should
23 be evaluated and met. Listed below are the requirements typically considered in the validation of a
18
24 quantitative method of chemical analysis . A screening or confirmation method may require different,
25 usually fewer, parameters.
26 Accuracy- determined during method development, confirmed during validation Formatted: Bullets and Numbering
27 Analytical range- determined during method development, confirmed during validation
28 Intermediate precision – may be determined during method development or during method
29 validation
30 Limit of Detection (LOD) – determined during method development, confirmed during validation
31 Limit of Quantification (LOQ) - determined during method development, confirmed during
32 validation
33 Linearity- determined during method development, confirmed during validation
7
1 Matrix Effects - usually completed during method development phase, confirmed during validation
2 Measurement Uncertainty- determined during method development, confirmed during validation
3 Repeatability of detection system - may be completed during method development phase
4 Repeatability of method- determined during method development, confirmed during validation
5 Reproducibility (if appropriate) – by collaborative study, after single laboratory validation is
6 complete
7 Ruggedness - completed during method development phaseRuggedness - completed during Formatted: Bulleted + Level: 1 + Aligned at:
0.25" + Tab after: 0.25" + Indent at: 0.5"
8 method development phase
9 Selectivity - usually completed during method development phase, confirmed during validation
10 Matrix Effects - usually completed during method development phase, confirmed during validation
11 Sensitivity – usually assessed during method development phase, confirmed during validation
12 Limit of Detection (LOD) – determined during method development, confirmed during validation Formatted: Indent: Left: 0.25"
13 Limit of Quantification (LOQ) - determined during method development, confirmed during validation
14 Analytical range- determined during method development, confirmed during validation
15 Linearity- determined during method development, confirmed during validation Formatted: Bulleted + Level: 1 + Aligned at:
0.25" + Tab after: 0.25" + Indent at: 0.5"
16 Stability of analyte in standard solution - completed at start of method development
17 Stability of analyte in matrix - assessed once method is validated
18 Stability of analyte in extract/digest - assessed during method development Formatted: Bulleted + Level: 1 + Aligned at:
0.25" + Tab after: 0.25" + Indent at: 0.5"
19 Formatted: Bullets and Numbering
20 Accuracy- determined during method development, confirmed during validation Formatted: Indent: Left: 0.25"
21 Repeatability of detection system - may be completed during method development phase
22 Repeatability of method- determined during method development, confirmed during validation
23 Intermediate precision – may be determined during method development or during method validation
8
1 Reproducibility (if appropriate) – by collaborative study, after single laboratory validation is complete
2 Measurement Uncertainty- determined during method development, confirmed during validation
3 Technical Guidelines & Approaches
4 There are several additional considerations which affect the experimental design and specifically
5 the choice of matrices and analytes for validation of method performance. In a regulatory environment,
6 such as testing of foods for the presence of element contaminants or essential nutrients, there are many
7 sample materials which potentially require testing. Resources may not be available to fully validate each
8 analytical method for all analytes and matrices to which it may be applied. Therefore, the concepts of
11,27
9 representative commodity (matrix) and representative analytes have been proposed to facilitate
10 method validation and routine application. Using this approach, for example, in validating a method for
11 application to “fish”, representative matrices are salmon for “high fat” finfish, tilapia for “low fat”, shrimp for
12 “crustaceans”. Apples may be the representative matrix for apples and pears, oranges for “citrus fruit”
13 and strawberries for “berries”, while head lettuce may represent “leafy vegetables” and carrots may
14 represent “root crops”. One could also reference the food triangle when validating methods for multiple
15 foods as with this approach, foods can be divided into categories based on carbohydrate, fat, and protein.
16 Once the method has been validated for an element or elements on the “representative commodity”, it is
17 considered to be applicable to all commodities represented by that matrix until performance issues are
18 observed when the method is applied for the first time to a less commonly analyzed member of the group.
19 When this happens, further work is required to adapt and validate the method for that application.
20 Representative analytes may also be used when validating methods for elements in foods. For example,
21 with ICP-MS, you may choose a low, medium, and high mass element to represent all elements in a
22 multi-analyte screen.
23 Calibration using the analytical function or internal standardization approach usually assumes and
24 requires the availability of a representative blank test sample. However, situations may be encountered
25 when a material is unavailable for a particular commodity which is free of naturally incurred analyte.
26 Ideally, in such situations, a “representative commodity” which is free of the analyte can be chosen as a
27 surrogate material for the validation or to represent the commodity grouping of which the material is
28 considered a member. A surrogate material must closely relate to the matrix undergoing validation and
29 be blank of the analyte of interest to allow for efficient validation experiments to be completed. Use of
30 surrogates for the validation of methods for elements in foods has been well documented in the literatureI Comment [G.3]: I’m not sure it well documented
for elements. We’ usually perform spiek recovery
31 Examples of use of surrogates in method validation experiments include a natural water CRM used to experiments on the matrix of interest.
42
32 assess method performance in the absence of a CRM for vinegar and Brown Bread BCR CRM 191 as a
33 surrogate for honey (based on carbohydrate content) to assess method accuracy as a honey CRM was
43
34 unavailable . In some situations, there is no such material available and mixing of materials may be
35 required to approximate the composition of the target commodity. The following sections provide some
9
1 approaches which may be used when blank test sample material is unavailable for use in method
2 validation or for method calibration.
3
4 Linear Range and Calibration Curve
5 A typical chemical measurement process at trace concentrations includes the evaluation of two
6 types of calibration, one involving the determination of the detector response to changing concentrations
7 of pure standard (instrument response), while the second assesses the response to changes in analyte
44
8 concentration in the presence of matrix components and reagents (method performance) . In the case
9 of analsysis of elements in foods, standard addition techniques can also be used when a matrix effect has
10 been observed. The instrumental detection limit (IDL) is obtained using analyte standard solutions,. It Comment [G.4]: I’d like to make this a little
different without the referenced definitation. We
11 should be specified whether the reagents in these solutions match those of the analytical solutions. as the have the calibration curve (functin) from standard
solutions (that may contain small amounts of
12 detector responses may differ when the analyte standards are measured in the presence of different analyte) and a standard additions curve (function?)
for comparison. We compare the slopes for
13 reagents or because of the presence of the analytes of interest in the reagents. It should also be specified agreement.
14 whether calibration experiments are conducted using analyte processed through the method or analyte
15 added to prepared standard reagent blanks. These approaches using standards are commonly seen
16 when the method of “external calibration” is applied in a method, yet they will not necessarily yield the
17 same calibration results.
18
19 When the calibration procedure involves determining the response of the detection system to the
20 analyte(s) in the presence of matrix material, it is best described by the term analytical function.
21 Detection and quantification limits derived from this approach to calibration are the “method” detection
22 and quantification limits and provide a more accurate portrayal of the actual performance capabilities of
23 an analytical method. Since they are intended to reflect any interferences or matrix enhancement or
24 suppression effects, as well as analyte recovery from the matrix during the performance of the analytical
25 method, the detection and quantification limits determined from these experiments are in most cases (the
26 exception being when there are matrix enhancement effects on the detector response to the analyte)
27 higher than the equivalent instrumental detection limits and quantification determined using pure analyte,
28 or pure analyte in the presence of method reagents. This method of calibration is used when the standard
29 curve is generated using analyte in the presence of matrix and is variously referred to as use of “matrix
30 fortified” or “matrix matched” standards, meaning that the standards may be either added to matrix
31 (preferably blank matrix, if available) prior to processing for analysis or added to an extract or digest of
32 such matrix following processing. As when pure standards are either processed through a method, or
33 added to a reagent blank processed through the method, the results from the two approaches are not
34 necessarily the same. When blank matrix is not available and the method of standard additions is used to
35 generate a calibration curve, the same considerations apply.
36
10
1 An internal standard may be used with either approach to calibration. In either case, it is assumed
2 that the recovery of internal standard is the same as that of the analyte(s) measured, so it is important to
3 characterize this relationship. Many methods in current use for trace organic chemicals incorporate
4 isotope-labelled versions of the target analytes and in such cases it is expected that recoveries will be
5 identical for the analyte and, for example, a deuterated analogue. However, when an internal standard of
6 this nature is used in a multi-analyte method, the assumption of equivalence of recovery may not be
7 warranted.
8
9 There is no clear consensus in the scientific literature on the definition of “matrix matched” and
10 frequently questions may be left in the mind of readers as to precisely how standards were prepared for
11 calibration of a method. Some authors use the term “matrix matched” when the analytical standards are
12 spiked into blank matrix prior to extraction, while others use the term referring to the spiking of the
13 standards into a matrix extract. The latter approach should be used during methods development to
14 examine for matrix effects by comparing the response of pure standard solutions to the response of
15 standards prepared at the same concentrations in an extract of digest of representative blank test
16 sample.. Any differences in response observed may be attributed to matrix enhancement or suppression
17 effects. A second experiment can then be conducted to assess if there are also differences in response
18 related to analytical recovery. To make this comparison, compare the response to the standards prepared
19 in blank matrix digest with the response when the standards are spiked into blank matrix test sample prior
20 to digestion. Any differences observed may be attributed to analytical recovery. Thus, preparation of
21 standards by addition to blank matrix extract provides a correction for matrix effects, while preparation of
22 standards by spiking of blank matrix prior to extraction provides a correction for both matrix effects and
45
23 method recovery . Since there are currently no accepted definitions for these terms, which are used in
24 various contexts by different authors in different published papers, suggested definitions, at least to
25 qualify the meaning of these terms as used within this paper, are provided in Appendix I.C. Equally, when
26 authors refer to “external calibration”, it is not always clear whether they have used pure standards in a
27 solvent, pure standards prepared in a reagent blank or pure standards taken through the method. The
28 approach used should be clearly stated.
29
30 As noted above, the means by which the standards used in preparation of the calibration curve
31 are prepared may be very significant, as different methods of preparation and treatment of the standards
32 may influence the results. Ideally, the curve obtained when standards are prepared in the presence of
33 matrix should be assessed relative to the curve obtained when pure standards or standards prepared in
34 the presence of a reagent blank to determine if the sample treatment or the presence of matrix materials
35 has an influence on the detector response.
36
11
1 For example, various calibration strategies were evaluated in a study of the application of
46
2 ICP/MS analysis to the determination of arsenic, lead and selenium in wine . In this study, samples were
3 analyzed using calibrants prepared in surrogate solutions, with the initial calibrants prepared in a 10%
4 ethanol solution, then compared with results obtained when calibration was with standards prepared in
5 10% (w/w) alcohol and various amounts (500–2000mg/L) of potassium nitrate and nitric acid (0.1–0.5%,
6 w/w). Potassium typically is present in wine at concentrations in the g/L range, so this surrogate solution
7 used for preparation of calibration standards was intended to provide a closer approximation of the
8 behaviour of the elements being analyzed in undigested wine. It was found that analyte response was
9 about 40% higher in wine than in standards prepared in aqueous ethanol, but that addition of potassium
10 to the ethanol standards did not produce the same signal intensity as seen in the presence of the matrix.
11 The authors concluded that external calibration was not suitable for this analysis and that the method of
12 standard addition should be used. External calibration using aqueous standards was compared with
13 addition of analyte to a 0.2% (m/v) soil slurry solution for the determination of arsenic and selenium in
47
14 soils and sludges by ICP/MS . It was observed that the slopes increased when the standard addition
15 technique was used, leading to more accurate results. The importance of appropriate calibration
16 procedures and the selection of appropriate blank materials is also discussed in a paper on the
48
17 application of ICP/MS for the determination of trace elements in environmental samples . The authors
18 note that, among other issues, the slope of the calibration curve may be biased by the highest point on
19 the curve, which can particularly affect the accuracy of determination of low concentrations of analytes
20 when an extended calibration range is used. An inadequate number of procedure blank determinations
21 may also provide an insufficient basis for background subtraction. The discussion in this paper deals
22 primarily with external calibration, but the same issues are relevant to other calibration approaches. Comment [G.5]: Should references be provided?
Response: Some material with references moved
23 from the literature review to illustrate the point under
discussion.
24 Calibration Options
25
26 The European Commission Decision 2002/657/EC specifies microwave digestion for “elemental
49
27 confirmatory methods” and calibration by external standard or standard addition . These two approaches
28 are the ones most commonly found in published methods or applications, with the external standard
29 method of calibration, usually in combination with the use of an internal standard to correct for instrument
30 drift, being the more prevalent approach for analysis of elements in
,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66, 67,68
31 foods . External standard calibration also was the preferred
32 approach to calibration in a sampling of ICP/MS methods applied to non-food and environmental
69,70,71,72,73,74,75,76,77 78,79,80,81,82,83
33 matrices and clinical application . “Matrix-matched” external standard
34 calibration was used in the determination of total arsenic and total selenium concentrations in fish tissues
84
35 obtained from retail sources , while in a study to assess the applicability of ICP/MS to the determination
36 of metals in composite diet samples, matrix also was fortified with standards and analyzed to determine if
85
37 matrix effects were observed when compared to results for the same pure standards in solution . The
12
1 authors reported erratic recovery for the determination of arsenic and barium in fatty samples and
2 difficulty in quantification of cadmium due to the presence of high incurred concentrations of cadmium in
3 the material.
4
5 The next most common approach used in elemental analysis of foods was the method of
4648,86,87,88,89,90,91
6 standard addition . However, some studies have investigated multiple calibration
7 approaches. For example, both external standard calibration and standard addition were assessed In
42,92,93,94
8 several studies of elements in food matrices . In a study to determine concentrations of 26
9 potential elemental contaminants in wine using external calibration, recoveries were estimated by
6870
10 fortification of sample material (standard addition) . A recent study using laser ablation ICP/MS for the
11 determination of lead in blood samples included a comparison of calibration using aqueous standard
95
12 solutions with standards prepared in a matrix-matched solution prepared using a blood CRM .
13
14 Typically, no matter which calibration approach was used, method performance was assessed by
15 analysis of certified reference materials, comparison with alternate methods and/or participation in
16 available proficiency testing rounds. However, although external calibration, combined with internal
17 standards, appears to be the most widely used approach to instrument calibration, the other papers cited
18 above suggest additional experiments which may be conducted to assess issues such as potential matrix
19 effects on the analytical signal and on analyte recovery. Experiments using surrogate materials, standard
20 addition techniques and, when available, blank matrix, can provide a better understanding of the method
21 performance.
22
23 Assessment of calibration approaches
24
25 In trace organic analysis, external calibration (with use of internal standard to correct for recovery)
26 or preparation of “matrix fortified” calibration curves (also frequently with use of internal standard) appear
27 more prevalent than in trace element analysis. When methods for organic chemicals use mass
28 spectrometers as the detection system, the assessment of matrix effects on detector response becomes
29 especially important, while recovery of analyte is also usually an important issue. The International Union
96
30 of Pure & Applied Chemistry has issued guidance on recovery correction , but no equivalent consensus
31 on a standardized approach to the assessment of matrix effects on detector response has yet been
32 achieved. However, a systematic approach to the assessment of matrix effects and the differentiation of
4544
33 matrix effects from method recovery has been proposed for bioanalytical chemical analyses . In either
34 organic or elemental analyses, instrument calibration may be achieved by use of external or internal
35 standard calibrations, but the calibration approach used should be clearly stated.
36
13
1 Based on the literature reviewed, external standard calibration, with addition of internal standards
2 to correct for instrument drift, is the method of choice in most circumstances involving elemental analysis.
3 Although analytical recovery and matrix effects are not discussed in a number of the papers reviewed, it
4 is recognized that both are issues for the determination of trace elements by ICP/MS or other
5 instrumental techniques applied to digests of sample material. In ICP/MS analysis, internal standards are
6 frequently used to correct for both effects, even if the effects are not specifically characterized in the
7 report of the work. Several authors have used either standard addition or fortification of blank matrix
8 (when available) for method calibration. Inherently, there is no apparent reason why the calibration
9 procedures used in trace element analysis should differ significantly from those used in trace organic
10 analysis. The availability of true matrix test sample blanks can be an issue in either case and should be
11 dealt with appropriately. “Best practices” should therefore include a clearly described process to
12 determine whether external calibration, internal calibration using fortified matrix or standard addition has
13 been selected for method calibration. Matrix effects should be assessed using fortified digests of blank
14 test sample matrix or suitable surrogates when blank matrix is not available. Recoveries should be
15 assessed by fortification of matrix prior to extraction or digestion. The term “matrix matched” should be
16 used to describe experiments in which blank matrix test sample extracts or digests are fortified with
17 standards prior to instrumental analysis, while “matrix fortified” should refer to experiments in which the
18 matrix is fortified with standards prior to extraction or digestion. Certified reference materials should be
19 used, when available, to assess method accuracy and the effectiveness of the calibration procedures
20 used in compensating for recovery issues or matrix effects.
21
22 Analytical range of method and linearity
23
24 Linear range is determined by the injection of standard solutions in order to determine at what
25 level the instrument response no longer conforms to a linear equation (y = mx + b). This is determined in
26 the following manner:
27 Injections of standard solutions (minimum six) made up in similar reagents as the analytical test
28 solutions.
29 The concentrations of the solutions must be evenly spaced to determine the precise level at
30 which the calibration curve is no longer linear.
31 The range of concentration should encompass the expected concentration range from routine
32 samples if known.
33 The amount or concentration of the element injected is plotted against the instrument response to
34 determine the linear portion of the curve.
14
1 The instrument linear range is used to determine the analyte concentration range for which the method
2 will be fit for purpose. The calibration curve should be properly evaluated as the highest standard
3 solution may have a large impact on the line of best fit for the curve. Other points to consider are whether
4 to force the calibration curve through the origin, through the standard blank, or no forcing whatsoever.
4884
5 The approach used by the analyst in the laboratory can have drastic effects on reported results .
2
6 Linearity can be presented with R to determine if curve is linear over the concentration range chosen.
7 Matrix effect
8 The lab sample matrix may alter the results or create an enhanced or suppressed response from
9 the detector. Prior to experiments to determine if matrix effects are evident, several possibilities can be
10 explored to compare the results obtained for pure standard solutions in comparison to pure standard
11 solutions taken through the method, matrix matched and matrix fortified standards, The following
12 comparisons can then be made:
13 Pure standard solutions taken through the method compared to pure standard solutions made
14 prior to instrumental determinations is indicative of any losses of analyte which are related to the
15 method, while enhanced results may indicate reagent contamination.
16 Matrix fortified standards (standards added to “blank” test sample matrix prior to digestion)
17 compared with pure standards made prior to instrumental determinations provides an indication
18 of the combined effect of matrix enhancement/suppression effects and losses/gains related to the
19 method.
20 Matrix matched standards (standards added to “blank” test solutions after digestion) compared to
21 pure standard solutions made prior to instrumental determinations, provides an indication of
22 matrix enhancement/suppression effects only.
23 Blank Matrix
24 In order to determine matrix effect, calibration curves of pure standard solutions and matrix
25 matched standard solutions must be prepared and compared. The matrix matched calibration curves are
26 prepared by using extracted/digested blank matrix test portion solution as diluent. Prior to instrumental
27 determination, fortify the test solutions with aliquots of standards to provide the required concentration in
28 the final test solution to be equivalent to that of pure standard solutions. The standards are analysed by
29 duplicate or triplicate injections. When the results (as indicated by the slopes of the curves) obtained for
30 matrix-matched standard solutions are different than the results obtained for pure standard solutions, than
31 matrix suppression or enhancement effects are evident. Calibration curves for the pure standard solutions
32 and matrix fortified standard solutions are prepared by plotting the average response of the standard
33 solution against the standard concentration (Figure 1). As shown in Figure 1, matrix effects are not
34 evident, therefore, the use of pure standard solutions for the quantification of Hg in tuna is warranted.
15
1 Differences (>10%) of the slope of the matrix matched calibration curve in relation to that of the pure
2 standard curve, or significant changes in the instrument responses for corresponding standards indicates
3 that the matrix does indeed affect the instrument response. If this is the case, routine analysis will have
4 to be performed using matrix matched standards or possibly standard additions. Comment [M6]: NEED TO FIX GRAPH
Hg in Tuna - Neat vs Matrix Fortified Standards
0.25
y = 0.0138x + 0.0023
0.2 R2 = 0.9996
Instrument Response
y = 0.0136x + 0.0035
0.15
R2 = 0.9998
0.1 Pure Standards
Matrix Fortified Standards
0.05
0
0 5 10 15 20
Concentration (µg/L)
5
6 Figure 1: Matrix Effects Experiments for Total Hg in Tuna
7
8 No Blank Matrix
9 If blank lab sample matrix cannot be found, such as the case in many elemental analysis
10 techniques, a different approach is needed. First, test material must be characterized to determine the
11 analyte concentration in tissue by conducting a total of 20 determinations over 4 days. Then, prepare a
12 solution with similar analyte concentration to the matrix under investigation. Run matrix and prepared
13 solution with varying fortification levels (ie. 3 levels) in the same analytical run and repeat on a second
14 day. Plot theoretical analyte concentration versus instrument response for both matrix and solution on
15 the same graph (Figure 2). As shown in Figure 2, matrix effects are evident, therefore, the use of matrix
16 matched standards or standard additions is needed for the quantification. If the slopes of the curves
17 diverge by >10% or final fortification level concentrations show a >10% difference, then a matrix effect is
18 evident. If curves do not diverge (<10% difference in slopes), or final fortification level concentrations
19 show a <10% difference then no matrix effects are evident. Comment [M7]: FIX GRAPH
16
1
As+3 in Pears - Neat vs Matrix Fortified Standards
2
3
50000
4
y = 2721.1x - 34.931
R2 = 1
Instrument Response
5 Neat
Standards
6 25000 y = 2238.4x + 33.419
Matrix
R2 = 1
Fortified
Standards
7
8
0
0.0000 2.0000 4.0000 6.0000 8.0000 10.0000 12.0000 14.0000 16.0000
9
Theoretical Standard Concentration (ng/mL)
10
+3
11 Figure 2: Matrix Effects Experiments for Speciated Arsenic (As ) in Pears
12 Analyte Stability
13 While the stability of the analyte and the recovery of analyte from matrix may both be issues of
14 considerable concern in dealing with the analysis of organic chemicals, particularly residues and
15 contaminants in biological matrices, it is generally considered that stability of the analyte and recovery
16 from the original sample matrix, typically after a chemical digestion process, are of less concern in
17 elemental analysis. However, there can be exceptions, particularly when it is necessary to differentiate
18 between different chemical forms or species of an element. For example, conversion between oxidation
97 98
19 states of chromium has been observed , . Thus, part of the validation strategy for elemental analysis,
20 particularly when the analysis involves the speciation of different oxidation states of an element, should
21 include a demonstration that the species targeted is stable and recovered without loss or conversion
22 during the processing prior to instrumental analysis. Peer reviewed literature can also be a valuable
23 resource with respect to determining the stability of the analyte in solution.
24 Limit of Detection and Limit of Quantification
25 There are several approaches and multiple procedures typically used to determine LOD and LOQ
5,99
26 which have been presented in recent publications . It is recommended to use an approach that is
27 common to the field of analytical chemistry you are practicing and would be accepted by other scientific
28 colleagues. Whatever the technique used must be defended scientifically for the circumstances of the
17
1 method and provided that the basis for the estimates is clearly stated. Typically, LOD determinations can
2 be grouped into three common approaches; (1) an evaluation of the noise/background of matrix and/or
3 method blanks; (2) the use of the calibration curve and y-intercept, (3) and matrix spiking experiments. In
4 the more familiar approach, the LOD and LOQ are calculated as multiples of the standard deviation of the
5 mean response for a blank (typically, 3x for LOD and either 6x or 10x for LOQ). The alternative approach
6 estimates the LOD and LOQ based on method precision, so that LOD is based on “the rounded value of
7 the reproducibility relative standard deviation when it goes out of control (where 3 σR = 100%; σR = 33%,
8 rounded to 50% because of the high variability)”, while the LOQ is set at the concentration where σR =
9 25%. This latter approach may give a more practical and realistic estimate, as it is based on performance
10 where known concentrations of analyte are present and the method goes “out of control” for quantitative
11 purposes. The more commonly used alternative attempts to make an estimate based on “typical” blank
12 signals processed through a data system and is based on an approach that was used when data from
13 instruments typically were recorded on a strip-chart recorder with little or no intermediate modification of
14 the detector output.
15 There are three approaches, involving analysis of different types of materials, which have been reported
16 in papers on ICP/MS analysis of foodstuffs and environmental samples for the determination of limits of
17 detection and quantification (LOD, LOQ), The developers of an ICP/MS method for the determination of
18 arsenic, cadmium, lead and mercury in animal-derived foodstuffs (meat, fish, milk and milk products)
19 reported that they were able to identify suitable representative “blanks” which contained very low
20 concentrations of the target analytes and based the method LOD and LOQ on fortification of these
100
21 materials (test sample fortification as opposed to fortification of matrix test solutions) . These authors
22 noted the importance of the availability of homogeneous blank material for assessment of method
23 performance limits. Method performance was validated by procedures which included the analysis of
24 certified reference materials, comparison with results of other validated methods and participation in
25 proficiency tests (FAPAS). Instrumental detection and quantitation limits were based on method reagent
26 blanks, while diluted milk was used for assessment of method LODs and LOQs in a recent study of trace
Error! Bookmark not defined.
27 elements in milk . Similarly, the authors of a recent method for the analysis of
28 cadmium and lead in animal offal using ICP/MS calculated method performance characteristics (LOD,
101
29 LOQ) using fortified offals which were “blank” . Instrumental LODs were estimated using digestion
30 blanks (ng/mL), while method LOD’s and LOQ’s were estimated using digested milk (ng/g) in another
9073 5856
31 recent study . LOQs have also been estimated by dilution of CRMs .
32
Error!
33 An alternative approach, calculation of LOD from digestion blanks, was used by Chan et al
Bookmark not defined.51
34 . This approach was also used by Cubadda et al, who determined limits of detection
35 and quantification by analysis of digests of matrix blanks, with the result then adjusted for dilution factors
5250
36 applied to matrix digests to estimate these factors for the elements in matrix . Similar approaches to
Error! Bookmark not defined.53, 6058, Error! Bookmark not
37 the calculation have been described in numerous other papers
18
defined.79, Error! Bookmark not defined.Error! Bookmark not defined.
1 . In a recent study of selenium species and
2 concentrations in surface waters on the Canadian prairies, method detection limits for the ICP/MS
102
3 analysis were calculated from the standard deviation of the lowest calibrated concentration .
4 Blank Matrix
5 The limit of detection (LOD) must be determined for each analyte for which the method is
6 validated. This is done preferably by evaluating the noise level of 5 blank test samples per run on 4
7 separate instrument runs (n=20). One approach that could be used to determine the LOD for the analyte
8 in the matrix is by calculating the average noise of the 20 observations + 3SD. A procedure for estimation
103
9 of the LOD and the LOQ from the y-intercept of the calibration curve is used in many laboratories , as it
10 is considered to provide a more realistic estimate of these parameters than a direct calculation from the
11 observed noise level. The third approach is to determine the concentration at which the relative standard
5
12 deviation exceeds the requirements for quantitative analysis . With some techniques, a method reagent
13 blank taken through the method may be the only means of evaluating background noise. In this case 5
14 reagent blank samples per run on 4 separate instrument runs (n=20) would be completed and 3 standard
15 deviations of the background analyte level may be used as a good indicator of LOD, while 10 standard
16 deviations may be used to estimate the LOQ. Using this approach, the LOQ is approximately 3 times the
17 LOD.
18 No Authentic Blank Matrix
19 With some techniques, a reagent blank taken through the method may be the only means of
20 evaluating background noise. In this case 5 method reagent blank samples per run on 4 separate
21 instrument runs (n=20) would be completed and 3 standard deviations of the background analyte level
22 may be used as a good indicator of LOD, while 10 standard deviations may be used to estimate the LOQ.
23 Using this approach, the LOQ is approximately 3 times the LOD.
24 Since all approaches may give varied results for LOQ, an experiment could be conducted where
25 solutions of the analyte of interest are prepared at increasing intervals between the lowest and highest
26 calculated LOQ. If multiple injections of a particular solution has acceptable precision then this
27 concentration would be indicative of the LOQ.
28
29 Method Recovery
30 The recovery of the analyte(s) by the method for each validated matrix is to be determined by the
20
31 analysis of that matrix fortified with a specified amount of the analyte(s) . Certified reference materials
32 (CRM) representing a closely related matrix to the material undergoing validation should be used when
33 available to assess method recovery in combination with spiking experiments. Several studies in the
19
1 literature have supported the use of CRMs to access method recovery. In a recent report on monitoring
2 of the arsenic, lead and mercury content of traditional herbal preparations method performance was
104
3 verified by analysis of a certified reference material. . Method accuracy and precision were assessed
4 against guidelines issued by AOAC International for single laboratory validation of methods used in the
105
5 analysis of botanicals and dietary supplements . In a recent investigation of the effects of cooking on
6 concentrations of arsenic, cadmium, lead and mercury in foods, the authors reported use of a certified
106
7 reference material to assess method performance . In another study, the analysis of CRMs were used
107
8 to assess method performance and recoveries, based on the mean values for the elements in the
9 certified reference materials, ranged from 84-114%, demonstrating that, as in trace organic analysis, it
10 cannot be assumed that recovery is always 100%. Recovery studies are to be carried out on a minimum
11 of three fortification levels. These levels should be chosen depending on the intended use for the
12 method, and whether authentic blank matrix can be found. Five replicated analyses at each fortification
13 level shall be carried out on 3 separate days. Calculate the mean, standard deviation and % relative
14 standard deviation for each of the three levels.
15 Blank Matrix with ML/Target Level
16 If authentic blank material is available, and there is a published concentration of importance (ie.
17 Canadian ML for mercury in fresh tuna is 0.5 μg/g) then spike levels should be a factor of this ML. Spike
18 at ½ML, 1ML, and 2ML with each level replicated 5 times over three days. A CRM can also be used here
19 for additional information (if it contains appropriate concentration of the element of interest).
20 Blank Matrix with no ML/Target Level
21 If authentic blank material is available, and there is not a published concentration of importance
22 then spike levels should be a factor of the LOD. Spike at 3LOD, 10LOD, and the tissue equivalent
23 concentration of the upper limit of the calibration curve. Each level will be replicated 5 times over three
24 days. A CRM can also be used here for additional information (if it contains appropriate concentration of
25 analyte of interest).
26 No Blank Matrix with ML/Target Level
27 If authentic blank material cannot be found, a surrogate matrix (one in which low or non-
28 detectable levels of the analyte(s) of interest are present) may be used to fulfill validation requirements. If
29 an appropriate surrogate matrix cannot be found, spike solution is added to previously characterized
30 tissue so that target concentration(s) (background level + spike added) of tissue are equal to ½ ML, 1ML
31 and 2ML with each level replicated 5 times over three days. A CRM can also be used here for additional
32 information (if it contains appropriate concentration of analyte of interest).
20
1 No Blank Matrix with no ML/Target Level
2 If authentic blank matrix cannot be found and there is no published ML or concentration of
3 interest a surrogate matrix (one in which low or non-detectable levels of the analyte(s) of interest are
4 present) may be used to fulfill validation requirements. If an appropriate surrogate matrix cannot be
5 found, spike levels will be determined based on the previously characterized analyte concentration(s). A
6 low, medium, and high spike level will be used for this study, ie. spike equivalent to ½X, 1X, and 2X (or
7 upper limit of the calibration curve) of the analytical range of the characterized tissue concentration. A
8 CRM can also be used here for additional information (if it contains appropriate concentration of element
9 of interest).
10 In all scenarios, spiking at the LOQ may be required for verification purposes (if possible). Other
11 spiking levels may be used in place of those prescribed above. The objective is to have at least three
12 fortification levels (low, medium, and high) so confidence can be gained in the method’s and analysts’s
13 ability to recover the element of interest in the analytical test portion over a specific concentration range.
14 Calculate average spike recovery for each level, standard deviation and percent relative standard
15 deviation (%RSD) and compare to “Guidelines for establishing numeric values for analytical method
5
16 performance criteria, as recommended by the Codex Alimentarius Commission .”
17 Repeatability
18 There are two types of repeatability that are to be determined. The first type is a function of the
19 instrument. Instrument repeatability is determined by repeat injections of the standard solutions as well
20 as a fortified sample, naturally incurred material, or a CRM at least one level. The second type is the
21 method repeatability. It is determined by replicate digestion and analysis of a fortified, incurred material
22 at or near each of the fortification levels. Certified reference materials (CRM) representing a closely
23 related matrix to the material undergoing validation should be used when available to assess method
24 repeatability in combination with experiments with fortified matrix.
25 Instrument Repeatability
26 Inject each of the standard solutions that are used to prepare the working calibration curve as
27 well as an incurred or fortified test sample at one of the spike levels 5 times. These injections should be
28 done in random order to reduce any bias. Calculate average, standard deviation and percent relative
29 standard deviation (%RSD). In most cases, precision should not exceed 10%RSD for replicate injections.
30 If this is the case, the analyst must investigate why the instrument is not repeatable.
31 Method Repeatability
21
1 Prepare pools of test sample material with levels of the analyte(s) at or near the same
2 concentrations that were used for the method recovery studies. This may be completed by using incurred
3 material or by fortifying test sample material (blank or incurred) with the required amount of the analyte(s).
4 Prepare five replicate test portions of each of these test samples and analyse on the same day. This
5 process is to be repeated on two more days. A CRM can also be used here for additional information (if it
6 contains appropriate concentration of analyte of interest). Calculate average, standard deviation and
7 percent relative standard deviation (%RSD and compare to “Guidelines for establishing numeric values
5
8 for analytical method performance criteria, as recommended by the Codex Alimentarius Commission .”
9
10 Intermediate Precision
11 This parameter is used to determine if there are biases in the method. The bias can come from
12 the analyst, instrumentation, or other sources. To study this parameter prepare pools of test sample
13 material with, either incurred or fortified, levels of the analyte(s) at or near the same concentrations that
14 were used for the recovery and repeatability studies. The same material should be used as was prepared
15 for the repeatability studies if sufficient is remaining. Certified reference materials (CRM) can also be
16 used for the evaluation of intermediate precision. The CRM used should be a closely related matrix as
17 the material undergoing validation.
18 At a minimum the study must be carried out by an additional analyst over three separate days. The
19 second analyst is to prepare all fresh reagents and the test samples are to be analysed in 5 replicates
20 over three separate days by the second analyst. If multiple instruments are available then the study by
21 the second analyst must be carried out on the second instrument, to take into account any instrument
22 bias.
23
24 Measurement Uncertainty
25 The uncertainty of a result from a chemical analysis can be caused by many steps in the process.
26 In practice the uncertainty on the result may arise from many possible sources, including examples such
27 as incomplete definition, sampling, matrix effects and interferences, environmental conditions,
28 uncertainties of masses and volumetric equipment, reference values, approximations and assumptions
20 108
29 incorporated in the measurement method and procedure, and random variation. ,
30 A document which provides extensive guidance on the estimation of measurement uncertainty in
109
31 analytical methods is available from Eurachem . Rather than attempting to calculate the uncertainty
22
1 from each factor independently and combining the results, an approach is to the look at the methodology
2 as a whole and group the uncertainty into two categories: Accuracy and Precision.
3
4 Data sets that are to be considered for Accuracy are; recovery, CRM data, PT samples etc. Data
5 to be included with precision are; intermediate precision, in-house check samples, CRM data, etc. The
6 relative uncertainty (MU) for the method is calculated by determining square root of the sum of the
7 squares of the respective relative uncertainties for accuracy and precision.
8
2 2
9 MU (RU (accuracy) RU ( precision)
10
11 Ruggedness
12
13 The ruggedness of an analytical method is the resistance to change in the results produced by an
14 analytical method when minor deviations are made from the experimental conditions described in the
15 procedure. The ruggedness of a method is tested by deliberately introducing small changes to the
16 procedure and examining the effect on the results. Methods should be ruggedness tested as the last
17 stage of method development, prior to method validation. Ruggedness testing should not be used to
18 determine critical control points (these should be determined earlier during method development) and
19 critical control points should not be included in ruggedness testing, as they are known to have a
20 significant impact on the analysis. Ruggedness testing does not need to be performed for each matrix
21 tested as this examination of matrix effects should be performed in method development. The matrix
22 used for ruggedness testing should be as representative as possible of the proposed workload, i.e. the
23 most common matrix, or material representative of the matrix components (non-fat, low-fat, high-fat).
24
25 Examples of variables to be tested for elemental analysis include:
26 pH
27 temperature (digestion)
28 acid concentrations
29 reagents (age, source, concentrations)
30 delays in continuing the method at different stages
31 analytical portion mass
32 extraction/digestion (time, technique, solvents/acids)
33 different instruments
34 different instrument parameters
35
23
31
1 The easiest approach is to use Youden’s factorial approach , where seven variables can be
2 combined in a specific manner to determine the effects of all seven variables using eight combinations in
3 a single experiment. If the method has fewer variables to be tested, then blanks can be included, or
4 variables can be examined individually. The experiment should also be repeated on two separate days in
5 order to eliminate the possibility of a single sample affecting the outcome. Values for each sample should
6 be spike recoveries or concentrations if incurred/fortified tissue is being used.
7
8
Sample Factor Combinations Measurement
1 ABCDEFG s
2 ABcDefg t
3 AbCdEfg u
4 AbcdeFG v
5 aBCdeFg w
6 aBcdEfG x
7 abCDefG y
8 abcDEFg z
9
10 To determine effect of individual factor,
11
12 Effect of A and a: [(s + t + u + v)/4] – [(w + x + y + z)/4] = J
13 This simplifies to: (4A/4) – (4a/4) = J
14 Effect of B and b: [(s + t + w + x)/4] – [(u + v + y + z)/4] = K
15 Effect of C and c: [(s + u + w + y)/4] – [(t + v + x + z)/4] = L
16 Effect of D and d: [(s + t + y + z)/4] – [(u + v + w + x)/4] = M
17 Effect of E and e: [(s + u + x + z)/4] – [(t + v + w + y)/4] = N
18 Effect of F and f: [(s + v + w + z)/4] – [(t + u + x + y)/4] = O
19 Effect of G and g: [(s + v + x + y)/4] – [(t + u + w + z)/4] = P
20
21 After calculating the differences between factors (J-P) examine those values. Small changes in
22 factors with larger differences can lead to significant changes in results. Determine which factors create
23 statistically significant changes by performing a two-sample t-test assuming equal variance for each
24 factor. If the p-value is <0.05 the factor is significant, if the p-value is >0.15 the factor is not significant,
25 and if 0.05<p<0.15 the factor may be significant.
26
27 If factors are determined to be significant the procedural instructions dealing with those factors
28 should be made more specific and the ruggedness testing repeated, including those factors with
24
1 intermediate p-values (0.05<p<0.15). These factors may be determined to be critical control points, and if
2 this is the case the procedural instructions should be changed to reflect an acceptable variance.
3
4 If intermediate precision data are available then an additional comparison may be made.
5 Comparing the standard deviation of the method as determined in intermediate precision testing to the
6 standard deviation of the differences of factors examined during ruggedness testing may reveal that a
7 combination of factors has a significant effect on the method even though no individual factors have a
8 significant effect on the method. In this case the procedural instructions should be made more specific
9 and the ruggedness testing should be repeated using the more specific instructions.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
25
1
2
3
4 Appendix 1: Example of an Experimental Method Validation Plan for Tin in Canned Foods
5
6
7 Method Title: Validation of a Method for the Determination of Tin (Sn) in Canned Pears
8 Project Participants: Analyst 1, Analyst 2
9 Start Date: January 1, 2009
10 Projected Completion Date: March 31, 2009
11 Instrument(s): ICP-OES
12 ML/Target Level: 250 μg/g
13 NOTE: For this validation plan, tin levels were negligible in the pear matrix
14
15 Linearity Survey:
16 Analysis of standard solutions prepared at equal concentration intervals (minimum 6) to
17 determine at what concentration the calibration curve is no longer linear. Should include expected
18 concentration of samples if known.
19
20 Analytical Range:
21 Determined using the analytical solution quantification limit as the lower limit and the upper end of
22 the linear range as the upper limit
23
24 Matrix Effects:
25 Blank pear material is digested as per the method. The resulting digest is then used as diluent to
26 prepare a calibration curve. A pure standard solution curve (ie. 10% HCl) is also prepared. The
27 neat and matrix matched standards are run on the same day on the ICP-OES to determine if
28 slope of curves are similar. If similar, then pure standards can be prepared for the remainder of
29 the validation experiments and for quantifying samples. If slopes are different (>10% difference),
30 then matrix matched calibration curves or methods of standard additions will need to be prepared
31 for the remainder of the validation experiments and for quantifying samples.
32
33 LOD/LOQ:
34 Since authentic blank pears can be found, run 20 blank matrix tissue samples through digestion
35 over 4 days and measure the noise level for each. Calculate the average noise and standard
36 deviation of the 20 data points. LOD may be calculated several ways such as LOD = 3SD or
37 LOD = noise + 3SD, or using the y-intercept approach. LOQ = 3LOD. Method detection limit
26
1 (MDL) would then be calculated by taking into account typical dilution factors used and sample
2 weight into account. Since all approaches may give varied results for LOQ, an experiment will be
3 conducted to verify the calculated LOD. Spike at 2-5X the expected LOD with atleast 7 replicates
4 within the run.
5
6 Stability:
7 Analyte in standard solution:
8 Compare a freshly prepared working standard solution to one that has been made and stored.
9 Measure at intervals over a specified time period to determine Sn stability in solution.
10
11 Analyte in matrix:
12 Run a canned pear sample at specified time intervals over the time that the test sample would
13 typically be stored to see if Sn levels degrade or concentrate. Ensure standards used for the
14 calibration curve are not degraded or expired.
15
16 Analyte in sample digest:
17 Digest a sample with a known concentration of tin and measure daily for a period of a week.
18
19 Recovery:
20 Since blank pear matrix is available and an ML exists, fortify blank matrix at 3 levels: ½ ML, 1ML
21 and 2ML. Run five samples per level, on 3 separate days. CRM, if available, will also be used to
22 give an indication of method recovery.
23
24 Repeatability:
25 Instrument:
26 Inject each of the standards of the calibration curve, as well as a CRM/fortified sample/incurred
27 sample, five times in random order on the same day.
28
29 Method:
30 Make pooled pear test sample material at three levels ½ ML, 1ML and 2ML. Run five replicates
31 per each level over 3 separate days. If a CRM exists for Sn in canned pears (or canned fruit)
32 incorporate this into these experiments as well.
33
34
35 Intermediate Precision:
36 A second analyst in the same laboratory repeats the procedure for method repeatability as above.
37 If possible use a different ICP-OES instrument, different reagents, etc..
27
1
2 Reproducibility:
3 Other laboratories would analyze pooled material at 3 levels as well CRMs if available.
4
5 Measurement Uncertainty:
6 Data from recovery/repeatability experiments will be used to determine accuracy and precision.
7 Upon implementation of the method, these values will be updated.
8
9 Other:
10 Other analyses may be required to further validate the method for use such as analysis of CRMs,
11 inter-laboratory samples, proficiency samples and in-house check samples add to the method validation
12 data and should be included as this data is available.
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
28
1
2
3 References
1 ISO (19992005). ISO/IEC-17025(E): General requirements for the competence of calibration and
testing laboratories. 2nd Edition. International Organization for Standardization, Geneva.
2 CAC (1997) Guidelines for the assessment of the competence of testing laboratories involved in
the import and export control of food; CAC/GL 27-1997 (amended 2006). Joint FAO/WHO Food
Standards Program; http://www.codexalimentarius.net/download/standards/355/CXG_027e.pdf;
accessed February 23, 2010.
3 Thompson, M., & Wood, R. (1995). Harmonized Guidelines for Internal Quality Control in Analytical
Chemistry Laboratories. Pure & Appl. Chem. 67: 649-666.
4 Thompson, M., Ellison, S.L.R, & Wood, R. (2006) The International Harmonized Protocol for the Formatted: Font: 9 pt
Proficiency Testing of (Chemical) Analytical Laboratories”, Pure & Appl. Chem. 78: 145-196.
Formatted: Indent: Left: 0", Hanging: 0.39"
Thompson, M. and Wood, R. 1993. International Harmonized Protocol for Proficiency Testing of
(Chemical) Analytical Laboratories. Pure & Appl. Chem. 65: 2132-2144. Formatted: Font: 9 pt
th
5 CAC (2009). Codex Alimentarius Commission Procedural Manual, 19 ed., Joint FAO/WHO Food
Standards Program; ftp://ftp.fao.org/codex/Publications/ProcManuals/Manual_19e.pdf; accessed
April 6, 2010.
6 CAC (1995) Protocol for the design, conduct and interpretation of method performance studies;
CAC/GL 64-1995. Joint FAO/WHO Food Standards Program;
http://www.codexalimentarius.net/download/standards/10918/CXG_064e.pdf; accessed February
23, 2010.
7 CAC (1997) Food control laboratory management: recommendations; CAC/GL 28-1995, Rev.1-
1997; Joint FAO/WHO Food Standards Program;
http://www.codexalimentarius.net/download/standards/356/CXG_028e.pdf; accessed February 23,
2010.
8 CAC (2001) Harmonized IUPAC guidelines for the use of recovery information in analytical
measurement; CAC/GL 37-2001. FAO/WHO Food Standards Program;
http://www.codexalimentarius.net/download/standards/376/CXG_037e.pdf; accessed February 23,
2010.
9 CAC (2003) Harmonized IUPAC guidelines for single-laboratory validation of methods of analysis;
CAC/GL 49-2003. Joint FAO/WHO Food Standards Program;
http://www.codexalimentarius.net/download/standards/10256/CXG_049e.pdf; accessed February
23, 2010.
10 CAC (1997) Harmonized guidelines for internal quality control in analytical chemistry laboratories;
CAC/GL 65-1997. Joint FAO/WHO Food Standards Program;
http://www.codexalimentarius.net/download/standards/10920/CXG_065e.pdf; accessed February
23, 2010.
11 CAC (2003) Guidelines on good laboratory practice in residue analysis; CAC/GL 40-1993, Rev.1-
2003. Joint FAO/WHO Food Standards Program;
http://www.codexalimentarius.net/download/standards/378/cxg_040e.pdf; accessed February 23,
2010.
29
12 CAC (2005) Guidelines on the use of mass spectrometry (MS) for identification, confirmation and
quantitative determination of residues; CAC/GL 56-2005. Joint FAO/WHO Food Standards
Program; http://www.codexalimentarius.net/download/standards/10185/cxg_056e.pdf; accessed
February 23, 2010.
13 CAC (2006) Guidelines on estimation of uncertainty of results; CAC/GL 59-2006. Joint FAO/WHO
Food Standards Program;
http://www.codexalimentarius.net/download/standards/10692/cxg_059e.pdf; accessed February
23, 2010.
14 CAC (2009) Guidelines for the design and implementation of national regulatory food safety
assurance programme associated with the use of veterinary drugs in food producing animals;
CAC/GL 71-2009. Joint FAO/WHO Food Standards Program;
http://www.codexalimentarius.net/download/standards/11252/CXG_071e.pdf; accessed February
23, 2010.
15 CAC (2009) Guidelines for settling disputes over analytical (test) results; (CAC/GL 70-2009). Joint
FAO/WHO Food Standards Program;
http://www.codexalimentarius.net/download/standards/11256/cxg_070e.pdf; accessed February
23, 2010.
16 CAC (2008). ALINORM 08/31/23; Report of the twenty-ninth session of the Codex Committee on
Methods of Analysis and Sampling, Budapest, Hungary, 10 - 14 March 2008, Appendix II, pp. 31-
33; http://www.codexalimentarius.net/download/report/699/al31_23e.pdf; accessed March 24,
2009.
17 IUPAC (2009) Project 2009-006-1-500: Experimental Requirements for Single-Laboratory
Validation; Ellison, S.L.R. (chairman); http://www.iupac.org/web/ins/2009-006-1-500; accessed
February 1, 2010.
18 Thompson, M., Ellison, S.L.R., and R. Wood. 2002. Harmonized Guidelines for Single Laboratory
Validation of Methods of Analysis. (IUPAC Technical Report). Pure Appl. Chem., Vol. 74(, No. 5), Formatted: Font: Bold
pp. 835–855.
Formatted: Normal, Indent: Left: 0",
Hanging: 0.39", Space After: 12 pt, Don't
adjust space between Latin and Asian text,
19 ISO (1994). ISO 8402: 1994, Quality management and quality assurance -- Vocabulary Don't adjust space between Asian text and
International Standards Organization, Geneva ; withdrawn 2000-12-08. numbers
20 Eurachem (1998). The Fitness for Purpose of Analytical Methods - A Laboratory Guide to Method
Validation and Related Topics. http://www.eurachem.org/guides/valid.pdf; accessed March 24,
2009.
21 CAC (2009) Guidelines on Analytical Terminology (CAC/GL 72-2009). Codex Alimentarius
Commission, Joint FAO/WHO Food Standards Program;
http://www.codexalimentarius.net/download/standards/11357/cxg_072e.pdf; accessed January 28,
2010.
22 ICCVAM (2003) Guidelines for the nomination and submission of new, revised and alternative test
methods. http://iccvam.niehs.nih.gov/SuppDocs/SubGuidelines/SD_subg034508.pdf; accessed
February 23, 2010.
30
23 VIM (2008) International Vocabulary of Metrology – Basic and general concepts and associated
rd
terms, 3 edition, JCGM 200.
24 AOAC Int. (2002) AOAC Guidelines for Single Laboratory Validation of Chemical Methods for
Dietary Supplements and Botanicals, AOAC International, Gaithersburg, MD;
http://www.aoac.org/Official_Methods/slv_guidelines.pdf; accessed February 25, 2010.
25 ICH (2005) Validation of analytical procedures: text and methodology, ICH Harmonized Tripartite
Guideline Q2(R1), International Conference on Harmonisation of Technical Requirements for
Registration of Pharmaceuticals for Human Use; http://www.ich.org/LOB/media/MEDIA417.pdf;
accessed February 25, 2010.
26 FAO (1998) “Validation of analytical methods for food control”, Report of a Joint FAO/IAEA Expert
Consultation, December 1997, FAO Food and Nutrition Paper No. 68, FAO, Rome.
27 Alder, L, Holland, PT, Lantos, J, Lee, M, MacNeil, JD, O’Rangers, J, van Zoonen, P, Ambrus, A.
2000. Guidelines for Single-Laboratory Validation of Analytical Methods for Trace-level
Concentrations of Organic Chemicals in Principles and Practices of Method Validation, Fajgelj, A, &
Ambrus, A (ed.). ISBN 0-85404-783-2. The Royal Society of Chemistry, Cambridge, UK, pp. 179-
248.
28 Horwitz, W. (1988) Protocol for the design, conduct and interpretation of collaborative studies. Pure
& Appl. Chem., 60 (6): 855-864.
29 Horwitz, W. (1994) Protocol for the design, conduct and interpretation of method-performance
studies. Pure & Appl. Chem., 67 (2): 331-343.
30 AOAC Int. (2010) Guidelines for Collaborative Study Procedures To Validate Characteristics of a
Method of Analysis. AOAC International, Gaithersburg, MD.
http://www.aoac.org/Official_Methods/Collaborative_Study_Validation_Guidelines.pdf; accessed
February 11, 2010.
31 Youden, W.J., & Steiner, E.H. (1975) Statistical Manual of the AOAC, pp.33-36.
32 CAC (20072009). Codex General Standard for Contaminants and Toxins in Foods - Codex Stan
193-1995, Rev.3 5-20097; http://www.codexalimentarius.net/web/standard_list.jsp; accessed
August 8, 2010..
33 EU (2006). Commission Regulation (EC) No 1881/2006 of 19 December 2006 setting maximum
levels for certain contaminants in foodstuffs. Official Journal of the European Union, L364: 5-24.
34 EU (2001). Commission Directive 2001/22/EC of 8 March 2001 laying down the sampling methods
and the methods of analysis for the official control of the levels of lead, cadmium, mercury and 3-
MCPD in foodstuffs. Official Journal of the European Communities, L77: 14-21.
.
35 HC (2009). Food & Drug Act and Regulations, B.15.003;
http://laws.justice.gc.ca/en/showtdm/cr/C.R.C.-c.870; accessed March 25, 2009.
36 HC (2007). Canadian Standards ("Maximum Limits") for Various Chemical Contaminants in Foods,
Health Canada, Ottawa, ON, Canada; http://www.hc-sc.gc.ca/fn-an/securit/chem-
chim/contaminants-guidelines-directives-eng.php; accessed March 25, 2009.
37 U.S. Food and Drug Administration, Office of Enforcement. (Rev. 1995, March5-2005). Compliance Formatted: Indent: Left: 0", Hanging: 0.39"
Ppolicy Gguides, Ssec. 540.600 Fish, Shellfish, Crustaceans and Other Aquatic Animals - Fresh,
Frozen or Processed - Methyl Mercury (CPG 7108.07), p. 288;
31
http://www.fda.gov/ICECI/ComplianceManuals/CompliancePolicyGuidanceManual/ucm074510.htm Field Code Changed
; accessed August 8, 2010.
.
Formatted: Indent: Left: 0", Hanging: 0.39"
38 IUPAC (1999) Workshop Report: International Workshop on Principles and Practices of Method
Validation, November 4-6, 1999, Budapest, Hungary;
http://old.iupac.org/symposia/conferences/method_validation_4nov99/report.html, accessed June
1, 2010.
39 Horwitz, W., Kamps, L. R. and Boyer, K. W. (1980) Quality assurance in the analysis of foods and
trace constituents. J. Assoc. Off. Anal.Chem., 63: 1344-1354.
40 Horwitz, W. and Albert, R. (1996) Reliability of the Determinations of Polychlorinated Contaminants
(Biphenyls, Dioxins, Furans) J. AOAC Int., 79: 589-621.
41 Thompson, M. (2000) Recent trends in inter-laboratory precision at ppb and sub-ppb
concentrations in relation to fitness for purpose criteria in proficiency testing. Analyst 125: 385-386.
42 Ndung’u, K. Hibdon, S., & Flegal, A.R. (2004) Determination of lead in vinegar by ICP-MS and Formatted: Font: (Default) Arial, Not
GFAAS: evaluation of different sample preparation procedures. Talanta 64: 258–263. Superscript/ Subscript
Formatted: Font: (Default) Arial
43 Caroli, S., Forte, G., Iamiceli, A.L., & Galoppi, B. (1999) Determination of essential and potentially
Formatted: English (Canada)
toxic trace elements in honey by inductively coupled plasma-based techniques. Talanta 50: 327-
336. Formatted: Not Superscript/ Subscript
Formatted: Indent: Left: 0", Hanging: 0.39"
44 AOAC International (2002) Validation Guidelines. Appendix D: Guidelines for Collaborative Study
Procedures To Validate Characteristics of a Method of Analysis, AOAC International, Gaithersburg,
MD; http://www.aoac.org/Official_Methods/Collaborative_Study_Validation_Guidelines.pdf;
accessed June 1, 2010.
45 Matuszewski, B.K., Constanzer, M. L., & C. M. Chavez-Eng (2003) Strategies for the Assessment
of Matrix Effect in Quantitative Bioanalytical Methods Based on HPLC-MS/MS. Anal. Chem. 75:
3019-3030.
46 Grindlay, G., Mora, J., Gras, L., & de Loos-Vollebregt, M.T.C. (2009) Ultratrace determination of
Pb, Se and As in wine samples by electrothermal vaporization inductively coupled plasma mass
spectrometry. Anal. Chim. Acta 652: 154–160.
47 Tseng, Y.J., Liu, C.-C., & Jiang, S.-J. (2007) Slurry sampling electrothermal vaporization inductively
coupled plasma mass spectrometry for the determination of As and Se in soil and sludge. Anal.
Chim. Acta 588: 173–178.
48 Salomon, S., Jenne, V., & Hoenig, M. (2002) Practical aspects of routine trace element
environmental analysis by inductively coupled plasma–mass spectrometry. Talanta 57:157–168.
49 EU (2002) Commission Decision of 12 August 2002 implementing Council Directive 96/23/EC
concerning the performance of analytical methods and the interpretation of results (2002/657/EC ),
Official Journal of the European Communities, L221:8-36.
50 Rodushkin, I. (1998) Capabilities of high resolution inductively coupled plasma mass spectrometry
for trace element determination in plant sample digests. Fresenius J. Anal. Chem. 362: 541-546.
51 Benkhedda, K. Robichaud, A., Turcotte, S., Béraldin., F.J., & Cockell, KA. (2009) Determination
of Total Iodine in Food Samples Using Inductively Coupled Plasma–Mass Spectrometry. J. AOAC Formatted: German (Germany)
Int. 92: 1720-1727.
32
52 Cubadda, F., Raggi, A., Testoni, A., & Zanasi, F. (2002) Multielemental Analysis of Food and
Agricultural Matrixes by Inductively Coupled Plasma-Mass Spectrometry. J. AOAC. Int. 85: 113- Formatted: German (Germany)
121.
53 Chan, K.C., Yip, Y.C., Chu, H.S. & Sham, W.C. (2006). High-throughput determination of seven Formatted: German (Germany)
trace elements in food samples by inductively coupled plasma – mass spectrometry. J. AOAC Int.
Formatted: German (Germany)
89 (2): 469-479.
54 Harrington, C.F., Elahi, S., Merson, S.A., & Ponnampalavanar, P. (2004) Quantitative Analysis of
Iron-Containing Protein Myoglobin in Different Foodstuffs by Liquid Chromatography Coupled to
High-Resolution Inductively Coupled Plasma Mass Spectrometry. J.AOAC. Int. 87: 253-258.
55 Ashoka, S. Peake, B.M., Bremner, G., Hageman, K.J., & Reid, M.R. (2009) Comparison of
digestion methods for ICP-MS determination of trace elements in fish tissues. Anal. Chim. Acta
653: 191–199.
56 Giannenas, I., Nisianakis, P., Gavriil, A., Kontopidis, G., & Kyriazakis, I. (2009) Trace mineral
content of conventional, organic and courtyard eggs analysed by inductively coupled plasma mass
spectrometry (ICP-MS). Food Chem. 114: 706–711. Formatted: German (Germany)
57 Batista, B.L., Grotto, D., Rodrigues, J.L., de Oliveira Souza, V.C., & Barbosa Jr., F. (2009)
Determination of trace elements in biological samples by inductively coupled plasma mass
spectrometry with tetramethylammonium hydroxide solubilization at room temperature. Anal. Chim.
Acta 646: 23–29.
58 Miksa, I.R., Buckley, C.L., Carpenter, N.P., & Poppenger, R.H. (2005). Comparison of selenium
determination in liver samples by atomic absorption spectroscopy and inductively coupled plasma –
mass spectrometry. J. Vet. Diagn. Invest. 17: 331-340. Formatted: German (Germany)
59 Falandycz, J., Szymczyk-Kobrzynska, K., Brzostowski, A., Zalewski, K., & Zasadowski, A. (2005). Formatted: German (Germany)
Concentrations of heavy metals in the tissues of red deer (Cervus elaphus) from the region of
Warmia and Mazury, Poland. Food Add. Contam. 22 (2): 141-149.
60 Stürup, S., & Büchert, A. (1996) Direct determination of copper and iodine in milk and milk powder
in alkaline solution by flow injection inductively coupled plasma mass spectrometry. Fresenius J. Formatted: German (Germany)
Anal. Chem. 354: 323-326.
61 Güler, C., & Alpaslan, M. (2009) Mineral content of 70 bottled water brands sold on the Turkish
market: Assessment of their compliance with current regulations. J. Food Comp. & Anal. 22: 728–
737.
62 Krachler, M. & Shotyk, W. (2009) Trace and ultratrace metals in bottled waters: Survey of sources
worldwide and comparison with refillable metal bottles. Sci. Tot. Envir. 407: 1089-1096.
63 Coelhoa, N.M.M., Coelho, L.M., de Lima, E.S., Pastor, A. , & de la Guardia, M. (2005)
Determination of arsenic compounds in beverages by high-performance liquid chromatography-
inductively coupled plasma mass spectrometry. Talanta 66: 818–822. Formatted: German (Germany)
64 Nisianakis, P., Giannenas, I., Gavriil, A., Kontopidis, G., & Kyriazakis, I. (2009) Variation in Trace
Element Contents Among Chicken, Turkey, Duck, Goose, and Pigeon Eggs Analyzed by
Inductively Coupled Plasma Mass Spectrometry (ICP-MS). Biol. Trace Element Res. 128: 62-71. Formatted: German (Germany)
33
65 Benincasa, C., Lewis, J., Perri, E., Sindona, G., & Tagarelli, A. (2007). Determination of trace
elements in Italian virgin olive oils and their characterization according to geographical origins by
statistical analysis. Anal. Chim. Acta 585: 366-370.
66 Krachler, M., Prohaska, T., Koellensperger, Rossipal, G.E., & Stingeder, G. (2000) Concentrations
of Selected Trace Elements in Human Milk and in Infant Formulas Determined by Magnetic Sector
Field Inductively Coupled Plasma–Mass Spectrometry Biol. Trace Element Res. 76: 97-112. Formatted: German (Germany)
Formatted: German (Germany)
67 Juhlshamn, K., Maage, A., Skarr Norli, H., Grobecker, K.H., Jorhem, L., & Fecher, P. (2007).
Determination of arsenic, cadmium, mercury and lead by inductively couple plasma/mass
spectrometry in foods after pressure digestion: NKML collaborative study. J. AOAC Int. 90 (3): 844-
856.
68 Catarino, S., Curvelo-Garcia, A.S., Sousa, R.B. (2006) Measurements of contaminant elements of
wines by inductively coupled plasma-mass spectrometry: A comparison of two calibration
approaches. Talanta 70:1073-1080. Formatted: German (Germany)
69 Melaku, S., Dams, R., & Moens, L. (2005) Determination of trace elements in agricultural soil
samples by inductively coupled plasma-mass spectrometry: Microwave acid digestion versus aqua
regia extraction. Anal. Chim. Acta 543: 117–123.
70 Xie, H., Huang, K., Liu, J., Nie, X. & Fu, L. (2009) Determination of trace elements in residual oil by
high-resolution inductively coupled plasma mass spectrometry. Anal. Bioanal. Chem. 393:2075–
2080.
71 Arslan, Z., Ertas, N., Tyson, J.F., Uden, P.C., & Denoyer, E.R. (2000) Determination of trace
elements in marine plankton by inductively coupled plasma mass spectrometry (ICP-MS) Fresenius
J. Anal. Chem. 366: 273-282.
72 Lee, K.H., Oshima, M., & Motomizu, S. (2002) Inductively coupled plasma mass spectrometric
determination of heavy metals in sea-water samples after pre-treatment with a chelating resin disk
by an on-line flow injection method. Analyst, 127: 769–774.
73 Rodushkin, I., Engström, E., & Baxter, D.C. (2010) Sources of contamination and remedial
strategies in the multi-elemental trace analysis laboratory. Anal. Bioanal. Chem. 396:365–377.
74 Gramowska, H., Krzyzaniak, I., Baralkiewicz, D., & Goldyn, R. (2010) Environmental applications of
ICP-MS for simultaneous determination of trace elements and statistical data analysis. Environ.
Monit. Assess. 160:479–490.
75 Hein, A., Tsolakidou, A., Iliopoulos,I., Mommsen, H., Buxeda i Garrigós, J., Montanac, G., &
Kilikoglou, V. (2002) Standardisation of elemental analytical techniques applied to provenance
studies of archaeological ceramics: an interlaboratory calibration study. Analyst 127, 542–553.
76 Waddell, R.J.H., NicDaéid, N., & Littlejohn, D. (2004) Classification of ecstasy tablets using trace
metal analysis with the application of chemometric procedures and artificial neural network
algorithms. Analyst 129: 235-240.
77 Mudiam, M.R., Kumar, S.A., Mahadevan, S., Ghosh, P., Sarin, R.K., & Beedu, S.R. (2005)
Quantitative Evaluation of 28 Mineral Elements by Inductively Coupled Plasma/Mass Spectrometry
and Its Application in Source Identification of Indian Opium. J.AOAC. Int. 88: 1469-1484.
78 D’Ilio, S., Violante, N., Di Gregorio, M., Senofonte, O., & Petrucci, F. (2006) Simultaneous
quantification of 17 trace elements in blood by dynamic reaction cell inductively coupled plasma
34
mass spectrometry (DRC-ICP-MS) equipped with a high-efficiency sample introduction system.
Anal. Chim.Acta 579: 202–208.
79 Sarmiento-González, A., Marchante-Gayón, J.M., Tejerina-Lobo, J.M., Paz-Jiménez, J. & Sanz-
Medel, A. (2008) High-resolution ICP–MS determination of Ti, V, Cr, Co, Ni, and Mo in human
blood and urine of patients implanted with a hip or knee prosthesis. Anal. Bioanal. Chem. 391:
2583–2589.
80 Gellein, K., Lierhagen, S., Brevik, P.S., Teigen, M., Kaur, P., Singh, T., Flaten, T.P., & Syversen, T.
(2008) Trace Element Profiles in Single Strands of Human Hair Determined by HR-ICP-MS. Biol.
Trace Elem. Res. 123: 250-260.
81 Noël, L., Guérin, T., Frémy, J.-M., Huet, H., & Kolf-Clauw, M. (2003) Optimized Simultaneous
Determination of Several Elements in Human Intestinal Caco-2 TC7 Cells by Inductively Coupled
Plasma-Mass Spectrometry after Closed VesselMicrowave Digestion. J.AOAC. Int. 86: 1225-1231.
82 Krachler, M., Heisel, C., & Kretzer, J.P. (2009) Validation of ultratrace analysis of Co, Cr, Mo and Ni
in whole blood, serum and urine using ICP-SMS. J. Anal. At. Spectrom. 24: 605-610.
83 Zeiner M., Ovari, M., Zaray, G., & Steffan, I. (2009) A study of the exposure of various population
groups to platinum in Vienna (Austria) through urine analysis by inductively coupled plasma-mass
spectrometry. Microchem. J. 93: 22–24.
84 Reyes, L.H., Mar, J.L.G., Rahman, G.M.M., Seybert, B., Fahrenholz, T., Kingston, H.M.S. (2009)
Simultaneous determination of arsenic and selenium species in fish tissues using microwave-assisted
enzymatic extraction and ion chromatography-inductively coupled plasma mass spectrometry. Talanta 78:
983-990.
85 Melnyk, L.J., Morgan, J.N., Fernando, R., Pellizzari, E.D., & Akinbo, O. (2003). Determination of
metals in composite diet samples by inductively coupled plasma – mass spectrometry. J. AOAC
Int. 86 (2): 439-448.
86 D’Ilio, S., Petrucci, F., D’Amato, M., Di Gregorio, M., Senofonte, O., & Violante, N. (2008) Method
validation for determination of arsenic, cadmium, chromium and lead in milk by means of dynamic
reaction cell inductively coupled plasma mass spectrometry. Anal. Chim Acta 624: 59–67.
87 Juhlshamn, K., Dahl, L., & Eckhoff, K. (2001). Determination of iodine in seafood by inductively Formatted: German (Germany)
coupled plasma/mass spectrometry. J. AOAC Int. 84 (6): 1976-1983.
88 Chen, F-Y, & Jiang, S-J. (2009) Slurry Sampling Flow Injection Chemical Vapor Generation
Inductively Coupled Plasma Mass Spectrometry for the Determination of As, Cd, and Hg in
Cereals. J. Agric. Food Chem. 57: 6564–6569.
89 Vela, N.P., & Heitkemper, D.T. (2004) Total Arsenic Determination and Speciation in Infant Food
Products by Ion Chromatography-Inductively Coupled Plasma-Mass Spectrometry. J. AOAC Int.
87: 244-252.
90 Frazzoli, C., & Bocca, B. (2008) Validation, uncertainty estimation and application of a sector field
ICP MS-based method for As, Cd and Pb in cow’s milk and infant formulas. Microchim. Acta 162:
43–50.
91 Jorhem, L., (strand, C. Sundstr(m, B., Baxter, M., Stokes, P., Lewis, J., & Petersson Graw(, K. Formatted: Font: (Default) Arial
(2008) Elements in rice from the Swedish market: 1. Cadmium, lead and arsenic (total and
inorganic). Food Add. Contam. 25: 284–292.
35
92 Caroli, S., Forte, G., Iamiceli, A.L., & Galoppi, B. (1999) Determination of essential and potentially
toxic trace elements in honey by inductively coupled plasma-based techniques. Talanta 50: 327-
336.
93 Fingerová, H., & Koplík, R. (1999) Study of minerals and trace element species in soybean flour.
Fresenius J. Anal. Chem. 363: 545-549.
94 Leiterer, M., Einax, J.W., Löser, C., & Vetter , A. (1997) Trace analysis of metals in plant samples
with inductively coupled plasma – mass spectrometry. Fresenius J. Anal. Chem. 359: 423-426.
95 Hsieh, H.-F, Chang, W.-S., Hsieh, Y.-K., Wang C.-F. (2009) Lead determination in whole blood by
laser ablation coupled with inductively coupled plasma mass spectrometry. Talanta 79:183–188.
96 Thompson, M., Ellison, S.L.R., Fajgelj, A., Willetts, P. & Wood, R. (1999) Harmonised guidelines for
the use of recovery information in analytical measurement. Pure & Appl. Chem., 71: 337–348.
97 Ma, H.-L. & Tanner, P.A. (2008) Speciated isotope dilution analysis of Cr(III) and Cr(VI) in water by Formatted: Italian (Italy)
ICP-DRC-MS. Talanta, 77: 189-194.
98 Pettine, M., Capri, S. (2005) Digestion treatments and risks of Cr(III)-Cr(VI) interconversions during
Cr(VI) determination in soils and sediments - A review. Anal. Chim. Acta, 540: 231-238.
99 Stöckl, D., D’Hondt, H., & Thienpont, L.M. (2009) Method validation across the disciplines—Critical Formatted: Font: Arial, 10 pt
investigation of major validation criteria and associated experimental protocols. J. Chromatogr. B,
Formatted: Font: Arial, 10 pt, French
877: 2180–2190. (Canada)
100 Noel, L., Dufailly, V., Lemaiheau, N., Vastel, C. & Guerin, T. (2005). Simultaneous analysis of
cadmium, lead, mercury and arsenic in foodstuffs of animal origin by inductively coupled
plasma/mass spectrometry after closed vessel microwave digestion: validation study. J. AOAC Int.
88 (6): 1811-1821.
101 Forte, G., & Bocca, B. (2007). Quantification of cadmium and lead in offal by SF-ICP-MS: method
development and uncertainty estimate. Food Chem., 105: 1591-1598.
102 Hu, X., Wang, F., & Hanson, M.L. (2009) Selenium concentration, speciation and behavior in
surface waters of the Canadian prairies. Sci.Tot. Environ. 407: 5869–5876.
103 Miller J.C., Miller J.N., 1988. Statistics for Analytical Chemistry, 2nd Edition, New York, Ellis Formatted: English (Canada)
Horwood Limited.
104 Martena, M.J. , Van Der Wielen, J.C.A., Rietjens, I.M.C.M., Klerx, W.N.M., De Groot, H.N., &
Konings, E.J.M. (2010) Monitoring of mercury, arsenic, and lead in traditional Asian herbal
preparations on the Dutch market and estimation of associated risks. Food Add. Contam. 27: 190–
205.
105 AOAC International (2002) AOAC Guidelines for Single Laboratory Validation of Chemical Methods
for Dietary Supplements and Botanicals. AOAC International, Gaithersburg, MD.
http://www.aoac.org/Official_Methods/slv_guidelines.pdf; accessed February 18, 2010
106 Perell(, G., Mart(-Cid, R., Llobet, J.M., Domingo, J.E.L. (2008) Effects of Various Cooking Processes on
the Concentrations of Arsenic, Cadmium, Mercury, and Lead in Foods. J. Agric. Food Chem. 56: 11262–
11269.
Formatted: English (Canada)
36
107 Melaku, S., Dams, R., & Moens, L. (2005) Determination of trace elements in agricultural soil
samples by inductively coupled plasma-mass spectrometry: Microwave acid digestion versus aqua
regia extraction. Anal. Chim. Acta 543: 117–123.
108 Standards Council of Canada. 2005. CAN-P-4E (ISO/IEC 17025:2005) General Requirements for
the Competence of Testing and Calibration Laboratories.
109 Ellison, S.L.R., Rosslein, M., & Williams, A., ed. (2000). Quantifying Uncertainty in Analytical
Measurement, Second Edition. EURACHEM / CITAC Guide CG4;
http://www.eurachem.org/guides/QUAM2000-1.pdf; accessed March 27, 2009.
Appendix I: Definitions for analytical terms used in method validation recommendations contained in this
document. Unless otherwise referenced, the definitions are taken from the Guidelines on Analytical
Terminology (CAC/GL 72-2009), Codex Alimentarius Commission, Joint FAO/WHO Food Standards
Program; http://www.codexalimentarius.net/download/standards/11357/cxg_072e.pdf; accessed January
28, 2010.
References included with each definition are to the original sources of the definitions included in the
Codex guideline, CAC/GL 72-2009.
Accuracy: The closeness of agreement between a test result or measurement result and a reference
value.
Notes:
37
The term “accuracy,” when applied to a set of test results or measurement results, involves a
combination of random components and a common systematic error or bias component.
When applied to a test method, the term accuracy refers to a combination of trueness and
precision.
Reference:
ISO Standard 3534-2: Vocabulary and Symbols Part 2: Applied Statistics, ISO, Geneva, 2006
Analyte: The chemical substance sought or determined in a sample.
Note:
This definition does not apply to molecular biological analytical methods.
Reference:
Codex Guidelines on Good Laboratory Practice in Residue Analysis (CAC/GL 40-1993)
Applicability: The analytes, matrices, and concentrations for which a method of analysis may be used
satisfactorily.
Note:
In addition to a statement of the range of capability of satisfactory performance for each factor, the
statement of applicability (scope) may also include warnings as to known interference by other
analytes, or inapplicability to certain matrices and situations.
Reference:
th
Codex Alimentarius Commission, Procedural Manual, 17 Edition, 2007
Bias: “The difference between the expectation of the test result or measurement result and the true value.
In practice conventional quantity value (VIM, 2007) can be substituted for true value.
Notes:
Bias is the total systematic error as contrasted to random error. There may be one or more
systematic error components contributing to bias. A larger systematic difference from the accepted
reference value is reflected by a larger bias value.
The bias of a measuring instrument is normally estimated by averaging the error of indication over
the appropriate number of repeated measurements. The error of indication is the: “indication of a
measuring instrument minus a true value of the corresponding input quantity”.
Expectation is the expected value of a random variable, e.g. assigned value or long term average
{ISO 5725-1}
Reference:
ISO Standard 3534-2: Vocabulary and Symbols Part 2: Applied Statistics, ISO, Geneva, 2006
Calibration: Operation that, under specified conditions, in a first step, establishes a relation between the
values with measurement uncertainties provided by measurement standards and corresponding
indications with associated measurement uncertainties and in a second step uses this information
to establish a relation for obtaining a measurement result from an indication.
Notes:
A calibration may be expressed by a statement, calibration function, calibration diagram, calibration
curve, or calibration table. In some cases it may consist of an additive or multiplicative correction of
the indication with associated measurement uncertainty.
Calibration should not be confused with adjustment of a measuring system often mistakenly called
“self calibration,” or with verification of calibration.
Often the first step alone in the above definition is perceived as being calibration.
38
Reference:
rd
VIM, International Vocabulary of Metrology – Basic and general concepts and associated terms, 3
edition, JCGM 200: 2008.
Certified reference material (CRM): Reference material accompanied by documentation issued by an
authoritative body and providing one or more specified property values with associated uncertainties and
traceability, using valid procedures.
Notes:
Documentation is given in the form of a “certificate” (see ISO guide 30:1992).
Procedures for the production and certification of certified reference materials are given, e.g. in ISO
Guide 34 and ISO Guide 35.
In this definition, “uncertainty” covers both measurement uncertainty and uncertainty associated
with the value of the nominal property, such as for identity and sequence. Traceability covers both
metrological traceability of a value and traceability of a nominal property value.
Specified values of certified reference materials require metrological traceability with associated
measurement uncertainty {Accred. Qual. Assur., 2006}
ISO/REMCO has an analogous definition {Accred. Qual. Assur., 2006} but uses the modifiers
metrological and metrologically to refer to both quantity and nominal properties.
Reference:
rd
VIM, International Vocabulary of Metrology – Basic and general concepts and associated terms, 3
edition, JCGM 200: 2008
New definitions on reference materials, Accreditation and Quality Assurance, 10:576-578, 2006.
Conventional quantity value: quantity value attributed by agreement to a quantity for a given purpose.
Notes:
The term “conventional true quantity value” is sometimes used for this concept, but its use is
discouraged.
Sometimes a conventional quantity value is an estimate of a true quantity value.
A conventional quantity value is generally accepted as being associated with a suitably small
measurement uncertainty, which might be zero.
Reference:
rd
VIM, International Vocabulary of Metrology – Basic and general concepts and associated terms, 3
edition, JCGM 200: 2008
Critical value (LC): The value of the net concentration or amount the exceeding of which leads, for a
given error probability α, to the decision that the concentration or amount of the analyte in the
analyzed material is larger than that in the blank material. It is defined as:
Pr (>LC | L=0) ≤ α
Where is the estimated value, L is the expectation or true value and LC is the critical value.
Notes:
The definition of critical value is important for defining the Limit of Detection (LOD).
The critical value Lc is estimated by
LC = t1-ανso,
Where t1-αν is Student's-t, based on ν degrees of freedom for a one-sided confidence interval of 1-α
and so is the sample standard deviation.
If L is normally distributed with known variance, i.e. ν = ∞ with the default α of 0.05, LC = 1.645so.
A result falling below the LC triggering the decision “not detected” should not be construed as
demonstrating analyte absence. Reporting such a result as “zero” or as < LOD is not
recommended. The estimated value and its uncertainty should always be reported.
References:
ISO Standard 11843: Capability of Detection-1, ISO, Geneva, 1997
Nomenclature in evaluation of analytical methods, IUPAC, 1995
39
Error: Measured quantity value minus a reference quantity value.
Note:
The concept of measurement ‘error’ can be used both: when there is a single reference value to
refer to, which occurs if a calibration is made by means of a measurement standard with a
measured value having a negligible measurement uncertainty or if a conventional value is given, in
which case the measurement error is not known and if a measurand is supposed to be represented
by a unique true value or a set of true values of negligible range, in which case the measurement
error is not known.
Reference:
rd
VIM, International Vocabulary of Metrology – Basic and general concepts and associated terms, 3
edition, JCGM 200: 2008
Expanded measurement uncertainty: product of a combined standard measurement uncertainty and a
factor larger than the number one.
Notes:
The factor depends upon the type of probability distribution of the output quantity in a measurement
model and on the selected coverage probability.
The term factor in this definition refers to a coverage factor.
Expanded measurement uncertainty is also termed expanded uncertainty.
Reference:
rd
VIM, International Vocabulary of Metrology – Basic and general concepts and associated terms, 3
edition, JCGM 200: 2008.
Fitness for purpose: Degree to which data produced by a measurement process enables a user to
make technically and administratively correct decisions for a stated purpose.
Reference:
Eurachem Guide: The fitness for purpose of analytical methods: A laboratory guide to method
validation and related topics, 1998
HorRat: The ratio of the reproducibility relative standard deviation to that calculated from the Horwitz
equation,
-0.15
Predicted relative standard deviation (PRSD)R =2C :
HorRat(R) = RSDR/PRSDR ,
HorRat(r) = RSDr/PRSDR
Where C is concentration expressed as a mass fraction (both numerator and denominator
expressed in the same units).
Notes:
The HorRat is indicative of method performance for a large majority of methods in chemistry.
-6
Normal values lie between 0.5 and 2. (To check proper calculation of PRSDR, a C of 10 should
give a PRSD of 16 %.)
R
If applied to within-laboratory studies, the normal range of HorRat(r) is 0.3-1.3.
For concentrations less than 0.12 mg/kg the predicted relative standard deviation developed by
Thompson (The Analyst, 2000), 22% should be used.
References:
A simple method for evaluating data from an inter-laboratory study, J AOAC, 81(6):1257-1265,
1998
Recent trends in inter-laboratory precision at ppb and sub-ppb concentrations in relation to fitness
for purpose criteria in proficiency testing, The Analyst, 125:385-386, 2000
Inter-laboratory study: A study in which several laboratories measure a quantity in one or more
“identical” portions of homogeneous, stable materials under documented conditions, the results of
which are compiled into a single document.
Notes:
40
The larger the number of participating laboratories, the greater the confidence that can be placed
in the resulting estimates of the statistical parameters. The IUPAC-1987 protocol (Pure & Ap pl.
Chem., 66, 1903-1911(1994)) requires a minimum of eight laboratories for method-performance
studies.
Reference:
th
Codex Alimentarius Commission, Procedural Manual, 17 Edition, 2007.
Limit of Detection (LOD): The true net concentration or amount of the analyte in the material to be
analyzed which will lead, with probability (1-β), to the conclusion that the concentration or amount
of the analyte in the analyzed material is larger than that in the blank material. It is defined as:
Pr (≤LC | L=LOD) =
Where is the estimated value, L is the expectation or true value and LC is the critical value.
Notes:
The limit of detection LOD is estimated by,
LOD ≈ 2t1-νϭo [where = ],
Where t1-ν is Student's-t, based on ν degrees of freedom for a one-sided confidence interval of 1-
and ϭo is the standard deviation of the true value (expectation).
LOD = 3.29 ϭo, when the uncertainty in the mean (expected) value of the blank is negligible, =
= 0.05 and L is normally distributed with known constant variance. However, LOD is not defined
simply as a fixed coefficient (e.g. 3, 6, etc.) times the standard deviation of a pure solution
background. To do so can be extremely misleading. The correct estimation of LOD must take into
account degrees of freedom, and , and the distribution of L as influenced by factors such as
analyte concentration, matrix effects and interference.
This definition provides a basis for taking into account exceptions to simple case that is described,
i.e. involving non-normal distributions and heteroscedasticity (e.g. “counting” (Poisson) processes
as those used for real time PCR).
It is essential to specify the measurement process under consideration, since distributions, ϭ’s and
blanks can be dramatically different for different measurement processes.
At the limit of detection, a positive identification can be achieved with reasonable and/or previously
determined confidence in a defined matrix using a specific analytical method.
References:
ISO Standard 11843: Capability of Detection-1, ISO, Geneva, 1997
Nomenclature in evaluation of analytical methods, IUPAC, 1995
Guidance document on pesticide residue analytical methods, Organization for Economic
Cooperation and Development, 2007
Limit of Quantification (LOQ): A method performance characteristic generally expressed in terms of the
signal or measurement (true) value that will produce estimates having a specified relative standard
deviation (RSD), commonly 10% (or 6%). LOQ is estimated by:
LOQ = kQ ϭQ, kQ = 1/RSDQ
Where LOQ is the limit of quantification, ϭQ is the standard deviation at that point and kQ is the
multiplier whose reciprocal equals the selected RSD. (The approximate RSD of an estimated ϭ,
based on ν-degrees of freedom is 1/ √2ν.)
Notes:
If ϭ is known and constant, then ϭQ = ϭo, since the standard deviation of the estimated quantity is
independent of concentration. Substituting 10% in for kQ gives:
LOQ = (10 * ϭQ) = 10 ϭo
In this case, the LOQ is just 3.04 times the limit of detection, given normality and = = 0.05.
41
At the LOQ, a positive identification can be achieved with reasonable and/or previously determined
confidence in a defined matrix using a specific analytical method.
This definition provides a basis for taking into account exceptions to the simple case that is
described, i.e. involving non-normal distributions and heteroscedasticity (e.g. “counting” (Poisson)
processes as those used for real time PCR).
References:
Nomenclature in evaluation of analytical methods, IUPAC, 1995
Guidance document on pesticide residue analytical methods, Organization for Economic Co-
operation and Development, 2007
Linearity: The ability of a method of analysis, within a certain range, to provide an instrumental response
or results proportional to the quantity of analyte to be determined in the laboratory sample. This
proportionality is expressed by an a priori defined mathematical expression. The linearity limits are
the experimental limits of concentrations between which a linear calibration model can be applied
with an acceptable uncertainty.
Reference:
th
Codex Alimentarius Commission, Procedural Manual, 17 Edition, 2007
Measurand: Quantity intended to be measured.
Notes:
The specification of a measurand requires knowledge of the kind of quantity, description of the
state of the substance carrying the quantity, including any relevant component and the chemical
entities involved.
In chemistry, ‘analyte’ or the name of a substance or compound are terms sometime used for
measurand. This usage is erroneous because these terms do not refer to quantities.
Reference:
rd
VIM, International Vocabulary of Metrology – Basic and general concepts and associated terms, 3
edition, JCGM 200: 2008
Measurement method: Generic description of a logical organization of operations used in a
measurement.
Note:
Measurement methods may be qualified in various ways such as: substitution measurement
method, differential measurement method, and null measurement method; or direct measurement
method, and indirect measurement method.
Reference:
rd
VIM, International Vocabulary of Metrology – Basic and general concepts and associated terms, 3
edition, JCGM 200: 2008
Measurement procedure: Detailed description of a measurement according to one or more
measurement principles and to a given measurement method, based on a measurement model
and including any calculation to obtain a result.
Notes:
A measurement procedure is usually documented in sufficient detail to enable an operator to
perform a measurement.
A measurement procedure can include a statement concerning a target measurement uncertainty.
A measurement procedure is sometimes called a standard operating procedure (SOP).
Reference:
rd
VIM, International Vocabulary of Metrology – Basic and general concepts and associated terms, 3
edition, JCGM 200: 2008
Measurement uncertainty: Non-negative parameter characterizing the dispersion of the values being
attributed to a measurand, based on the information used.
Notes:
42
Measurement uncertainty includes components arising from systematic effects, such as
components associated with corrections and the assigned values of measurement standards, as
well as the definitional uncertainty. Sometimes estimated systematic effects are not corrected for
but, instead associated measurement uncertainty components are incorporated.
The parameter may be, for example, a standard deviation called standard measurement
uncertainty (or a given multiple of it), or the half-width of interval having a stated coverage
probability.
Measurement uncertainty comprises, in general many components. Some of these components
may be evaluated by Type A evaluation of measurement uncertainty from the statistical distribution
of the values from a series of measurements and can be characterized by experimental standard
deviations. The other components which may be evaluated by Type B evaluation of measurement
uncertainty can also be characterized by standard deviations, evaluated from assumed probability
distributions based on experience or other information.
In general, for a given set of information, it is understood that the measurement uncertainty is
associated with a stated quality value attributed to the measurand. A modification of this value
results in a modification of the associated uncertainty.
Reference:
rd
VIM, International Vocabulary of Metrology – Basic and general concepts and associated terms, 3
edition, JCGM 200: 2008
Method-Performance Study: An inter-laboratory study in which all laboratories follow the same written
protocol and use the same test method to measure a quantity in sets of identical test samples. The
reported results are used to estimate the performance characteristics of the method. Usually these
characteristics are within-laboratory and among-laboratories precision, and when necessary and
possible, other pertinent characteristics such as systematic error, recovery, internal quality control
parameters, sensitivity, limit of quantification, and applicability.
Notes:
The materials used in such a study of analytical quantities are usually representative of materials to
be analyzed in actual practice with respect to matrices, amount of test component (concentration),
and interfering components and effects. Usually the analyst is not aware of the actual composition
of the test samples but is aware of the matrix.
The number of laboratories, number of test samples, number of determinations, and other details of
the study are specified in the study protocol. Part of the study protocol is the procedure which
provides the written directions for performing the analysis.
The main distinguishing feature of this type of study is the necessity to follow the same written
protocol and test method exactly.
Several methods may be compared using the same test materials. If all laboratories use the same
set of directions for each method and if the statistical analysis is conducted separately for each
method, the study is a set of method-performance studies. Such a study may also be designated
as a method-comparison study.
Reference:
th
Codex Alimentarius Commission, Procedural Manual, 17 Edition, 2007
Precision: The closeness of agreement between independent test/measurement results obtained under
stipulated conditions.
Notes:
Precision depends only on the distribution of random errors and does not relate to the true value or
to the specified value.
The measure of precision is usually expressed in terms of imprecision and computed as a standard
deviation of the test results. Less precision is reflected by a larger standard deviation.
Quantitative measures of precision depend critically on the stipulated conditions. Repeatability and
reproducibility conditions are particular sets of extreme conditions.
Intermediate conditions between these two extreme conditions are also conceivable, when one or
more factors within a laboratory (intra-laboratory e.g. the operator, the equipment used, the
43
calibration of the equipment used, the environment, the batch of reagent and the elapsed time
between measurements) are allowed to vary and are useful in specified circumstances.
Precision is normally expressed in terms of standard deviation.
Reference:
ISO Standard 3534-2: Vocabulary and Symbols Part 2: Applied Statistics, ISO, Geneva, 2006
ISO Standard 5725-3: Accuracy (trueness and precision) of measurement methods and results
Part 3: Intermediate measures of the precision of a standard measurement method, ISO, Geneva,
1994
Recovery/recovery factors: Proportion of the amount of analyte, present in, added to or present in and
added to the analytical portion of the test material, which is presented for measurement.
Notes:
Recovery is assessed by the ratio R = Cobs / Cref of the observed concentration or amount Cobs
obtained by the application of an analytical procedure to a material containing analyte at a
reference level Cref .
Cref will be: (a) a reference material certified value, (b) measured by an alternative definitive
method, (c) defined by a spike addition or (d) marginal recovery.
Recovery is primarily intended for use in methods that rely on transferring the analyte from a
complex matrix into a simpler solution, during which loss of analyte can be anticipated.
Reference:
Harmonized guidelines for the use of recovery information in analytical measurement, 1998
Use of the terms “recovery” and “apparent recovery” in analytical procedures, 2002
Reference material: Material, sufficiently homogeneous and stable with respect to one or more specified
properties, which has been established to be fit for its intended use in a measurement process or in
examination of nominal properties.
Notes:
Examination of a nominal property provides a nominal property value and associated uncertainty.
This uncertainty is not a measurement uncertainty.
Reference materials with or without assigned values can be used for measurement precision
control whereas only reference materials with assigned values can be used for calibration and
measurement trueness control.
Some reference materials have assigned values that are metrologically traceable to a
measurement unit outside a system of units. In a given measurement, a given reference material
can only be used for either calibration or quality assurance.
The specification of a reference material should include its material traceability, indicating its origin
and processing. {Accred. Qual. Assur., 2006}
ISO/REMCO has an analogous definition that uses the term measurement process to mean
examination which covers both measurement of a quantity and examination of a nominal property.
References:
rd
VIM, International Vocabulary of Metrology – Basic and general concepts and associated terms, 3
edition, JCGM 200: 2008
New definitions on reference materials, Accred. Qual. Assur., 10:576-578, 2006
Reference value: Quantity value used as a basis of comparison with values of quantity of the same kind.
Notes:
A reference quantity value can be a true quantity value of a measurand, in which case it is
unknown, or a conventional quantity value in which case it is known.
A reference quantity value with an associated measurement uncertainty is usually provided with
reference to
a) a material, e.g. a certified reference material
b) a reference measurement procedure
c) a comparison of measurement standards.
Reference:
44
rd
VIM, International Vocabulary of Metrology – Basic and general concepts and associated terms, 3
edition, JCGM 200: 2008
Repeatability (Reproducibility): Precision under repeatability (reproducibility) conditions.
Reference:
ISO 3534-1 Statistics, vocabulary and symbols-Part 1: Probability and general statistical terms,
ISO, 1993
ISO Standard 78-2: Chemistry – Layouts for Standards – Part 2: Methods of Chemical Analysis,
1999)
th
Codex Alimentarius Commission, Procedural Manual, 17 Edition, 2007
AOAC International methods committee guidelines for validation of qualitative and quantitative food
microbiological official methods of analysis, 2002.
Repeatability conditions: Observation conditions where independent test/measurement results are
obtained with the same method on identical test/measurement items in the same test or measuring
facility by the same operator using the same equipment within short intervals of time.
Note:
Repeatability conditions include: the same measurement procedure or test procedure; the same
operator; the same measuring or test equipment used under the same conditions; the same
location and repetition over a short period of time.
Reference:
ISO Standard 3534-2: Vocabulary and Symbols Part 2: Applied Statistics, ISO, Geneva, 2006
Repeatability (Reproducibility) limit: The value less than or equal to which the absolute difference
between final values, each of them representing a series of test results or measurement results
obtained under repeatability (reproducibility) conditions may be expected to be with a probability of
95%.
Notes:
The symbol used is r [R]. {ISO 3534-2}
When examining two single test results obtained under repeatability (reproducibility) conditions, the
comparison should be made with the repeatability (reproducibility) limit, r [R] = 2.8ϭr[R]. {ISO 5725-
6, 4.1.4}
When groups of measurements are used as the basis for the calculation of the repeatability
(reproducibility) limits (now called the critical difference), more complicated formulae are required
that are given in ISO 5725-6: 1994, 4.2.1 and 4.2.2.
Reference:
ISO Standard 3534-2: Vocabulary and Symbols Part 2: Applied Statistics, ISO, Geneva, 2006
ISO 5725-6 “Accuracy (trueness and precision) of a measurement methods and results—Part 6:
Use in practice of accuracy value”, ISO, 1994
th
Codex Alimentarius Commission, Procedural Manual, 17 Edition, 2007
Repeatability (reproducibility) standard deviation: Standard deviation of test results or measurement
results obtained under repeatability (reproducibility) conditions.
Notes:
It is a measure of the dispersion of the distribution of the test or measurement results under
repeatability (reproducibility) conditions.
Reference:
ISO Standard 3534-2: Vocabulary and Symbols Part 2: Applied Statistics, ISO, Geneva, 2006
Repeatability (reproducibility) relative standard deviation (coefficient of variation):
Repeatability (reproducibility) standard deviation divided by the mean.
RSDr[R] is computed by dividing the repeatability (reproducibility) standard deviation by the mean.
Notes:
Relative standard deviation (RSD) is a useful measure of precision in quantitative studies.
45
This is done so that one can compare variability of sets with different means. RSD values are
independent of the amount of analyte over a reasonable range and facilitate comparison of
variabilities at different concentrations.
The result of a collaborative test may be summarized by giving the RSD for repeatability (RSDr)
and RSD for reproducibility (RSDR).
The RSD is also known as coefficient variation.
Reference:
ISO Standard 3534-2: Vocabulary and Symbols Part 1: General statistical terms used in probability,
ISO, Geneva, 2006
AOAC International methods committee guidelines for validation of qualitative and quantitative food
microbiological official methods of analysis, 2002.
Reproducibility conditions: Observation conditions where independent test/measurement results are
obtained with the same method on identical test/measurement items in different test or
measurement facilities with different operators using different equipment.
Reference:
ISO Standard 3534-2: Vocabulary and Symbols Part 2: Applied Statistics, ISO, Geneva, 2006
Result: Set of values being attributed to a measurand together with any other available relevant
information
Notes:
A result of measurement generally contains ‘relevant information’ about the set of values, such that
some may be more representative of the measurand than others. This may be expressed in the
form of a probability density function.
A result of measurement is generally expressed as a single measured value and a measurement
uncertainty. If the measurement uncertainty is considered to be negligible for some purpose, the
measurement result may be expressed as a single measured value. In many fields, this is the
common way of expressing a measurement result.
In the traditional literature and in the previous edition of the VIM, result was defined as a value
attributed to a measurand and explained to mean an indication or an uncorrected result or a
corrected result according to the context.
Reference:
rd
VIM, International Vocabulary of Metrology – Basic and general concepts and associated terms, 3
edition, JCGM 200: 2008
Robustness (ruggedness): A measure of the capacity of an analytical procedure to remain unaffected
by small but deliberate variations in method parameters and provides an indication of its reliability
during normal usage
Reference:
ICH Topic Q2 Validation of Analytical Methods, the European Agency for the Evaluation of
Medicinal Products: ICH Topic Q 2 A - Definitions and Terminology (CPMP/ICH/381/95), 1995
Harmonized guidelines for single laboratory validation of methods of analysis, Pure and Appl.
Chem., 2002
Selectivity: Selectivity is the extent to which a method can determine particular analyte(s) in a mixture(s)
or matrice(s) without interferences from other components of similar behaviour.
Note:
Selectivity is the recommended term in analytical chemistry to express the extent to which a
particular method can determine analyte(s) in the presence other components. Selectivity can be
graded. The use of the term specificity for the same concept is to be discouraged as this often
leads to confusion.
Reference:
Selectivity in analytical chemistry, IUPAC, Pure Appl Chem, 2001
Codex Alimentarius Commission, Alinorm 04/27/23, 2004
th
Codex Alimentarius Commission, Procedural Manual, 17 Edition, 2007
46
Sensitivity: Quotient of the change in the indication of a measuring system and the corresponding
change in the value of the quantity being measured.
Notes:
The sensitivity can depend on the value of the quantity being measured
The change considered in the value of the quantity being measured must be large compared with
the resolution of the measurement system.
Reference:
rd
VIM, International Vocabulary of Metrology – Basic and general concepts and associated terms, 3
edition, JCGM 200: 2008
Surrogate: Pure compound or element added to the test material, the chemical and physical behaviour of
which is taken to be representative of the native analyte.
Reference:
Harmonized guidelines for the use of recovery information in analytical measurement, 1998
Systematic error: Component of measurement error that in replicate measurements remains constant or
varies in a predictable manner.
Notes:
A reference value for a systematic error is a true quantity value, or a measured value of a
measurement standard of negligible measurement uncertainty, or a conventional value.
Systematic error and its causes can be known or unknown. A correction can be applied to
compensate for a known systematic error.
Systematic error equals measurement error minus random measurement error.
Reference:
rd
VIM, International Vocabulary of Metrology – Basic and general concepts and associated terms, 3
edition, JCGM 200: 2008
Trueness: The closeness of agreement between the average of an infinite number of replicate measured
quantity values and a reference quantity value.
Note 1: Measurement trueness is not a quantity and thus cannot be expressed numerically, but measures
for closeness of agreement are given in ISO 5725.
Note 2: Measurement trueness is inversely related to systematic measurement error, but is not related to
random measurement error.
Note 3: Measurement accuracy should not be used for 'measurement trueness' and vice versa.
Reference:
rd
VIM, International Vocabulary of Metrology – Basic and general concepts and associated terms, 3
edition, JCGM 200: 2008
True value: Quantity value consistent with the definition of a quantity.
Notes:
In the error approach to describing measurement, a true quantity value is considered unique and in
practice unknowable. The uncertainty approach is to recognize that, owing to the inherently
incomplete amount of detail in the definition of quantity, there is not a single true quantity value, but
rather a set of quantity values consistent with the definition of a quantity. However, this set of
values is, in principle and in practice unknowable. Other approaches dispense altogether with the
concept of true quantity value and rely on the concept of metrological compatibility of measurement
results for assessing their validity.
When the definitional uncertainty associated with the measurand is considered to be negligible
compared to the other components of the measurement uncertainty the measurand may be
considered to have an essentially “unique” true value.
Reference:
rd
VIM, International Vocabulary of Metrology – Basic and general concepts and associated terms, 3
edition, JCGM 200: 2008
Validation: Verification, where the specified requirements are adequate for an intended use.
47
Reference:
rd
VIM, International Vocabulary of Metrology – Basic and general concepts and associated terms, 3
edition, JCGM 200: 2008
Validated Test Method: An accepted test method for which validation studies have been completed to
determine the accuracy and reliability of this method for a specific purpose.
Reference:
ICCVAM Guidelines for the nomination and submission of new, revised and alternative test
methods, 2003
Validated range: That part of the concentration range of an analytical method which has been subjected
to validation.
Reference:
Harmonized guidelines for single-laboratory validation of methods of analysis, 2002
Verification: Provision of objective evidence that a given item fulfils specified requirements.
Notes:
When applicable method uncertainty should be taken into consideration.
The item may be e.g. a process, measuring procedure, material, compound or measuring system.
The specified requirement may be that a manufacturer’s specifications are met.
Verification in legal metrology, as defined in VIM and in conformity assessment in general pertains
to the examination and marketing and/or issuing of a verification certificate for a measuring system.
Verification should not be confused with calibration. Not every verification is a validation.
In chemistry, verification of the identity of the entity involved or of the activity, requires a description
of the structure and properties of that entity or activity.
References:
rd
VIM, International Vocabulary of Metrology – Basic and general concepts and associated terms, 3
edition, JCGM 200: 2008
B. Definitions from other sources (referenced in the main body of this document).
Accepted Limit (AL): Concentration value for an analyte corresponding to a regulatory limit or guideline
value which forms the purpose for the analysis, e.g. MRL, MPLmaximum permissable level; trading
standard, target concentration limit (dietary exposure assessment), acceptance level (environment), etc.
For a substance without an MRL or for a banned substance there may be no AL (effectively it may be
zero or there may be no limit) or it may be the target concentration above which detected residues levels
Error! Bookmark not defined.Error! Bookmark not defined.Error!
should be confirmed (action limit or administrative limit).
Bookmark not defined.
Calibration function: The functional (not statistical) relationship for the chemical measurement process,
relating the expected value of the observed (gross) signal or response variable to the analyte
amount.Error! Bookmark not defined.43
48
Intermediate Precision: The precision of an analytical procedure expresses the closeness of agreement
between a series of measurements obtained from multiple sampling of the same homogeneous
sample under the prescribed conditions. Intermediate precision expresses within-laboratories
Error! Bookmark not defined.Error!
variations: different days, different analysts, different equipment, etc.
Bookmark not defined.Error! Bookmark not defined.
Lowest Calibrated Level (LCL): Lowest concentration of analyte detected and measured in calibration of
the detection system. It may be expressed as a solution concentration or as a mass ratio in the test
Error! Bookmark not defined.Error! Bookmark not defined.Error!
sample and must not include the contribution from the blank.
Bookmark not defined.
Linear Range: The range of analyte concentrations over which the method provides test results
Error! Bookmark not defined.43
proportional to the concentration of the analyte .
Error! Bookmark not defined.43
Matrix: The components of the sample other than the analyte .
Matrix Effect: The combined effect of all components in the sample other than the analyte on the
measurement of the quantity. If a specific component can be identified as causing an effect then
Error! Bookmark not defined.43
this is referred to as interference .
Matrix-matched Calibration: Calibration using standards prepared in an extract of the commodity
analysed (or of a representative commodity). The objective is to compensate for the effects of co-
extractives on the determination system. Such effects are often unpredictable, but matrix-matching
Error! Bookmark not
may be unnecessary where co-extractives prove to be of insignificant effect.
defined.Error! Bookmark not defined.Error! Bookmark not defined.
See additional comments and definitions in
Section C.
Representative Analyte: Analyte chosen to represent a group of analytes which are likely to be similar in
their behaviour through a multi-residue analytical method, as judged by their physico-chemical properties
Error! Bookmark not defined.Error!
e.g. structure, water solubility, Kow, polarity, volatility, hydrolytic stability, pKa etc.”
Bookmark not defined.Error! Bookmark not defined.
Represented Analyte: Analyte having physico-chemical properties which are within the range of
Error! Bookmark not defined.Error! Bookmark not defined.Error! Bookmark not defined.
properties of representative analytes.”
Representative Commodity: Single food or feed used to represent a commodity group for method
validation purposes. A commodity may be considered representative on the basis of proximate sample
composition, such as water, fat/oil, acid, sugar and chlorophyll contents, or biological similarities of
Error! Bookmark not defined.Error! Bookmark not defined.Error! Bookmark not defined.
tissues etc.”
49
C. Terms for which no consensus-based definitions have been issued by authoritative scientific
organizations, such as IUPAC or ISO.
Matrix Fortified Calibration Curve: Known quantities of the target analyte are added to replicate
extracts of a blank representative matrix to provide a range of concentrations of analyte in matrix
prior to extraction or digestion to generate a calibration curve. This curve is used to determine the
effect of the matrix on the response of the analyte. See also Matrix Matched.
Matrix Matched: The current consensus in literature publications is to refer to “matrix matched” when
fortified blank matrix is extracted and carried through the method to generate a calibration curve, in
contrast to the definition for “.Matrix-matched Calibration” given in Section B, taken from a
consultation held in 1999. The procedure of adding known concentrations of analyte to extracts of
blank matrix is now usually referred to as “matrix fortified”, while adding the known concentrations
of analyte to blank tissue prior to extraction is now conventionally termed “matrix matched
calibration”. “matrix fortified” standards are used to examine matrix effects, while “matrix matched”
standards and calibration are used to correct for matrix effects.
In metals testing, “matrix matched” typically refers to matching diluent concentrations of standards
to that of the sample digest. Other elements that are known to be present in sample digest may be
added as well.
A recent IUPAC project to update terminology used in mass spectrometery has not addressed
usage of the terms “matrix fortified” and “,\matrix matched”.
Surrogate matrix: When authentic blank tissue does not exist, a surrogate may be used for validation
experiments. This would consist of a closely related matrix (i.e., similar chemical composition)
which may have low or non-detected levels of the analyte(s) of interest. For biological matrices,
surrogates should have similar contents of protein, fat, carbohydrate, moisture and salt.
50
Appendix 2: Guidelines for establishing numeric values for analytical method performance
criteria, as recommended by the Codex Alimentarius Commission5:
Method Applicability The method has to be applicable for the specified Formatted: Line spacing: 1.5 lines
provision, specified commodity and the specified level(s)
(maximum and/or minimum) (ML). The minimum
applicable range of the method depends on the specified
level (ML) to be assessed, and can either be expressed in
terms of the reproducibility standard deviation (sR) or in
Formatted: Line spacing: 1.5 lines, Adjust
terms of LOD and LOQ. space between Latin and Asian text, Adjust
space between Asian text and numbers
Minimum For ML ≥ 0.1 mg/kg, [ML - 3 sR , ML + 3 sR ] Formatted: Line spacing: 1.5 lines, Adjust
space between Latin and Asian text, Adjust
applicable range For ML < 0.1 mg/kg, [ML - 2 sR , ML + 2 sR ] space between Asian text and numbers
Formatted: Line spacing: 1.5 lines
sRa = standard deviation of reproducibility
Formatted: Line spacing: 1.5 lines
Limit of For ML ≥ 0.1 mg/kg, LOD ≤ ML · 1/10 Formatted: Line spacing: 1.5 lines, Adjust
space between Latin and Asian text, Adjust
Detection (LOD) For ML < 0.1 mg/kg, LOD ≤ ML · 1/5 space between Asian text and numbers
Limit of For ML ≥ 0.1 mg/kg, LOQ ≤ ML · 1/5 Formatted: Line spacing: 1.5 lines, Adjust
space between Latin and Asian text, Adjust
space between Asian text and numbers
Quantification (LOQ) For ML < 0.1 mg/kg, LOQ ≤ ML · 2/5
Formatted: Line spacing: 1.5 lines
Precision For ML ≥ 0.1 mg/kg, HorRat value ≤ 2 Formatted: Line spacing: 1.5 lines
For ML < 0.1 mg/kg, the RSDTR < 22%. Formatted: Line spacing: 1.5 lines, Adjust
space between Latin and Asian text, Adjust
b space between Asian text and numbers
RSDR = relative standard deviation of reproducibility
Formatted: Line spacing: 1.5 lines, Adjust
Recovery (R) Concentration Ratio Unit Recovery space between Latin and Asian text, Adjust
space between Asian text and numbers
(%)
Formatted: Line spacing: 1.5 lines
100 1 100% (100 98 – 102 Formatted: Line spacing: 1.5 lines
Formatted: Line spacing: 1.5 lines
g/100g)
Formatted: Line spacing: 1.5 lines, Adjust
≥10 10-1 ≥ 10% (10 98 – 102 space between Latin and Asian text, Adjust
space between Asian text and numbers
g/100g) Formatted: Line spacing: 1.5 lines
≥1 10 -2
≥ 1% (1 g/100g) 97 - 103 Formatted: Line spacing: 1.5 lines
Formatted: Left, Line spacing: 1.5 lines
51
≥0.1 10-3 ≥ 0.1% (1 mg/g) 95 – 105
-4
0.01 10 100 mg/kg 90 – 107
-5
0.001 10 10 mg/kg 80 – 110
0.0001 10-6 1 mg/kg 80 – 110
-7
0.00001 10 100 μg/kg 80 – 110
-8
0.000001 10 10 μg/kg 60 – 115
-9
0.0000001 10 1 μg/kg 40 – 120
Other guidelines are available for expected recovery
ranges in specific areas of analysis. In cases where
recoveries have been shown to be a function of the matrix
other specified requirements may be applied.
Trueness Other guidelines are available for expected recovery Formatted: Line spacing: 1.5 lines
Formatted: Line spacing: 1.5 lines, Adjust
ranges in specific areas of analysis. space between Latin and Asian text, Adjust
space between Asian text and numbers
In cases where recoveries have been shown to be a
function of the matrix other specified requirements may be
applied.
For the evaluation of trueness preferably certified
reference material should be
used. Formatted: Line spacing: 1.5 lines
a
The sR should be calculated from the Horwitz / Thompson equation. When the Horwitz /
Thompson equation is not applicable (for an analytical purpose or according to a regulation) or
when “converting” methods into criteria then it should be based on the sR from an appropriate
method performance study.
b
The RSDR should be calculated from the Horwitz / Thompson equation. When the Horwitz / Formatted: Tab stops: Not at 0.39"
Thompson equation is not applicable (for an analytical purpose or according to a regulation) or
when “converting” methods into criteria then it should be based on the RSDR’s from an
appropriate method performance study.
Formatted: Indent: Left: 0", First line: 0",
Space After: 0 pt, Adjust space between Latin
and Asian text, Adjust space between Asian
text and numbers, Tab stops: Not at 0.63"
52
Related docs
Other docs by HC120727022149
The regulation replaces an existing regulation that deals with waste treatment and
Views: 0 | Downloads: 0
