Combining and Reporting Analytical Results
By A. Fajgelj, M. Belli, U. SansoneThe Royal Society of Chemistry
Copyright © 2007 The Royal Society of Chemistry
All rights reserved.
ISBN: 978-0-85404-848-9Contents
Environmental Metrology: the Italian Approach M. Belli, 1,
Comparability and Quality of Experimental Data Under Different Quality Systems S. Caroli, 8,
The Role of Traceability in Sustainable Development: the UNIDO Approach O. Loesener, L. M. Dhaoui, 14,
Quality Assurance of Chemical Measurements — Metrological or Management Effort M. Prošek, A. Golc-Wondra, and M. Fir, 26,
Combination of Results from Several Measurements — An Everlasting Problem W. Hässelbarth and W. Bremser, 34,
Metrological Characteristics of the Conventional Measurement Scales for Hydrogen and Oxygen Stable Isotope Amount Ratios: The δ-Scales M. Gröning, 62,
Experience with Metrological Traceability and Measurement Uncertainty in Clinical Chemistry A. Kallner, 73,
Traceability of pH in a Metrological Context G. Meinrath, M.F. Camões, P. Spitzer, H. Bühler, M. Máriássy, K. Pratt, and C. Rivier, 85,
Determination of PCBs in Organic Solutions: An Example of Traceability Chain M. Sega and E. Amico di Meane, 92,
Quality Control of pH Measurements Considering Activity and Concentration Scales: Uncertainty Budget of Primary and Secondary Apparatuses P. Fisicaro, E. Ferrara, E. Prenesti, and S. Berto, 96,
Metrology in Complex Situations: Experiences with Thermodynamic Data G. Meinrath, S. Lis, and A. Kufelnicki, 104,
Some Traceability Problems in Analytical Assays of interest in Thermal Metrology F. Pavese, 110,
Developments in Uncertainty Evaluation: The Activity of JCGM/WG1 W. Bich and F. Pennecchi, 117,
How to Combine Results Having Stated Uncertainties: To MU or Not to MU? D. L. Duewer, 27,
Uncertainty and Traceability: The View of the Analytical Chemist A. Sahuquillo and G. Rauret, 143,
Propagating Non-normally Distributed Uncertainty — The Ljungskile Code A. Ödegaard-Jensen, G. Meinrath, and Ch. Ekberg, 154,
A Systematic Approach in the Evaluation of Uncertainty in Analytical Chemistry — Application to ICP-AES Analysis P. Carconi, R. Gatti, G. Zappa, and C. Zoani, 161,
Self-Referring Procedure for the Full Cell Calibration of a Dynamic Gas Devider R. Beltramini Boveri, 168,
Investigation of Uncertainty Related to Measurement of Particulate Organic Pollutants A. Cecinato, C. Balducci, A. Di Merino Di Bucchianico, 173,
Micronucleus Test in Fish Peripheral Erythrocytes: Variability in the Microscope Scoring D. Conti, S. Barbizzi, V. Bellaria, A. Pati, S. Balzamo, M. Belli, 176,
Evaluation of intrinsic Uncertainty in the k0-NAA T. Bucar and B. Smodiš, 187,
Reporting of Uncertainty in Environmental Monitoring of Radionuclides B. Varga and S. Tarján, 195,
The Use of Reference Materials in International Refeence Measurement Systems and for Comparison of Analytical Data H. Emons, 205,
The CSM Approach to the Calculation of the Uncertainty in XRF Analysis of Low- and High-Alloyed Steels E. Celia and F. Falcioni, 216,
Comparison of Different Approaches to Evaluate Proficiency Test Data A. Shakhashiro, A. Fajgelj, and U. Sansone, 220,
Distribution of Proficiency Testing Results and Their Comparability I. Kuselman, 229,
Inter-Laboratory Comparison: the APAT Approach P. de Zorzi, S. Balzamo, S. Barbizzi, S. Gaudino, A. Pali, S. Rosamilia, M. Belli, 240,
Proficiency Testing in the Biomedical Field: from Definitions of Targets to Use of Data from End-Users M. Patriarca, I. Altieri, M. Castelli, F. Chiodo, A. Semeraro and A. Menditto, 248,
Measurement Uncertainty Its Role in Proficiency Testing Scheme — Case Study ILC Waste Water M. Cotman, A. Drolc, and M. Roš, 263,
The Proficiency Testing of Laboratories: A First Approach to Implement Bayesian Methods in the Assessment of Performance R. Núñez-Lagos, M. Barrera, and M. L. Romero, 269,
The En, Uscore, and Accuracy Parameters — A Topic to Debate Em. Cincu, I. Cazan, and V. Manu, 275,
Collaborative Study for Pesticides Residues Determination in Water Samples (Method 5060 APAT-IRSA CNR) — Project 4b L. 93/01 M. Antoci, S. Barbizzi, B. Bencivenga, D. Centioli, S. Finocchiaro, M. Fiore, F. Fiume, V. Giudice, M. Lorenzin, M. C. Manca, M. Morelli, E. Sesia, and M. Volante, 284,
Gross α/β Measurements in Drinking Waters by Liquid Scintillation Technique: Validation and interlaboratory Comparison Data I. Lopes and M. J. Madruga, 294,
Improvement of a Radiochemical Laboratory Through Fourteen Years Participation in a Intercomparison Program M. H. T. Taddei, 301,
Subject Index, 307,
CHAPTER 1
ENVIRONMENTAL METROLOGY IN ITALY: THE ROLE OF APAT
M. Belli
Agenzia per la Protezione dell'Ambiente e per i Servizi Tecnici (APAT), Servizio Laboratori, Misure ed Attivita di Campo, Via di Castel Romano, 100, 00128 Roma, Italy
1 INTRODUCTION
The objective of environmental monitoring is to quantify the condition of ecological systems in spatial and temporal differentiation. The organized and systematic measurement of selected variables provides the establishment of baseline data and the identification of both natural and human-induced changes in the environment. Therefore, spatially and temporally measured values have to correspond to real conditions and not to different measurement methods. In addition, the environmental monitoring enables to guide the formulation and the implementation of environmental management policies designed to protect human health and well-being, which includes ecological well-being. In general, the objectives of environmental monitoring programs can be summarized as follows;
• to verify the compliance with national or international environmental quality standards;
• to provide a basis for the implementation of environmental legislation;
• to assess human population and ecosystem exposure to pollution.
To reach these objectives, both at national or international level, a combination of a high amount of environmental data are necessary, collected in different period and coming from different sources. In the case of data used to provide information on the status of natural values and threatening processes, and to determine the type and magnitude of trends over time in assessing long-term trends, quality assurance, high precision and consistency of data, are of the utmost importance. The same considerations are valid if the monitoring data are used to evaluate the pattern of environmental contamination across a country or Europe. In addition, the assessment of compliance with national or international environmental quality standards, is vitally dependent on reliable environmental data. In this frame, the analytical laboratories have a great role, as the results of analytical measurements may be the basis upon which economic, legal or environmental management decisions are made, and they are essential in international trade, environmental protection, safe transportation, law enforcement, consumer safety and the preservation of human health. It is essential that such measurements are accurate, reliable, cost effective and defensible to ensure that correct decisions are made. This requires data determined with standardized methods, measurement results traceable to national or international standards with a stated measurements uncertainty and implementation of quality assurance/quality control (QA/QC) systems.
The environmental monitoring in Italy is performed at regional level and the data are then collected and combined by the Italian National Environmental Protection Agency (APAT), in order to assess trends and human and ecosystem exposure to pollution. Analytical data are produced by around one hundred laboratories belonging to the Italian Regional and Provincial Environmental Protection Agencies (ARPAs/APPAs) that in some cases use different approaches and different analytical methods. As focal point for the European Environmental Agency, APAT has the role to harmonize the environmental monitoring activities within the Italian territory, and to assure the comparability of the environmental data produced at regional level. To this end the following initiatives were promoted by the Environmental Metrology Service (Servizio Laboratori, Misure ed Attività di Campo) of APAT:
• the provision of matrix reference materials, similar to the test sample being measured in the regional laboratories (soil, sediments, compost, water, wastes, etc.), widely used for internal and external quality control activities. The analytes of interest are metals (for liquid and solid matrices), pesticides, organic pollutants and anions/cations (for liquid matrices);
• the establishment at national level of a permanent advisory group (GTP), according to the ISO guide 43-1 on proficiency testing scheme. GTP includes representatives from each regional agency. Quality managers, user of analytical data and laboratories are represented in the permanent advisory group;
• organization of proficiency tests and inter-laboratory comparison exercises;
• the establishment of a network of reference laboratories at national level (expert laboratories) for the different analytical field, in order to harmonise the different analytical methods used in environmental monitoring activities;
• the establishment of a link with the Italian National Institute of Metrology to disseminate the concepts of traceability in the Italian Regional and Provincial Environmental Protection Agencies laboratories, trough dedicated courses and equipment calibration.
2 REFERENCE MATERIAL PRODUCTION AND CHARACTERIZATION
Reference materials (RM) are one of the tools used to obtain comparability of analytical results. In recent times, there has been increasing interest worldwide in the accuracy, traceability and comparability of analytical measurements and the role that matrix reference materials play in the process. The ultimate goal of any measurement process is to ensure accuracy and to establish traceability to common universal reference points (preferably the SI) through an unbroken chain of comparisons. The use of reference materials is essential for method validations, calibration and internal/external quality control activities. The reference materials produced by APAT are closely matching in terms of matrix to the samples analyzed in the Italian environmental laboratories and are used mainly for the organization of inter-laboratory exercises. The organization of inter-laboratory exercises provides a continuous check on the comparability of environmental analytical results across Italy and identifies the determinations for which improvements are required. In addition, the reference materials produced by APAT may be used by the regional laboratories, as internal quality control materials, to check routinely if a measurement method is under statistical control.
Solid and liquid reference materials are produced following ISO Guide 35. As an example for solid materials, the raw material is dried at a constant temperature of +40°C in a ventilated oven. The material is then sieved through a 2 mm mesh sieve and the resulting fraction above 2mm is discarded. The fraction below 2mm is milled into powder (<90 micrometers) and homogenized over two weeks by mixing into a cylindrical drum placed on a roll-bed. The bulk homogeneity of the sample is checked by measuring the C and N concentrations on 10 sub-samples (10-15 g each), taken directly from the cylindrical drum. If the data of C and N content does not show any measurable heterogeneity (coefficient of variation below 1% for C and N), the material is bottled. The bottling is carried out in one day. Bottling is carried out in a way to prevent the possible segregation of fine particles (20 samples, each of about 30 g, are taken from the centre of the cylindrical drum immediately after stopping the rotation and placed into 10 pre-cleaned brown glass bottles). The drum is again rotated for a further 2 minutes and again 20 samples are taken in the same way and bottled. The sampling from the cylindrical drum and the bottling of the samples continued following this procedure until the material is finished. Each batch of RM is formed by about 1000 bottles.
The main characteristics of reference materials produced by APAT are homogeneity and stability of the analytes of interest. The within- and between-bottles variability is estimated by one-way analysis of variance (ANOVA) for the analytes of interest, as required by ISO Guide 35. The homogeneity of the material, or better the fitness for purpose of the material, is then confirmed using the results of the inter-laboratory exercises.
A short term stability study is carried out to demonstrate that the contents of the analytes of interest are not changed in the period of the inter-laboratory exercises. The stability study is carried out at different temperatures to study the effect of different temperatures on the properties of the material. Generally, the stability study is carried out at 40 and 20 °C and normally last 3 months. At the start of the stability test, 25 bottles are stored at a reference temperature (-18 °C) at which it is assumed that no instability can occur. Additional 5 bottles are stored at +20 °C and 5 bottles at +40 °C. After 1, 2 and 3 months, 5 bottles are transferred from -18 °C to +20 °C and 5 bottles from -18 °C to +40 °C After three months 3 sub-samples collected from each bottle are measured in one run (under repeatability conditions). As homogeneity tests, stability tests are carried out on the analytes of interest. For each temperature (+20 and +40 °C), the following parameters are assessed:
• standard deviations between the bottles stored at the same temperature for the same time interval, the mean value of concentration and the coefficient of variation (CV %);
• standard deviations between the mean values of concentration of bottles stored for different time periods, the mean values and the coefficient of variation (CV %);
• ratios of the mean values of measurements on bottles stored at +20 °C and +40 °C, respectively, and the mean values of measurements on samples stored at -18 °C for the same period;
• linear regression of the above mentioned ratios, the uncertainty contribution due to the material stability;
• analysis of variance (ANOVA) to assess the influence of the storage period at +20 °C and +40 °C on the stability of the material.
The characterization of the reference materials, defined as the assignment of concentration data to the interested analytes which approaches as closely as possible the "true value", together with uncertainty limits, is another step of prime importance for the use of reference materials. The key characteristic of a reference material is that the properties of interest are measured and assigned on the basis of the accuracy. The goal is the arrival at the best possible estimate of the unknown "true value". It implies the reliable assignment of a value to a property of a material. It encompasses selection of measurands, appropriate analytical methodologies, adequately calibrated and properly used.
The assignment of property value to the reference materials produced in the APAT laboratories is carried out following different strategies: measurements in one or two expert laboratories working independently using different analytical methods or more likely, collaborative analysis in a group of expert laboratories using one definitive method. Robust mean and robust standard deviation are calculated for each set of analytical results. The assigned values are then determined as the robust average of all laboratory mean values, while the expanded uncertainty range is calculated as reported in ISO 13528.
3 ORGANIZATION OF INTERLABORATORY EXERCISES
As above reported, one of the main purpose of the reference materials prepared in APAT Laboratory is the organization of inter-laboratory comparisons, to check the overall quality of the analytical results of the environmental laboratory across Italy. The main aims of the inter-laboratory exercises organized by APAT are: to get information on the comparability of the environmental analytical results across the Country, to define the analytical methods that are fitness-for-purpose for environmental monitoring activities and to assist the ARPA/APPA laboratories in meeting the requirements of the monitoring activities. Literature reports many different schemes to carry out inter-laboratory comparisons and proficiency testing in chemical and biological analyses, specially for scores assignment. Furthermore, environmental monitoring involves the analysis of a vast range of matrices with a wide number of analytes to be determined. In order to design inter-laboratory exercises and proficiency testing that are fit for purpose for environmental monitoring, a Technical Advisory Group (GTP) was established by APAT to ensure a smooth operation and the success of all inter-laboratory exercises. The support of an advisory group to the PT and inter laboratory comparison organisers is suggested also by the ISO Guide 43. The functions and the activities of the GTP coordinated by APAT are reported in another paper on this proceedings.
4 NETWORK OF REFERENCE LABORATORIES
In 2004 APAT has established a first network of reference laboratories to support APAT and ARPA/APPA system in environmental monitoring activities. The core task of this first network is the technical evaluation and validation of methods for pesticides, furan, dioxin and asbestos. Results achieved in method validation of a multiresidue analytical method for the determination of pesticide residues in water samples are reported elsewhere in this book.
Excerpted from Combining and Reporting Analytical Results by A. Fajgelj, M. Belli, U. Sansone. Copyright © 2007 The Royal Society of Chemistry. Excerpted by permission of The Royal Society of Chemistry.
All rights reserved. No part of this excerpt may be reproduced or reprinted without permission in writing from the publisher.
Excerpts are provided by Dial-A-Book Inc. solely for the personal use of visitors to this web site.
