Use of uncertainty information for estimation of certified values. Comparison of four approaches using data from a recent certification exercise

Four approaches for estimation of reference values and their respective uncertainties of characterisation were compared using data from the recently finalised certified reference materials ERM-EC680k and ERM-EC681k, elements in plastics. Reference values and uncertainties of characterisation were estimated as mean of laboratory means and their respective standard deviations, using equal weights and the weighting procedure of Mandel–Paule. In addition, two approaches taking into consideration uncertainty information reported by the participants, namely the consistency check and simulation procedure proposed by Cox for CCQM Key comparisons and an approach suggested by Pauwels et al. (Accred Qual Assur 3:180–184, 2000) were used. No difference between the equally-weighted and Mandel–Paule consensus means was observed and the reference value from the Cox approach was in all cases within ±2 u _char of each consensus mean. Uncertainties varied between the three approaches. Uncertainties derived from equally-weighted mean of means approach are on average 14% above uncertainties using the Mandel–Paule consensus mean, 36% above the uncertainties estimated by Pauwels et al., and 54% above the uncertainties from the Cox approach. Robustness of the uncertainty estimation against incorrect estimation of uncertainties was assessed. Assumption of a 50% uncertainty of the individual uncertainties resulted in an uncertainty of 30% of the uncertainty of characterisation. Differences between the four approaches are negligible for this dataset when combined with the uncertainty contribution from heterogeneity and stability as prescribed in ISO Guide 35. Electronic supplementary material The online version of this article (doi:) contains supplementary material, which is available to authorized users.     

Estimation of uncertainty for the determination of mercury in food by CVAAS

This paper is aimed at reporting a full validation of a Cold Vapour Atomic Absorption Spectrometry (CVAAS) method for mercury determination in fishery products. Method precision, trueness, limit of detection and limit of quantification are evaluated. The uncertainty of measurement is estimated following the bottom-up approach in conjunction with the validation data as suggested for complex analysis. The precision is evaluated using a pooled relative standard deviation and the recovery is added to the measurement uncertainty budget. Uncertainties of weight of sample, dilution factor, calibration, work solution and analyte amount are also assessed. The most relevant uncertainty sources result those associated to amount of mercury determined in the final sample solution, to method precision and to recovery. The relative contributions have different weights according to the analyst decision regarding recovery correction of results. In conclusion, the adopted CVAAS method fully complies with EU requirements. Presented at AOAC Europe/Eurachem Symposium March 2005, Brussels, Belgium     

A case study in the practical estimation of measurement uncertainty

This case study is written for analytical laboratories, in order to give support to the implementation of the concept of measurement uncertainty for routine measurements. The aim is to provide a practical, understandable and common way of performing measurement uncertainty calculations, based mainly on pre-existing quality control and validation data. Practical examples taken directly from environmental laboratory monitoring are presented and explained. However, the approach is very general and should be applicable to most testing laboratories in the chemical field. Following the protocol of evaluation illustrated in the case study, it is possible to ensure that most relevant uncertainty components associated with the method are covered. Contributions associated with sampling, homogenisation, sub-sampling, and so on, are, however, excluded.     

Estimates of uncertainty of measurement from proficiency testing data: a case study

The results obtained by a laboratory over a number of proficiency testing/external quality assessment schemes (PT/EQAS) rounds can give information on the uncertainty of its measurements for a given test, provided that conditions such as full coverage of the routine analytical range, traceability, and small uncertainty of the assigned values (compared to the spread of the results) are met and provided that systematic deviations and any other sources of uncertainty are considered. As organisers of the Italian EQAS (ITEQAS) in occupational and environmental laboratory medicine, we tested this hypothesis using as model data from well-performing laboratories taking part in ITEQAS for lead in blood over the last 2 years. We also investigated how different PT/EQAS features (frequency of trials and number of samples) would affect a laboratory estimate of its uncertainty. Such information can be helpful in improving PT/EQAS organisation and define, for a given test: (a) the state of the art of the uncertainty of current measurement procedures, (b) identify needs for improvement of analytical methodologies and (c) set targets for acceptable uncertainty values.Presented at the Eurachem PT Workshop September 2005, Portorož, Slovenia.Papers published in this section do not necessarily reflect the opinion of the Editors, the Editorial Board and the Publisher.     

Evaluation of the stability of ethanol in water certified reference material: measurement uncertainty under transport and storage conditions

This study simulated the transport and storage conditions of ethanol in water certified reference material (CRM) produced by the Chemical Metrology Division of Inmetro—DQUIM with the purpose of estimating the measurement uncertainty related to stability. The short-term stability study was performed on five different mass fractions (w) in terms of mg ethanol/g solution of the ethanol in water CRM. The nominal values are w = 0.5, 0.9, 1.1, 3.8 and 4.6 mg/g, at temperatures of 4 and 60 °C. On the other hand, the long-term stability study was developed on four different mass fractions (nominal values): w = 0.5, 0.9, 1.1 and 4.6 mg/g, at a temperature of 20 °C. This paper will show the data from the long-term stability study that took place over 52 weeks. The method used complies with ISO Guide 35, the BCR Guideline for Feasibility Studies and ISO Guide 34. According to the statistical parameters used in both studies, the stability of ethanol in water CRM was confirmed for all of the mass fractions studied.     

Purity determination and uncertainty evaluation of theophylline by mass balance method, high performance liquid chromatography and differential scanning calorimetry

Purity determination of pure organic substance is essential for the establishment of traceability to SI units. A mass balance method was employed to determine the purity of theophylline certified reference materials (CRM), compared with high performance liquid chromatography (HPLC) and differential scanning calorimetry (DSC). In the approach of the mass balance, the impurities were identified by ion trap time-of-flight mass spectrometer (IT-TOF-MS) and quantified by HPLC. The purity of theophylline CRM determined by mass balance method was 99.82% with an extended uncertainty of 0.1% (k = 2). The uncertainty evaluation of purity demonstrated that the accuracy of the mass balance method is better than that of HPLC and DSC. It indicated that the mass balance is suitable for the CRM and pharmaceutical standards.     

Production of artifact methylmercury during the analysis of certified reference sediments: use of ionic exchange in the sample treatment step to minimise the problem

Production of artifact methylmercury (MeHg⁺) during the analysis of two certified reference sediments, CRM-580 and IAEA-405, was investigated. Leaching of the analyte from the solid sample was achieved by ultrasound assisted acidic extraction. The aqueous leachate was either ethylated (NaBEt₄) or phenylated (NaBPh₄) using acetic/acetate or citric/citrate to buffer the solution. Preconcentration of the volatile compounds was carried out by extraction with an organic solvent (n-hexane) or solid phase microextraction (SPME). MeHg⁺ was finally separated and detected by gas chromatography with atomic emission or mass spectrometry detection (GC-MIP-AED or GC-MS). In all the cases the concentrations obtained for MeHg⁺ in the CRM-580 were significantly higher than the certified value. For the IAEA-405, however, the MeHg⁺ concentration found was always statistically indistinguishable from the certified value. Experiments were also conducted with synthetic samples, such as aqueous mixtures of MeHg⁺ and inorganic mercury (Hg²⁺) or silica-gel spiked with both compounds. The methylation rates found (defined as the percentage of Hg²⁺ present in the sample which methylates to give artifact MeHg⁺) ranged from not observable (in certain synthetic aqueous mixtures) to 0.57% (analysis of CRM-580 under certain conditions). As the amount of Hg²⁺ available in the sample seems to be the main factor controlling the magnitude of the artifact, several experiments were conducted using an ionic exchange resin (Dowex M-41) in order to minimise the concentration of this chemical in the reaction medium. First, a hydrochloric leachate of the sample was passed through a microcolumn packed with the exchanger. Second, the resin was mixed with the sample prior to extraction with HCl. In both cases, the predominant Hg²⁺ species, HgCl₄²⁻, was adsorbed on the resin, whereas MeHg⁺, mainly as MeHgCl, remained in solution. Following the second option, a new method to analyse MeHg⁺ in conflictive matrices like certain sediments was proposed. This approach produced better results for the CRM-580, but a MeHg⁺ concentration slightly, but statistically significant, higher than the reference value was still obtained.     

Comparison of digestion methods for ICP-MS determination of trace elements in fish tissues

A comparison is presented of six methods involving nitric acid in conjunction with other reagents to digest three certified marine biological samples (DOLT-3, DORM-3, IAEA-407) and a fish bone homogenate (prepared from Merluccius australis). An inductively coupled plasma-mass spectrometer with an octopole collision cell was used to determine up to 40 elements (Li, B, Na, Mg, Al, K, Ca, V, Cr, Mn, Fe, Co, Ni, Cu, Zn, As, Se, Rb, Sr, Y, Ag, Cd, Cs, Ba, La, Ce, Pr, Nd, Sm, Eu, Gd, Dy, Ho, Er, Tm, Yb, Lu, Pb, Th, U) in the digests. These results were assessed in terms of their accuracy and precision, and a flow chart was developed to aid the selection of the optimal digestion method. Although none of the methods was found to give accurate determinations for all elements in the different reference materials, a relatively simple method involving nitric acid and hydrogen peroxide heated in a domestic microwave oven gave the most acceptable results.     

Tributyltin determination in marine sediments: A comparative study of methods

Two independent speciation methods have been applied to the determination of tributyltin (TBT) in marine sediments: (1) acetic acid leaching/hydride generation/cold trapping/GC/quartz furnace AA; and (2) toluene–tropolone extraction/HPLC/GFAA. TBT determines in a typically moderately polluted sediment (Porto Vecchio Bay, Corsica, France) are in very good agreement, as also are the determination of total recoverable tin by the first method and by a direct one using strong acid leaching followed by GFAA. These cross-verifications lead one to conclude that both speciation methods are convenient and allow for valid determination of TBT in marine sediments, with no loss of TBT. Method-induced modification of tin speciation has not been observed.     

Practical digest for evaluating the uncertainty of analytical assays from validation data according to the LGC/VAM protocol

The estimation of the measurement uncertainty of analytical assays based on the LGC/VAM protocol from validation data is fully revisited and discussed in the light of the study of precision, trueness and robustness.     

Fractionation studies of mercury in soils and sediments: A review of the chemical reagents used for mercury extraction

Mercury in contaminated soils and sediments could be extracted by various chemical reagents in order to determine the different mercury species and partitions, providing useful information of toxicology, bioavailability and biogeochemical reactivity. Unfortunately, at present, neither specific extractants nor standard protocols exist for the isolation of particular mercury species. Although there has been considerable research focused on reagents for extracting mercury species, there is still little consensus. Thus, workers are advised to select the most appropriate reagent based on the nature of their sample, and to take all possible steps to validate the analyses performed. Therefore, the aim of this paper is to review the current reagents used for determining total mercury and its speciation as well as fractionation such as methylmercury, ethylmercury, elemental mercury, mercury sulphide and organically bound mercury by supposed selective (one reagent) and sequential (several reagents) extractions. The gathering information presented here bring to light the need for standard protocol for which the used chemical reagents should take into account the particular chemistry of mercury associated with specific properties of soil and sediment. Beside this required scheme, appropriate reference materials are also demanded.     

The determination of adsorbable organically bound halogens (AOX) in soil: interlaboratory comparisons and reference materials

The interlaboratory variability in the quantification of adsorbable organically bound halogens (AOX) in industrially contaminated soil is presented. Three consecutive rounds of a proficiency testing scheme, in which between 88 and 119 routine laboratories participated, yielded relative reproducibility standard deviations between 7 and 20% at AOX contents between 10.9 and 268 mg kg⁻¹. Nineteen laboratories with established proficiency were invited to participate in the certification of the AOX content in three soil reference materials meant for the internal quality control in analytical laboratories. The certified values are (1349 ± 59) mg kg⁻¹, (80 ± 7) mg kg⁻¹ and (102 ± 8) mg kg⁻¹, respectively.     

Speciation analysis of mercury in sediments, zoobenthos and river water samples by high-performance liquid chromatography hyphenated to atomic fluorescence spectrometry following preconcentration by solid phase extraction

A high-pressure microwave digestion was applied for microwave-assisted extraction (MAE) of mercury species from sediments and zoobenthos samples. A mixture containing 3 mol L⁻¹ HCl, 50% aqueous methanol and 0.2 mol L⁻¹ citric acid (for masking co-extracted Fe³⁺) was selected as the most suitable extraction agent. The efficiency of proposed extraction method was better than 95% with R.S.D. below 6%. A preconcentration method utilizing a “homemade” C18 solid phase extraction (SPE) microcolumns was developed to enhance sensitivity of the mercury species determination using on-column complex formation of mercury-2-mercaptophenol complexes. Methanol was chosen for counter-current elution of the retained mercury complexes achieving a preconcentration factor as much as 1000. The preconcentration method was applied for the speciation analysis of mercury in river water samples. The high-performance liquid chromatography-cold vapour atomic fluorescence spectrometric (HPLC/CV-AFS) method was used for the speciation analysis of mercury. The complete separation of four mercury species was achieved by an isocratic elution of aqueous methanol (65%/35%) on a Zorbax SB-C18 column (4.6 mm × 150 mm, 5 μm) using the same complexation reagent (2-mercaptophenol). The limits of detection were 4.3 μg L⁻¹ for methylmercury (MeHg⁺), 1.4 μg L⁻¹ for ethylmercury (EtHg⁺), 0.8 μg L⁻¹ for inorganic mercury (Hg²⁺), 0.8 μg L⁻¹ for phenylmercury (PhHg⁺).     

Method to reduce the memory effect of mercury in the analysis of fish tissue using inductively coupled plasma mass spectrometry

The routine determination of mercury (Hg) by inductively coupled plasma mass spectrometry (ICP-MS) is affected by a pronounced memory effect in the sample introduction system. This results in long washout times for the analyte, which affects the accuracy and reliability of the analytical procedure. By using a combination of flow injection sample introduction and a sulfur-containing compound in the carrier solution, it was possible to decrease the memory effect of mercury to that for the internal standard (rhodium). The carrier solution contained 2-mercaptoethanol (2-ME) and the developed method was evaluated using three different fish tissue certified reference materials: CRM 464 (BCR, Brussels); DORM-1; and DORM-2 (NRC, Canada). The samples were mineralized using a combination of concentrated nitric acid and hydrogen peroxide and heating in a closed microwave oven. The developed flow injection ICP-MS procedure gave values for total mercury in all three CRM materials in agreement with the certified concentration range. Cold vapour atomic fluorescence spectrometry (CV-AFS) confirmed the results from the developed method. The developed flow injection method had a detection limit (defined as three times the standard deviation of the blank concentration) for mercury of 5.1 μg l⁻¹.     

Geochemical investigation of alluvial sediments: validation of ICP-OES determination of heavy metals. A case study from the Utrata River Valley (central Poland)

The development and validation of a reliable analytical procedure for the determination of selected metals (Cd, Cr, Cu, Pb, Zn and Mn) in sediments accumulated in the Utrata River (Poland) is described. The aqua regia extraction followed by inductively coupled plasma optical emission spectrometry (ICP-OES) was used for this purpose. The optimized analytical procedure was validated, and adequate quality control actions were implemented in order to provide reliable data. The precision under’ within-laboratory’ reproducibility conditions was estimated from duplicate analysis. Certified reference material (CRM) was used in order to evaluate the accuracy of the results regarding the sewage sludge amended soil CRM 143R. The detection limits for all elements of interest were well-below their content in the investigated sediment samples. The obtained reliable data could be used for assessment of the relationship between human economic activity in the past and the geochemical features of the sediments.     

Determination of methylmercury in marine sediment samples: Method validation and occurrence data

The determination of methylmercury (MeHg) in sediment samples is a difficult task due to the extremely low MeHg/THg (total mercury) ratio and species interconversion. Here, we present the method validation of a cost-effective fit-for-purpose analytical procedure for the measurement of MeHg in sediments, which is based on aqueous phase ethylation, followed by purge and trap and hyphenated gas chromatography–pyrolysis–atomic fluorescence spectrometry (GC–Py–AFS) separation and detection. Four different extraction techniques, namely acid and alkaline leaching followed by solvent extraction and evaporation, microwave-assisted extraction with 2-mercaptoethanol, and acid leaching, solvent extraction and back extraction into sodium thiosulfate, were examined regarding their potential to selectively extract MeHg from estuarine sediment IAEA-405 certified reference material (CRM). The procedure based on acid leaching with HNO₃/CuSO₄, solvent extraction and back extraction into Na₂S₂O₃ yielded the highest extraction recovery, i.e., 94 ± 3% and offered the possibility to perform the extraction of a large number of samples in a short time, by eliminating the evaporation step. The artifact formation of MeHg was evaluated by high performance liquid chromatography coupled to inductively coupled plasma mass spectrometry (HPLC–ICP–MS), using isotopically enriched Me²⁰¹Hg and ²⁰²Hg and it was found to be nonexistent. A full validation approach in line with ISO 17025 and Eurachem guidelines was followed. With this in mind, blanks, selectivity, working range (1–800 pg), linearity (0.9995), recovery (94–96%), repeatability (3%), intermediate precision (4%), limit of detection (0.45 pg) and limit of quantification (0.85 pg) were systematically assessed with CRM IAEA-405. The uncertainty budget was calculated and the major contribution to the combined uncertainty (16.24%, k = 2) was found to arise from the uncertainty associated with recovery (74.1%). Demonstration of traceability of measurement results is also presented. The validated measurement procedure was applied to the determination of MeHg incurred in sediments from a highly polluted and scarcely studied area in the Caribbean region.     

A multivariate study of the programmed temperature vaporization injection-gas chromatographic-mass spectrometric determination of polycyclic aromatic hydrocarbons Application to marine sediments analysis

The trace analysis of polycyclic aromatic hydrocarbons (PAHs) in environmental matrices requires highly sensitive and selective methods. The use of large-volume injection in gas chromatography with mass spectrometry detection allows the improvement of the sensitivity. In this work, one of the most popular large-volume injectors, programmed temperature vaporization (PTV)-splitless injection, was optimized to determine 26 alkylated and non-alkylated PAHs at ultratrace levels. The injection conditions that use two different injection volumes, 10 and 25 μl, were optimized applying experimental designs. Six experimental factors that could affect the vaporization efficiency were studied. Detection limits for each assayed volume were compared with splitless injection, both in full scan and in SIM mode. An improvement of 50- and 100-fold, for 10 and 25 μl, respectively, was achieved for PAHs in comparison with splitless injection. Finally, the method that was optimized for 25 μl was applied successfully to the analysis of a certified marine sediment SRM 1941b and a real marine sediment extracted by microwave assisted extraction.     

Non-chromatographic screening method for the determination of mercury species. Application to the monitoring of mercury levels in Antarctic samples

A simple non-chromatographic method for the determination of mercury (Hg²⁺), methylmercury (MeHg⁺), dimethylmercury (Me₂Hg), and phenylmercury (PhHg⁺) employing atomic fluorescence spectrometry (AFS) as detection technique was developed. Mercury species showed a particular behavior in the presence of several reagents. In a first stage SnCl₂ was employed for Hg²⁺ determination; in a second step, [Hg²⁺ + PhHg⁺] concentration was determined using SnCl₂ and UV radiation. MeHg⁺ decomposition was prevented adding 2-mercaptoethanol. In a third stage, [Hg²⁺ + PhHg⁺ + MeHg⁺] concentration was determined using K₂S₂O₈. Finally, the four species were determined employing NaBH₄. Reagents concentration and flow rates were optimized. The extraction technique of mercury species involved the use of 2-mercaptoethanol as ion-pair reagent. The limits of detection for Hg²⁺, PhHg⁺, MeHg⁺, and Me₂Hg were 1, 40, 68, and 99 ng L⁻¹ with a relative standard deviation of 1.5, 3.1, 4.7 and 5.8%, respectively. Calibration curve was linear with a correlation factor equal to 0.9995. The method was successfully applied to the determination of the mercury species in two Antarctic materials: IRMM 813 (Adamussium colbecki) and MURST-ISS-A2 (Antarctic Krill).     

Comparison of three sequential extraction procedures to describe metal fractionation in anaerobic granular sludges

In the last few decades, several sequential extraction procedures have been developed to quantify the chemical status of metals in the solid phase. In this study, three extraction techniques (modified [A. Tessier, P.G.C. Campbell, M. Bisson, Anal. Chem. 51 (1979) 844]; [R.C. Stover, L.E. Sommers, D.J. Silvera, J. Water Pollut. Con. F. 48 (1976) 2165]; and the Bureau Communautaire de Reference (BCR) [K.F. Mossop, C.M. Davidson, Anal. Chim. Acta 478 (2003) 111]) were applied to study the distribution of trace (Co, Ni, Zn and Cu) and major (Mn and Fe) elements in two different anaerobic granular sludges from full-scale methanogenic wastewater treatment plants. The Stover scheme displayed a higher number of fractions that induces a poor recovery compared to the other schemes. The sequential extraction scheme recommended by BCR and the modified Tessier scheme gave similar trends and are sufficiently repeatable and reproducible for application in fractionation studies. However, the BCR scheme seems to be of limited utilisation to study anaerobic matrixes because the extraction stage for its reducible fraction may release substantial amounts of trace elements bound to the organic/sulfides fraction, and consequently, the recovery of trace elements in the oxide fraction may be overestimated at the expense of the oxidisable fraction. As a final conclusion, the modified Tessier scheme seems to be the most suitable scheme to study the metal partitioning in anaerobic granular sludges.     

Comparison of strategies to quantify uncertainty of lead measurements in biological tissue at mg kg<Superscript>–1</Superscript> level

The expression of measurement uncertainty in analytical chemistry still presents some difficulties. According to the various guides and standards, several approaches are possible. This paper present two referenced approaches applied to the determination of Pb in a biological tissue at the mg kg^–1 level. Estimations were achieved, either from inter-laboratory precision data, or from the uncertainty propagation equation applied to the measurements of a single laboratory. When comparing final values, for an average concentration of about 2.0 mg kg^–1, expanded uncertainties range from ±0.05 mg kg^–1 up to ±0.31 mg kg^–1, i.e. by a factor of 6. Obviously, both approaches of estimating uncertainty do not cover the same sources of uncertainty and it is interesting to try to understand better these discrepancies in order to try to define realistic guidelines for analysts. Moreover the experimental results of the study are given where a wide number of techniques is applied. Received: 5 May 2002 Accepted: 17 June 2002 Correspondence to M. Feinberg