CRM matrix

The analytical procedure is checked by analyses of method blanlcs to assure that secondary contamination by the analytes to be determined is avoided or minimized. Because the water content of the CRM matrix to be analyzed may vary from one laboratory to another (dependent on the local humidity and temperature), the water content has to be determined. Accordingly, at least three independent samples are kept at I05°C for 2 h, then allowed to cool to ambient temperature in a desiccator and the water loss is determined. The certified values are generally reported on a dry mass basis. 

As an example of how the foregoing discussion might be used in practice, consider a situation in which a certified reference material (CRM) is being analyzed in order to test the ability of a laboratory to reproduce the certified value Xcert for the concentration of a target analyte in the CRM matrix assume that a sufficiently large number of replicate analyses was performed to justify use of the unmodified Gaussian model. This is a case where the value of [I (the true value of x) can reasonably be claimed to be known (= x ) the test amounts to... 

Valuable contributions were made by two Canadian agencies, particularly by the National Research Council Canada (NRCC) who, from about 1976, provided marine and marine biological CRMs certified for metals, metal species and organic constituents (Berman 1984 Willie 1997). More recently their Halifax laboratories have issued a highly respected range of CRMs for the determination of shellfish toxins. Another Canadian producer, the National Water Research Institute (NWRI) specialized in marine (water and sedimentary) CRMs, and from the late 1980 s their matrix materials certified also for organic compounds (Chau et al. 1979 Lee and Chau 1987). 

Further afield, in 1978 the Japanese National Institute for Environmental Studies (NIES) started the production of a series of biological and environmental matrix CRMs, certified for a number of trace elements (Okamoto and Fuwa 1985). Recently also the certification of metal species in some materials was reported (Okamoto and Yoshinaga 1999). 

A remarkable level of activity can be seen in China. The National Research Center for CRM (NRCCRM) was founded in 1980 and the certification and accreditation program for CBW RMs started in 1983 by co-operation with many Chinese Institutions. In 1993 around 60 RMs and CRMs were available (Chai Chifang 1993) and in 1999 the availability of about 1000 CRMs was reported, around 30 of them clinical, 100 environmental, 200 geological, and 300 metallic matrix materials (Rong and Min 1999). 

From the mid 1980 s the rise of Quahty Standards, Total Quahty Management and Accreditation schemes created a booming demand for RMs and CRMs. Thus, the use and production of matrix RMs rapidly increased the new IAEA database lists 56 producers from 22 cormtries and about 1640 RMs. The 1998 Comar database, which covers a much wider scope, hsts more than 200 producers and around 10 000 RMs see Chapter 8 for more details. 

Most CRMs are so-called matrix-CRMs , identifying that they have been made from material sampled in nature. For these materials, it is impossible to come up with property values traceable to SI, as preparation steps carmot directly be related to that. At best, a comparison can be made among methods, from which usually one is the technically best established method, and the results of such a comparison may flow into the establishment of the property values and their respective uncertainties. 

The most important document, accompanying a CRM is its certificate. ISO Guide 31 (1981) provides guidance for the establishment of certificates, labeling of CRMs, and certification reports. The certificate contains among other information the certified values and their respective uncertainties. As important as this information is the traceability statement, which defines to what references the CRM is traceable. Ideally, a CRM is traceable to a suitable (combination) of SI units. This is not always possible, so other stated references may appear here. Especially when certifying matrix reference materials, making the measurements traceable to SI does not imply that the CRM is traceable to SI as well. The steps necessary to transform the sample into a state that can be measured may have a serious impact on the traceability of the values, and thus on the traceability statement. 

ISO Guide 33 (1998) deals with other uses of RMs. It elaborates on various uses of RMs, excluding calibration, which is the subject of ISO Guide 32. In most cases, RMs are used as a quality control measure, i.e. to assess the performance of a measurement method. Most matrix RMs are produced with this purpose in mind. Other purposes of RMs are the maintenance of conventional scales, such as the octane number and the pH scale. ISO Guide 33 provides guidance on the proper use of RMs, and therefore it is together with ISO Guide 32 the most important document for users of CRMs. 

One example, a candidate matrix material of organotin species in marine water, had stability determined by storage for 120 days at 4°C in the dark, at ambient temperature, and exposed to daylight (Quevauviller and Donard 1991). Frequently storage at different temperatures over at least a i-year period are reported. Examples include organochlorine pesticides (OCPs) in BCR CRM 430, where pork fat was stored at -2o°C, -i-20°C, and -r37°C (van der Paauw et al. 1992). Storage at -20°C, -i-20°C, and -i-4o°C was performed for total and methyl Hg in BCR CRMs 463 and 464, tuna fish (Quevauviller et al. 1994), and metals in BCR CRM 600, EDTA and DTPA-extractable trace metal contents in calcareous soil (Quevauviller et al. 1998m). 

Table 3.6 presents examples for environmental and biological natural matrix CRMs from BCR, NIST and NRCC certified for elements using the above described analytical multi-method approaches. 

Tab. 3.6 Examples for BCR, NIST and NRCC natural biological and environmental matrix CRMs/ SRMs certified for elements...

Certified reference materials (CRMs) to validate measurements of organic constituents were introduced in the early 1980 s, more than a decade after the development of the first natural matrix CRMs for inorganic constituents. There are three types of CRMs to support measurements of organic constituents ... 

CRMs for Contaminants in Environmental Matrices For nearly two decades NIST has been involved in the development of SRMs for the determination of organic contaminants such as polycyclic aromatic hydrocarbons (PAHs), polychlorinated biphenyls (PCBs), and chlorinated pesticides in natural environmental matrices such as fossil fuels (Hertz et al.1980 Kline et al. 1985), air and diesel particulate material (May and Wise 1984 Wise et al. 2000), coal tar (Wise et al. 1988a), sediment (Schantz et al. 1990, 1995a Wise et al. 1995), mussel tissue (Wise et al. 1991 Schantz et al. 1997a), fish oil, and whale blubber (Schantz et al. 1995b). Several papers have reviewed and summarized the development of these environmental matrix SRMs (Wise et al. 1988b Wise 1993 Wise and Schantz 1997 Wise et al. 2000). Seventeen natural matrix SRMs for the determination of organic contaminants are currently available from NIST with certified and reference concentrations primarily for PAHs, PCBs, chlorinated pesticides, polychlorinated dibenzo-p-dioxins (PCDDs), and polychlorinated dibenzofiirans (PCDFs) see Table 3.11. 

The BCR has also produced a large number of environmental matrix CRMs for PAHs, PCBs, pesticides, and PCDDs/PCDFs as shown in Table 3.12. These matrices include both natural contaminant level matrices as well as natural matrices spiked with low and high levels of contaminants. When viewed together the NIST and BCR CRMs provide a wide range of environmental matrices in which a considerable number of analytes have been assigned certified and reference values. 

Tab. 3.12 BCR natural matrix CRMs for the determination of organic contaminants in environmen-...

Fig. 3.3 General scheme used by the BCR for the certification of natural matrix CRMs...

BCR Analytical Approach for the Certification of PAHs in Natural Matrix CRMs Prior to the certification analyses for the CRM, each participating laboratory has to prepare standard solutions of the analytes to be determined from certified reference compounds (purity >99.0 %) to calibrate their instruments for response and response linearity (multiple point calibration), detection limit, and reproducibility. In the case of PAH measurements, reference compounds of certified purity are used as internal standards, which are not present at a detectable concentration in the matrix to be analyzed (e.g. indeno[i,2,3-cd]fluoranthene (CRM 267), 5-methylchrysene (CRM 081R), benzo[f ]chry-sene (CRM 046), picene (CRM 168), and/or phenanthrene-dio). 

To assess homogeneity, the distribution of chemical constituents in a matrix is at the core of the investigation. This distribution can range from a random temporal and spatial occurrence at atomic or molecular levels over well defined patterns in crystalline structures to clusters of a chemical of microscopic to macroscopic scale. Although many physical and optical methods as well as analytical chemistry methods are used to visualize and quantify such spatial distributions, the determination of chemical homogeneity in a CRM must be treated as part of the uncertainty budget affecting analytical chemistry measurements. 

Particularly for direct microanalytical techniques using <10 mg of sample for analysis, it is highly desirable to obtain quantitative information on element- and compound-specific homogeneity in the certificates for validation and quality control of measurements. As the mean concentration in a CRM is clearly material-related, the standard deviation of this mean value should represent the element s distribution in this matrix rather than differences in the analytical procedures used. 

An important prerequisite for the use of CRMs as calibrants, at least for optical methods and particularly all AAS modes, is that they should match the matrix and level of analyte contents of the materials to be analyzed as closely as possible, so that potential matrix effects will be compensated if calibrant and sample material are affected by the applied method, e.g. the temperature program for furnace techniques, in the same way. Further it is very important for all methods that the CRMs used should not show a nugget effecf, i.e. particles with extremely high analyte content that can lead to a high analyte heterogeneity (Kurfiirst 1991 Kurfiirst et al. 

Within collaborative work on element concentrations in a number of biological reference materials using solid sampling and other analytical methods, calibration of Cd, Cu, Pb and Zn in BCR CRM 185 Bovine liver with solid CRMs was performed for each element with a reference material of the same matrix, NIST SRM 1577... 

Bovine liver and CRMs of a different matrix and good to acceptable results were obtained. (Schauenburg and Weigert 1991). 

With solid sampling-electrothermal vaporization-inductively coupled atomic emission spectrometry (SS-ETV-ICP-AES), Cu in two environmental CRMs was determined using a third CRM with similar matrix as calibrant. Comparison with a reference solution showed good agreement (Verrept et al. 1993). 

Solid sampling techniques for ICP emission spectrometry need to be calibrated by very carefully characterized RMs or CRMs, if available. For example, the determination of Pb by vaporization of very small sample amounts into an ICP source was shown to be independent from matrix effects by using different CRMs for cahbra-tion (Ohls 1989). 

The composition of steels or other metals is commonly analyzed by emission or X-ray spectrometry during and after the production process. Both methods have to be calibrated by solid samples. These are either exactly analyzed samples taken from the same process or synthetic melted mixtures of the matrix with added accompanying elements (RMs). Available CRMs are then used to control the slope of the calibration function. Today, available RMs and CRMs are increasingly and exclusively used in spectral laboratories as the chemical analysis became much restricted and typical control laboratories were totally closed (Slickers 1993). 

Many more common problems start because the new users do not really understand their analytical systems, a problem described as the Nintendo scientist syndrome by Jenks (1995). Inexperienced scientists are often not sufficiently discriminating in their selection and use of CRMs. But incorrect choice can also be due to the unavailability of suitable matched matrix CRMs, or surprisingly often because the laboratory believes it cannot aftbrd the ideal product. 

In trace organic analysis there is usually an extraction or clean-up process, rather than a sample dissolution. Here not only must the matrix effect be considered, but also the recovery yield of the extraction. Frequently an external spike standard is added, but there is often no way of knowing if the recovery of the spike standard matches the analyte in question. There is considerable evidence that the U S E P A method for VOA analysis (Minnich 1993) is subject to such error, as reported by Schumacher and Ward (7997). The analyst must always consider the possibility of such an error, especially when using CRMs to control methods that are applied in routine mode. 

The importance of particle size is directly proportional to the sub-sample size recommended by the analytical method. The larger the sub-sample size the larger the acceptable particle size. For sub-sample sizes of ig or greater a soil sieved through a imm screen is generally acceptable. Therefore if the sample is relatively coarse, e.g up to 2mm particles and the matrix CRM is an uniform sub-micron powder, it may be necessary to use a much larger sample from the material under test than for the CRM. 

Only a direct matrix match of sample and CRM, and the CRM s use as a direct calibrant will allow the user to demonstrate accuracy and subsequently traceability close to the uncertainties established during the CRM certification ( note matrixmatching may not be necessary with matrix-independent techniques). This reality places a significant burden on the CRM producers, since large uncertainties in the certified values may degrade the perceived value of the CRM. 

---