Introduction

The volume of nanomaterialsFootnote 1 on the market is growing as the nanotechnology industry develops, generating an increasing number of consumer products containing nanomaterials. Nanomaterials may have properties that are different from those of the corresponding macroscale materials (if the latter exist at all). In order to better understand the properties of nanomaterials an increasing number of chemical, physical and biological tests are performed, including e.g. toxicological and ecotoxicological tests. These efforts underpin current EU consumer protection and environmental legislation, such as REACH (EC 2006), which is applicable also to nanomaterials.

Manufactured nanomaterials are often particulate materials, produced for example as a powder or as a dispersion of particles in a liquid matrix. Traditional methods for the testing of chemical substances are not necessarily adequate for assessing the properties and effects of these particulate nanomaterials (OECD 2009). Whether a certain method is applicable to a specific nanomaterial must be verified and, where necessary, the method has to be modified and (re-) validated (OECD 2010).

Nanotechnologies rely on the combined expertise from different scientific disciplines, several of which have developed their own terminology. In addition, regulatory authorities have developed a number of specific terms relevant, for example, for safety assessment of nanomaterials. This sometimes leads to one term having more than one meaning, or to different terms covering similar concepts (e.g. the terms ‘measurement’ and ‘test method’, see “The terms ‘testing’ and ‘measurement’” and “Steps in a test or measurement process” sections). The main body of the paper will focus on the specific aspects of benchmark materials, and definitions of other, related terms are given in the Annex to this paper, which also explains a number of the basic concepts for materials testing and measurement used in this paper.

Benchmark materials are materials or substances (see also “Substances, materials and samples” section) used for the assessment of the applicability of existing methods to new kinds of materials, modification of existing methods, development of new methods, validation of methods and quality control of routinely used methods. Table 1 gives an overview of the requirements for these materials with respect to the homogeneity and stability of their properties and the corresponding assigned values for different aspects of method validation and quality assurance. More information on the requirements summarised in Table 1 is provided in the Annex.

Table 1 Technical requirements of materials for method validation and quality assurance
Full size table

This paper aims to provide more clarity on the significance and the complementarity of the different kinds of benchmark materials. It describes the main features of previously defined classes of such materials, including (certified) reference materials, and the differences between their intended uses. In addition, we introduce the term ‘representative test material’, and provide a frame for its use. Examples are drawn from the activities of the European Commission’s Joint Research Centre institutes IRMM (Institute for Reference Materials and Measurements, Geel, Belgium) and IHCP (Institute for Health and Consumer Protection, Ispra, Italy), which provide a range of materials for the development, validation and quality assurance of test methods for nanomaterials.

Reference materials

Generic reference material terminology

Materials used in method validation and quality control (see Table 1) have to be sufficiently homogeneous and stable, so that results obtained by testing different sub-samples, in different places, at different times and with different methods, can be meaningfully compared. For specific purposes, these materials also need to have reliable, assigned property value(s).

Several scientific communities have developed their own terminology for the various kinds of materials fulfilling these requirements, which hampers mutual understanding. To remedy this situation, a dedicated ISO (International Organization for Standardization) Committee dealing with Reference Materials (ISO/REMCO) was created. To ensure that REMCO addresses all generic measurement and testing needs, across the main scientific disciplines physics, chemistry and biology, REMCO established liaisons with among others the World Health Organization (WHO), the International Organization of Legal Metrology, the Pharmacopoeia Discussion Group, the World Association of Societies of Pathology and Laboratory Medicine, the International Federation of Clinical Chemistry and Laboratory Medicine, and the International Union of Pure and Applied Chemistry (IUPAC), but also with specific ISO Technical Committees, such as ISO/TC 229 Nanotechnologies.

ISO/REMCO introduced the following generic reference material definition (ISO 2008a):

A reference material is a ‘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’.

Homogeneity and stability must be determined for each of the measurands (the quantities intended to be measured), corresponding with the properties of interest, and a reference material (RM) is a RM only for these specified properties. RMs produced and used applying the conditions and terms described in ISO Guide 34 meet these requirements (ISO 2009).

As indicated in Table 1, materials with a known property value are indeed required for several calibration, validation and quality assurance applications. ISO/REMCO is promoting the term ‘certified reference material’ for these materials and defines it as follows (ISO 2008a):

A certified reference material is a reference material characterized by a metrologically valid procedure for one or more specified properties, accompanied by a certificate that provides the value of the specified property, its associated uncertainty, and a statement of metrological traceability.

The basic difference between a reference material (RM) and a certified reference material (CRM) is the status of the property values assigned to the material. Metrologically valid procedures for the production and certification of reference materials can be found for instance in ISO Guides 34 and 35 (ISO 2009, 2006). CRMs are cornerstones of modern quality assurance because they allow calibration of instruments and quality control of methods and laboratories based on metrological traceability (JCGM 2008), and therefore ensure comparability of test results (Emons et al. 2004).

Synonyms for reference materials

The terms ‘control material’ as defined by IUPAC (IUPAC 1995, 2012), ‘in-house reference material’ as defined by the United Nations Office on Drugs and Crime (UNODC 2009), as well as the term ‘quality control material (QCM)’ (Emons 2006), are application-specific terms for subgroups of reference materials (Emons et al. 2004).

The OECD test guidelines use the terms reference substance and reference chemical which sound very similar to the term reference material. OECD reference substances and reference chemicals are used in tests where observed effects of an unknown substance are compared to known effects of a known substance in biological test systems (see OECD Guidance Document 34, OECD 2005). Since reference substances are chemicals for which responses in the selected biological test system are already known, they can be used in validation processes. This type of application of a material is similar to the use made of certified reference materials.

According to the Mutual Acceptance of Data (MAD) agreement established for OECD test guidelines, new tests should be performed applying Good Laboratory Practice (GLP). GLP uses the terms reference item (also called control item) and test item. The latter means an object (substance) that is the subject of a study, whereas the former means any object (substance) used to provide a basis for comparison with the test item (EC 2004). When samples of a reference substance or chemical are used as reference test items, principles of OECD Guidance Document 34 and GLP apply in a manner similar to the ISO Guides for reference materials mentioned above.

The most useful reference substances meet all technical requirements of a CRM (see Table 1 and see annex). Usually (eco) toxicological tests do not use formally certified reference substances, which would have to be accompanied by a certificate carrying all relevant information. The outcome of (eco) toxicological tests depends strongly on many factors such as the selected species, intra-species variations (individuals in any population have different susceptibilities), and the test method applied. It could be difficult to summarise these in a certification process in a cost-effective way. On the other hand, it is often the case that (eco) toxicological effects of a certain dose of material depend primarily on the chemical composition of the material. In this case, the reference character of a reference substance is sufficiently defined by its chemical composition, including purity and impurities, or by its activity in a well-defined test system and a fine chemical with a known composition (including impurities) may be used as reference material.

In the specific case of nanomaterials, knowledge of only the elemental or molecular composition may be insufficient for use as a reference substance in safety assessment, as nanomaterials may have (eco) toxicological and other properties that depend also on their size, shape, surface area, etc. The variations in these properties could have a larger influence on the test results than variations in the elemental or molecular composition within or between batches or between producers of nominally the same material (see also “Need for a new term” section).

Reference materials to assure metrological traceability in nanomaterial testing

Assuring the quality of testing of nanomaterials requires the same type of reference materials as the ones discussed in “Generic reference material terminology” section. One important feature is the metrological traceability of test results obtained on nanomaterials (see also Annex, “Metrological traceability” section).

For nanomaterials, lists of endpoints (properties), relevant for their safety assessment, are being established, for example by the OECD Working Party on Manufactured Nanomaterials (WPMN) (OECD 2008) and in ISO/TC 229 (Nanotechnologies) Working Group 3 (Health, Safety and Environmental Aspects of Nanotechnologies) (ISO 2011). Many of the listed endpoints, both physico-chemical and (eco) toxicological ones, are method-defined: the value of the property measured is determined by how (by which method) it is measured (Roebben et al. 2010). One example is the size of a particle. For regularly shaped particles, one value (e.g. the diameter of a sphere) or a few values (e.g. the length, width and height of a cuboid) can define the particle size. However, for particles with irregular shapes the so-called apparent size values are obtained, depending on how the particle size is measured and the measurement is evaluated. In practice, size measurements are often indirect measurements (e.g. of a sedimentation or diffusion rate), and therefore, even for near-spherical particles, the result of the measurement depends on the applied method.

An example is ERM-FD100 (Braun et al. 2011), a CRM for which four different particle size values are certified, each corresponding to a different measurement technique (electron microscopy (EM), centrifugal liquid sedimentation (CLS), small-angle X-ray scattering (SAXS) and dynamic light scattering (DLS)). For each of these methods, the interlaboratory study to characterise ERM-FD100 has demonstrated that the particle size can be reproducibly determined in different laboratories, if the laboratories follow the agreed measurement procedure, including a correct calibration of the instrument. The method-defined nature of the certified property values has important implications for the use of test results, as it is in principle impossible to convert results from one method to another, e.g. convert size as measured by DLS into size as measured by SAXS. However, the added value of this CRM, ERM-FD100, is that it provides the tool that laboratories need to ensure that the results of their DLS, CLS, EM and SAXS measurements of (near-) spherical nanoparticles are traceable to the SI unit metre, and thereby comparable with other results obtained with the same methods. More recently, IRMM released ERM-FD304 (Franks et al. 2012), a colloidal silica with monomodal size distribution, alike ERM-FD100, but with a higher polydispersity than ERM-FD100. Due to the higher polydispersity the differences between the certified values for different particle size methods are more pronounced for ERM-FD304 than for ERM-FD100.

Representative test materials

Need for a new term

The two terms ‘reference material’ and ‘certified reference material’ cover all uses mentioned in Table 1. This means that the materials needed for calibration, method validation and quality control must fulfil the requirements for RMs or even for CRMs. The high number of method-defined material properties (e.g. in the field of nanotechnology) implies that there is an equally large need for reference materials. However, reference materials are scarce, even in well-established fields like particle size analysis (Linsinger et al. 2011), whereas for many other measurands no reference materials exist at all. Indeed, the decision to develop a reference material is not taken lightly, largely due to the considerable resources needed to produce a reference material. Such decision depends on the size of the possible user community and on the consensus in this user community about the type of material needed and on the parameters to be assigned or certified. Therefore, it is current praxis in all technologies (including nanotechnology) that analysts use materials other than reference materials as a provisional benchmark, e.g. awaiting additional testing to ensure that the candidate reference materials and candidate standard methods meet the necessary requirements. We propose to give these materials their own name and status and call them representative test materials, with the following definition:

A representative test material (RTM) is a material from a single batch, which is sufficiently homogeneous and stable with respect to one or more specified properties, and which implicitly is assumed to be fit for its intended use in the development of test methods which target properties other than the properties for which homogeneity and stability have been demonstrated.

A RTM is not a reference material for the tests for which it is intended to be used, because homogeneity and stability are not demonstrated for the corresponding measurand. An RM can therefore be an RTM if it is used outside the range of properties for which it is an RM. In any case, an RTM is more valuable than an ordinary test material, since it has been checked for homogeneity and stability in terms of one or more specified properties. RTMs are extremely useful tools in intra- or interlaboratory development of methods for which reference materials are not (yet) available.

Table 2 compares the essential characteristics of representative test materials with those of certified and non-certified reference materials. As seen from table 2, it is assumed for RMs as well as RTMs that the material is representative for the class of materials that are or will be subject to testing with a selected method. For the RMs, homogeneity and stability are assessed and demonstrated for the measurands corresponding with the method that will be evaluated with the reference material. In the case of RTMs, homogeneity and stability for the measurands of interest are implicitly assumed based on data from other methods.

Table 2 Essential characteristics of the new concept ‘representative test material’ compared to the existing concepts of reference material and certified reference material
Full size table

The authors have proposed the term RTM for use in a document being prepared by ISO/TC 229 (Nanotechnologies), where it is being discussed in the Joint Working Group 2 Measurement and Characterisation (ISO/AWI TS 16195), but the concept is not unique to nanomaterials. An example of a material that corresponds with the RTM concept is the IRMM-481 Peanut Test Material Kit (IRMM 2005), which consists of carefully prepared peanut powders, with a homogeneous particle size and amount per vial. This material was developed for use in the research to identify and quantify allergenic proteins in peanut matrices (Scaravelli et al. 2009). The product information sheet (IRMM 2005) clearly states that the material is not a reference material, but it does not give an alternative denomination.

Representative test materials for nanomaterials testing

Obstacles to the production of reference nanomaterials

There is a lack of comparability of reported results on nanomaterials which have the same elemental composition (silver nanoparticles, carbon nanotubes, colloidal silica, etc.). Clearly, a nominal or purely elemental composition, without structural and other characteristics, is not sufficient to fully describe materials, let alone nanomaterials (Warheit 2010; Johnston et al. 2009, 2010a, b). For example, the surface structure of nanoparticles may be affected by a different history of environments in which the material has been; or the particle size distribution of a particulate nanomaterial may be different from one batch to another. It has e.g. been recognised for carbon nanotubes (CNT) that certain aspect ratios could give the CNT asbestos-like properties, which are absent for other aspect ratios (Poland et al. 2008). Related to this last example, there are a number of open questions related to hazard assessment (Aschberger et al. 2011) regarding for example (1) which parameters should be tested for nanomaterials, (2) how to express the doses for the effects data (e.g. in terms of total mass or mass fraction, particle number, specific surface area), (3) which methods and test guidelines to use, (4) and which test conditions to use. Since there is no agreement yet on which measurands would be the most relevant ones, development of standardised methods and corresponding reference materials is hampered. As a result, only few reference nanomaterials have been produced (Roebben et al. 2011) other than some reference materials for particle size (NIST 2011; Linsinger et al. 2011). In the absence of a clear understanding and agreement on which would be the relevant characteristics of a suitable reference nanomaterial, and on which methods to use for its characterisation, representative test materials will have to suffice.

Representative test materials for OECD WPMN

In order to assess the relevance and reliability of applying existing OECD test guidelines, which were developed for the testing of the safety of chemicals, to nanomaterials, the OECD WPMN initiated a vast collaborative project, called ‘OECD WPMN Safety Testing of a Set of Representative Manufactured Nanomaterials’. In this project, a set of data will be generated for 13 commercially important nanomaterials: carbon nanotubes (single wall and multiwall), fullerenes, dendrimers, nanoclays and nanoparticles of aluminium oxide, silver, titanium dioxide, cerium dioxide, zinc oxide, silicon dioxide, iron and gold. The collective development of data in the first phase will support evaluation of the methods’ applicability to nanomaterials for physico-chemical characterisation and (eco) toxicological testing. The tests performed will be the basis for discussions on if and how existing OECD methods should be adapted to nanomaterials and the need to develop new methods. The Guidance Manual for Sponsors (OECD 2010) describes in detail how to address each endpoint agreed for the project. The concept of ‘representative test materials’ was implicitly adopted by this testing programme by recognising that using a common source of nanomaterials would exclude one of the main variables in (eco) toxicity testing of nanomaterials performed by different laboratories. Furthermore, by making these nanomaterials available for testing also in additional projects, a knowledge base on those materials is being built, paving the way for using them later to better correlate different properties and effects of the tested nanomaterials.

In order to assist the testing under the OECD WPMN, the IHCP set up the JRC repository of representative nanomaterials. The nanomaterials in the NM-series are made from portions of industrial materials, produced to commercial specifications, and each series comes from the same production batch. The further processing of the industrially sourced base materials into the NM-series of materials ensures that the material has been homogenised, before being sub-sampled into vials, under GLP conditions. For example, the between-vial heterogeneity was explicitly assessed in terms of the primary particle size for the zinc oxide materials NM-110 and NM-111, which are dry powders (Singh et al. 2011). Based on the evaluation of SEM measurements, the materials can be considered as homogeneous (between-bottle standard deviation less than 4 %) with regard to the Feret’s diameter of the primary particles. Also, the stability of the sub-sampled materials is monitored.

Another example is a colloidal dispersion of silver nanoparticles (NM-300). The amount of silver (measured with UV–vis), the number of particles and the size distribution for samples before storage and after 12 months of storage are statistically not different, demonstrating the stability of this aspect of the NM-300 material under the specified storage condition (Klein et al. 2011). Thus, with homogeneity and stability being assessed and monitored for at least one relevant material property, all vials from the NM material are similar in terms of the monitored parameter. Having shown that the material is homogeneous and stable for a certain parameter, the RTM concept assumes that this holds true also for other properties. As a consequence, if between-vial differences in test results are obtained with other methods or for other properties, they should not be attributed lightly to the heterogeneity of the test material. This is the essence of the RTM concept.

The JRC nanomaterials repository hosts many of the nanomaterials which are tested within the OECD WPMN and provides them to the participants wishing to investigate one or more aspects of the materials (IHCP 2011). Currently, the repository hosts multiwalled carbon nanotubes (4 types), nanoclay (1 type) and nanoparticles of silver (1 type, two batches), titanium dioxide (6 types), cerium dioxide (2 types), zinc oxide (2 types) and silicon dioxide (5 types). The nanomaterials can be provided to researchers looking for materials, which are well characterised and for which tests are being performed for endpoints potentially relevant also in a regulatory context. The knowledge base on these nanoscale representative test materials is increasing through the tests performed. It is conceivable that, for a number of the assessed properties, materials from the NM-series may become reference materials. The reference material status may include selected (eco) toxicological endpoints, an area where RMs have not been developed in any significant number.

Summary

The current paper clarifies the meaning of three categories of test materials: reference material (RM), certified reference material (CRM), and representative test material (RTM). The new term RTM is put forward in this paper to close a gap in existing terminology. RTM is proposed to cover those materials that are used as a benchmark, for example in inter- or intralaboratory development of new or modified methods, for which the homogeneity and stability studies required for a reference material cannot be or have not yet been performed.

These three types of materials (RTM, RM, CRM) cover the whole range of test materials typically needed in the development, validation and routine use of measurement and testing methods. The JRC is actively involved in work related to all three types of materials in the field of nanotechnology, as the IHCP is providing representative test materials to the OECD WPMN and the IRMM is producing both certified and non-certified reference materials for nanotechnology. Thus, the two institutes contribute to providing the measurement and testing community with the means to ensure the reliability of test results on nanomaterials.