nanomaterials

2. How can the characteristics of nanomaterials be described and analysed?

• 2.1 How are nanomaterials defined?
• 2.2 Which are the important physical and chemical properties of nanomaterials?
• 2.3 How are nanomaterials detected and analysed?
• 2.4 How are nanomaterials prepared for biological testing?

2.1 How are nanomaterials defined?

Although nanomaterials themselves are covered by the definition of substance within the REACH legislation (Regulation (EC) No 1907/2006) (European Commission 2006), currently the definition of what is "nano” is still under debate. Various organisations have proposed definitions of nanoscale using an upper limit of about 100 nm. It should be noted that most currently proposed definitions use the size of the primary particle/structure as a starting point. However, when a nanomaterial is in particulate form, the particles may be present either as single particles or as agglomerates or aggregates. Depending on the nanomaterial, the majority of the particles may even be agglomerates or aggregates. This may lead to the misinterpretation that agglomerates or aggregates of nanoparticles that have external dimensions well beyond 100 nm are not considered nanomaterials. Yet they retain specific physicochemical properties characteristic of nanomaterials, most likely due to their large specific surface area (SSA). The uncertainty regarding the presence of nanomaterials (either determined by size, <100 nm, or SSA >60 m2/g when calculated for <100 nm unit density spheres) in products becomes of major importance when the only information on the presence of a nanomaterial relies solely on the information provided by the manufacturer. Currently, it is frequently not possible to evaluate the nanomaterial contents of these products when the nanomaterial in question is mixed into a complex matrix of the finished product. This unresolved issue occurs in consumer products, particularly cosmetics and health care products, and also in food and feed products. All of these products contribute to the current exposure of the European population.

When describing a nanomaterial it is therefore important to describe not only the mean particle size but also the size of the primary particles. In addition, information on the presence of agglomerates and/or aggregates should be presented. When the mean particle size deviates (i.e. is larger) from the primary particle size this would indicate the presence of agglomerates/aggregates. This information should be included in the description of the nanomaterial and/or the product containing the nanomaterial. In addition to size, the specific surface area is a good metric to describe particulates. The specific surface area as determined by the BET method (Brunauer et al. 1938) has the advantage of being independent of the primary versus the agglomerated state.

Scientific toxicological data suggest that the total surface area of nanoparticles is a reasonable metric to describe toxicological responses in biological systems. The total surface area should not be confused with the specific surface area (SSA) where smaller particles have a larger SSA independent of whether they are present as primary, agglomerated or aggregated particles.

2.2 Which are the important physical and chemical properties of nanomaterials?

The main parameters of interest with respect to nanoparticle safety are: Physical properties

• Size, shape, specific surface area, aspect ratio
• Agglomeration/aggregation state
• Size distribution
• Surface morphology/topography
• Structure, including crystallinity and defect structure
• Solubility

Chemical properties

• Structural formula/molecular structure
• Composition of nanomaterial (including degree of purity, known impurities or additives)
• Phase identity
• Surface chemistry (composition, charge, tension, reactive sites, physical structure, photocatalytic properties, zeta potential)
• Hydrophilicity/lipophilicity

When nanomaterials are used in test systems, one has to be aware that some of the properties which need to be determined are largely dependent on the surrounding media and the temporal evolution of the nanomaterials. Thus, a primary focus should be to assess the nanomaterials in exactly the form/composition they have as manufactured, and in the formulation delivered to the end-user or the environment if the formulation contains free nanoparticles. nanomaterials can exist as nanopowders; suspended in air (ultrafine particles, nanoparticles, aerosols), suspended in liquid (colloids) and incorporated in solids. For biological safety evaluation, manufactured nanomaterials need to be dispersed in an appropriate media. The interaction between these media and the nanomaterials can have a profound influence on the behaviour of the suspension.

With the increasing number of newly emerging manufactured nanomaterials the importance of the potential dissolution kinetics needs to be emphasised. Since dissolution kinetics is frequently proportional to the surface area, nanomaterials are likely to dissolve much more rapidly than larger sized materials. This applies e.g. to silver nanoparticles which are increasingly used for their release of silver ions as anti-bactericidal agents. Yet the dissolution kinetics is not properly studied. The example of silver nanoparticles highlights the complexity of risk estimates of nanomaterials since adverse interactions of the silver nanoparticles with biological systems need to be distinguished from those interactions of the ionic silver. It should be emphasised that not all properties can be determined in every situation, nor is it necessary to do so.

2.3 How are nanomaterials detected and analysed?

Methods for the assessment of nanoparticles in the air (aerosols) and suspended in liquids or fluids have been further developed, and new methods have become available. Notably, similar to most advanced chemical analysis many of these methods involve research grade instruments requiring trained operating personnel and are not always straightforward to use in typical ‘public health’ settings. On the other hand, mobile and portable/handheld equipment is also becoming available, and an increasing number of studies have been performed and published in recent years (Mordas et al. 2008, Smith 2004). However, the wealth of these studies relates to the background of atmospheric nanoparticles, and little work in the context of manufactured nanoparticles has actually appeared. Furthermore, there is still a deficiency of comparable, reproducible and repeatable harmonised protocols for measuring and characterising nanomaterials (SCENIHR 2006). The ability to provide more routinely operable instruments, together with optimised protocols is important for providing meaningful and valid data that are comparable, reproducible, and repeatable, and which can produce a system of reliable risk identification, assessment and management. This requires defining the metrics most appropriate for hazard characterisation and exposure, including the methodology to perform the measurements. For a broader overview on the full portfolio of available methods for nanoparticle detection and analysis, the reader is refered to SCENIHR (2006).

For the measurement of particles in air, a number of methods are available. They have been developped since the 1980’s, are very reliable and often highly sensitive, but sometimes costly. Depending on the physicochemical parameters of a nanomaterial and including off-line analyses, many companies provide instrumentation able to characterise airborne particles down to the nanometer range. Experience is also available in the field of electron microscopy and microanalysis of nanoparticles in tissue sections and precipitated on substrates (e.g. Geiser et al. 2005). Other measurement methods, in particular optical techniques such as light scattering (Lindfors et al. 2004), can be applied to suspensions in various gaseous and liquid media and to solid matrices. The particle dynamics in suspension depend strongly on the medium of suspension.

Absorption and scattering microscopy of single metal nanoparticles allow for the tracking of nanoparticles suspended in the liquid phase (Van Dijk et al. 2006). This technology resulted in new equipment with the capability of optical tracking and identification of metal nanoparticles in fluids. The use of Condensation Nucleus Counters, well established in aerosol science, can now be routinely used to obtain information on nanoparticles, but is still unable to discriminate the particles from the background.

In an analytical sense, the most powerful method, real-time single particle mass spectrometry, has been further developed to provide a reliable method for the assessment of nanoparticles suspended in gases and liquids (by Electrospray Ionisation) with potential applicability to other fluids (Kane et al. 2001, Noble and Prather 2000). Here, a mass spectrum suitable for chemical analysis of the components of individual nanoparticles including the surface layer can be sampled and analysed. At least two commercial set-ups are currently available. All these analytical techniques have their specific reliability and sensitivity profiles and typically need to be combined to obtain reliable and specific assessments. Therefore, special consideration needs to be given to each methodology to verify the characterisation of the nanomaterials in the various phases. Typically for high performance analytical techniques, a number of generic issues need to be considered in the application of these methods for a specific case (e.g. accuracy, specimen preparation, role of substrate and presence of contamination).

2.4 How are nanomaterials prepared for biological testing?

A. The importance of dispersion

When manufactured nanoparticles are analysed in a clean sample which does not contain any other material, their physical-chemical properties can be studied (using the many instruments which are commercially available) with the level of precision required for their targeted production and testing. However, if nanoparticles are mixed within a matrix of different materials, as is the case for scientific and technological applications, consumer products and in toxicological and ecotoxicological samples, then it becomes exceedingly difficult to identify those nanoparticles since they may occur only in parts per 106 to 1012 of the surrounding matrix. In effect, the nanoparticles become "needles in the hay stack” which are extremely laboursome to find, identify and characterise.

It is well known from colloid science that nanoparticles can form agglomerates or aggregates, especially when they are kept as powder under dry conditions. This tendency to aggregate can create difficulties when testing the toxicity of nanoparticles. However, despite their tendency to aggregate, nanoparticles do not usually change their specific surface area. The total surface area is an important parameter for interactions with biological systems. Usually, a dry powder or a suspension in a water-based medium or some other fluid is used to administer the nanoparticle into the biological system. Several studies have made suggestions as to how best disperse the nanoparticles (Bihari et al. 2008, Buford et al. 2007, Sager et al. 2007). Best protocols may vary between the different nanomaterials. It seems obvious that there should be a best attempt to render the nanoparticle in a size that is relevant to the expected consumer/population exposure.

Dispersal methodologies suggested for particles using rational approaches include the use of albumin, a fairly bland and ubiquitous globular protein (Bihari et al. 2008), and lung lining fluid phospholipids (Wallace et al. 2007). Researchers must be aware that these coatings may alter the properties of the nanomaterial being tested and, therefore, the biological activity under consideration.

Synthetic detergents such as polyoxyethylene sorbitan monooleate (Wick et al. 2007) and Tween (Warheit et al. 2003) have been used to disperse nanoparticles for experimental purposes. Researchers must be aware that these additions may be toxic by themselves or act as an antioxidant (e.g. Tween). These additions should be taken into consideration when characterising the nanomaterials prepared for testing.

B. Reference nanomaterials, characterisation and test item preparation

"Reference material" (RM) is the generic name for the materials having a proven and sufficient homogeneity and stability in terms of a defined intended use. "Reference substance" or "reference chemical" are terms used in toxicology for materials that need to meet similar conceptual requirements but that are used for hazard identification, usually under GLP. Reference materials (RMs) need to be produced and used applying the conditions and terms standardised and described in ISO guides 30 to 35. When used in toxicology as test items, principles of OECD GD 34 and GLP apply mutatis mutandis (OECD 2005).

RMs can be used for different purposes, such as calibration, assessment of laboratory proficiency or test method performance. In toxicological assays for hazard identification reference substances/materials may be used for comparison with both positive (toxic) and negative responses. Currently, a small number of reference materials already exist in the field of nanomaterials and manufactured nanoparticles (e.g. gold nanoparticles from the National Institute of Standards and Technology (NIST), Gaithersburg, MD, USA and silica from the European Commission, Institute for Reference Materials and Measurements (IRMM), Joint Research Centre (JRC), Geel, Belgium). They are spherical model materials which are certified primarily for size and are used mainly to calibrate instruments which measure particle size. The absence of well-defined parameters to measure ("measurands") and of standardised test protocols is identified as a major obstacle for reference material production, because agreed and harmonised methods are needed.

A typical issue of information generated by measurements or studies is to combine a metrological part with knowledge about a reference material with an intended use in toxicological test systems under GLP. Toxicological test systems mimic routes (and scenarios) of exposure and typically require information about dosage. A study, examination or test, when successfully performed, generates a prediction of the effect of interest. In toxicological tests, reliability AND relevance both contribute to the overall predictive capacity and to the validity of a test for its purpose.

In practice, and in agreement with the requirements mentioned above, characterisation results should be obtained and used in their appropriate context scenarios. The information should be used for description of intrinsic and extrinsic properties. The metrological principles of the reference nanomaterials available so far cannot be used or extrapolated to toxicological tests and related results, but the information on the properties provides a reliable basis as starting point for the test and reference items used in such studies. The preparation and use of a reference material comprises two stages.

Stage 1 is the characterisation of the intrinsic properties of a reference nanomaterial, its stability and homogeneity. Physicochemical properties need to be determined. The physical state and preparation form of the material examined should thereby be relevant for production and use. Sample preparation steps corresponding to analytical sample preparation should be critically assessed with regard to being a determinant of the measurement result itself.

When a reference material is prepared for use in test systems for toxicological evaluation or environmental fate analysis, it will be brought into a matrix/media/vehicle depending on the type of test assays used. This comprises conditioning and choice of matrix components. The prepared test sample should thereby correspond to the requirements of the test method and preferably be representative for the identified exposure situation.

Stage 2 comprises the characterisation of properties following test sample preparation. Results depend on the protocol used and matrix components, which may be essential for a certain test system and form part of that test system. Several results may be gained for the same reference nanomaterial and its properties depending on the conditioning and matrix used.

Indeed, shape, size and surface area affect the hazards associated with nanomaterials, at least because these parameters affect the transport properties of the particles (absorption, distribution, and excretion). Reference nanomaterials have to be seen in their context of intended use. They are tools of Quality Assurance and method validation. They serve method harmonisation and standardisation, and performance assessment.