European Observatory for Nanomaterials

Since 2020 REACH has made explicit that nanoforms have to be registered. The registrants are required to provide a specific hazard dataset for each nanoform or set of nanoforms and to characterised them. This includes particle size distribution and other physical-chemical properties. Moreover, all toxicological (considering health effects) and ecotoxicological testing of chemicals (considering effects on biotic systems; such as OECD Test Guideline (TG) 201, 202 and 203) need to be adjusted to the behaviour of nanomaterials (see OECD Guidance Document (GD) 317). However, to do appropriate adaptations to the behaviour of nanomaterials the dispersion stability (OECD Test Guideline (TG) 318) and the dissolution rate (no OECD TG available; based on OECD GDs 29 and 318) should be known. Scientifically sound studies are necessary, especially if no TG is available.

Important points to consider during dispersion stability studies are for example,

1. I) For various nanomaterials the predefined particle number concentration results in a low mass concentration, which is below the limit of quantification of the available analytical methods. Especially, since we want to measure the test concentration over 6 h considering agglomeration, thus we need to be able to validly measure from 100% to at least 10% of the initial concentration. The example (Box) is showing a hypothetic material where the problem could arise. To avoid the described problem, the highest allowed particle number concentration should be aimed for such materials.
2. Example:
3. A material with a high density (approximately 7 g/cm3) and a particle diameter of 10 nm, needs to be applied at a mass concentration of 2-20 µg/L, since the TG 318 prescribes a particle number concentration between 0.5 and 5x1012 particles/L (Excel tool).
   If we aim for the higher concentration in the prescribed range, 10% of the initial concentration are approximately 2 µg/L, which is depending on the element well above the limit of quantification of ICP-MS (in case of inorganic nanomaterials). However, the nanoparticles might consist of multiple elements such as TiO2 or Fe3O4, hence the mass concentration of the element to be measured by ICP-MS is even lower than the concentration of the nanomaterial as such. In addition, it needs to be noted that we need for ICP-MS measurements a certain volume of sample (which is depending on the method) above 0.5 mL (the amount the OECD TG considers to be taken from the test vials), that causes further dilution.
4. Furthermore, the dilution for ICP-MS should be minimized as much as possible. Before even starting the study considerations and calculations about the analytical method and its limitation are very important.
5. II) To setup the study and conduct the required samplings takes a lot of time, that means the study can't be performed during normal working hours. The set up of the screening takes at least 2 h (1h in only pH equilibration in the test vials). Usually it takes longer, mainly to adjust all pH values. After set up, the test runs for 6 h and is followed by centrifugation and the last sampling. Depending on the nanomaterial, even the measurement of the concentration or sample conservation needs to be conducted the same day. In sum we are above 8 h (considering normal working hours). In fact, the amount of work to perform the test easily doubles if we take into account the time necessary for the set up and the need to take the samples every hour. That often leads to sampling during night times (just an observation based on experience), this needs to be considered, to avoid troubles. Here it can help to setup the stock dispersion the day before. 

For dissolution rate studies it is important to know, that 

1. I) The dissolution, of course, depends a lot on the medium used in the study. A simple medium gives good indication, but for more detailed results various different media should be used. The dissolution (rate) of a nanomaterial can be influenced for example by organic molecules (of various sources in test media of ecotox/tox tests) in the medium. If they coat the nanomaterial the release of ions can be lowered. Ions in the medium might form insoluble complexes which can bias the measurements of dissolved ions. The pH value highly influences the dissolution rate of multiple materials, therefore at least 3 pH values should be considered.
2. II) The chosen ultrafiltration device plays a significant role, and a blocking of the pores needs to be taken into account during planning. For dissolution rate studies it is recommended to use ultrafiltration devices that are able to hold back nanomaterials above 2-3 nm. If the investigated nanomaterial is a platelet with only one dimension in the nanometre range these devices can get blocked. The results might slightly differ depending on the nanomaterial and on the pore sizes of the membranes used: for example some small particles might pass those membranes with bigger pore sizes while those with smaller pore sizes will hold them back. Therefore, it is important to specify the used membranes in the report.
3. III) Potentially, dissolved nanomaterials can adsorb to the membrane or the tube after ultrafiltration, biasing the result of the study hence causing under estimation of the dissolution. For metal/metal oxide nanomaterials, some ideas exist to work around, but for organic nanomaterials it's much more challenging. For metal ions it is known that acidification lowers/eliminates adsorbance to sample vials. 1% HNO3 in the sample is usually enough to stabilize the ions in dissolution. To allow direct acidification an appropriate amount of concentrated pure acid can be pipetted to the sample collection part of the ultrafiltration device. However, for organic/carbon nanomaterials acidification might not help and solvents can potentially destroy/dissolve the plastic sample container (thorough investigations needed).
4. IV) The adsorbance to the membrane itself can be assessed by filtering some (specific or random) samples twice. If the concentration after second filtration is as high as after the first filtration no adsorbtion to the membrane needs to be considered (important: never acidify a sample prior filtration, that can influence the dissolution and adsorbtion!). The double filtration test might at least show if there is adsorbance or not for inorganic and organic nanomaterials. 
5. V) For carbon nanomaterials (for which in general only total organic carbon analyzer measurements can be performed) another problem occurs, the membrane releases carbon in relative high amounts, even after thorough rinsing with pure water. This background is even after thorough rinsing of the membranes with pure water approximately 0.5 mg/L, thus low release of carbon from the nanomaterial can't be detected.
6. VI) Complex test items (e.g. core-shell nanomaterials, minerals, organic molecules) might consist of different chemical elements (or dissolvable fractions) with different dissolution rate. This means all elements (or dissolvable fractions) should be measured if possible. Not for all elements a measurement is possible.

Although the guidance documents, as well as the available test guidelines, help laboratories to find an appropriate way to generate the required data for REACH registrations, there might be big variations between laboratories regarding the study design. Moreover, the laboratories take the responsibility to conduct scientifically sound studies until a practical TG is available. The practicability of TGs is of high importance. Since the studies need to be feasible at many labs and the results should be highly comparable on an interlaboratory level. That's why round robin tests (interlaboratory test performed independently several times) are performed during TG development. Historically testing started with metal and metal oxide nanomaterials mainly due to practical reasons and therefore the development of TGs for organic nanomaterials with all the challenges they represent is currently still lacking behind. That leads to open questions (and sometimes inapplicability) if other materials need to be assessed for registration. Of course, not every potential issue can be clarified during round robins and nanomaterials are a very broad group of materials with different properties and challenges, therefore it seems to be of high importance that these studies are conducted by experienced scientific personnel.
 

To leave a comment on this Nanopinion we welcome you to join the conversation in EUON LinkedIn discussion group.