Nanotechnology risk assessment could benefit from nanoparticle categorization framework
(Nanowerk Spotlight) It has become quite a familiar pattern for regular readers of nanoscience and nanotechnology literature: on some days you see articles that slam certain nanomaterials, and nanotechnology in general, as inherently risky due to a new scientific paper that reports alarming levels of toxicity of a particular nanoparticle. On other days you find research papers that report no apparent risk whatsoever with the same nanomaterials. These reports then get cherry-picked by various interest groups to fit their agendas - the cosmetics industry for instance claims their products and the nanoparticles in them are perfectly safe; environmental groups contend that nanomaterials are inherently risky and that rigorous testing is needed before large scale commercialization of nanoproducts can happen; some even call for a complete moratorium on nanomaterial production.
The result is a situation that is confusing as hell not only for laypersons but also for industry, researchers and the journalists who cover the field. Fact is, that every new technology is inherently risky - plenty of people are being injured or killed every year by electricity, cars, chemicals, or nuclear energy, just to name a few. In order to reap the benefits of a new technology and make it acceptable to society there has to be a general perception that the risks are fully understood, can be managed and it is clear who is responsible for what.
All of that is currently missing in nanotechnology. Although the speed and scope of nanotechnology risk research - and the emerging field of nanotoxicology - is picking up, a lot of this work is stand-alone research that is not being coordinated within a larger framework. A group of Danish scientists argues that nanomaterials must be categorized based on the location of the nanoscale structure in the system/material before their hazards can be assessed and they propose a categorization framework that enables scientists and regulators to identify the categories of nanomaterials systematically.
As an emerging science, nanotoxicology is expanding the boundaries of traditional toxicology from a testing and auxiliary science to a new discipline where toxicological knowledge of nanomaterials can be put to constructive use in therapeutics as well as the development of new and better biocompatible materials (see our Spotlight: Toxicology - from coal mines to nanotechnology). Until now, though, no one has been able to pin-point which properties determines or influences the inherent hazards of nanoparticles.
"We believe that in order to reap the many benefits offered by nanotechnology it is necessary that environmental and human health risks are considered in an early stage in product development" Steffen Foss Hansen tells Nanowerk. "However, before this can be done there is a need for clarification of terminology, e.g., the current literature addressing the potential hazards of nanomaterials show a strong tendency to use the terms ‘nanotechnology’ and ‘nanomaterials’ as synonyms of nanoparticles. Thus, the hazards related to nanotechnology/nanomaterials have so far predominantly been documented for specific nanoparticles, mainly TiO2 and carbon-based nanoparticles. However, the physical, chemical and biological properties of various nanomaterials differ quite substantially from that of specific nanoparticles, as do the expected routes of exposure, making it necessary to differentiate nanomaterials in order to identify the potential hazards and risks they pose."
Hansen, a researcher in the Institute of Environment & Resources at Technical University of Denmark (DTU), together with colleagues from DTU, developed a framework that can be applied to a suggested hazard identification approach and is aimed at identifying causality between inherent physical and chemical properties and observed adverse effects reported in the literature.
"We generated a database noting the reported inherent physical and chemical properties of the nanoparticles tested and the main effects observed" Hansen explains. "428 studies were noted in the database reporting on a total of 965 nanoparticles. We found that although a limited number of studies have been reported on ecotoxicity, more than 120 and 270 have been reported on mammalian toxicity and cytotoxicity, respectively. In general there was a lack of characterization of the nanoparticles studied and it was not possible to link specific properties of nanoparticles to the observed effects. Our study shows that future research strategies must have a strong focus on characterization of the nanoparticles tested."
Hansen and his colleagues suggest that nanomaterials should be categorized depending on the location of the nanoscale structure in the system. This leads to a division of nanomaterials into three main categories, which then can be further divided into subcategories:
Materials that are nanostructured in the bulk;
Materials that have nanostructure on the surface; and
Materials that contain nanostructured particles.
The categorization framework for nanomaterials. The nanomaterials are categorized according to the location of the nanostructure in the material. (Reprinted with permission from Informa)
Hansen points out that it is possible for a system to consist of nanostructured elements belonging to different categories in this framework. He gives the example of car catalysts used to remove NOx from car exhaust: "The chemical reaction that removes NOx is catalyzed by Platinum and Ruthenium nanoparticles of 2-3 nm size. These nanoparticles are bound to the surface of a support material. This corresponds to a category IIIa system according to the above figure. At the same time, the support material is a nanoporous material consisting mostly of γ-Al2O3 (70-85%) and other oxides such as cerium oxide or lanthanum oxide. Thus, the support structure is a category Ib system. A similar analysis would apply to for example fuel cells."
A major benefit of the proposed categorization framework is that it provides a tool for dividing nanosystems into identifiable parts and thereby facilitating evaluations of, for instance, relevant exposure routes or analysis of effect studies according to relevance of the material tested.
Another dimension that needs to be considered in assessing the toxicity of nanomaterials are their physical and chemical properties. Today, it still is an open questions which properties determine or influence the inherent hazards of nanoparticles.
"After an initial literature review, and when considering the information needed in order to describe a nanomaterial from a physical and chemical perspective when estimating the hazard of nanomaterials, we propose the following nine properties as being important" says Hansen: "(1) Chemical composition, (2) Size, (3) Shape, (4) Crystal structure, (5) Surface area, (6) Surface chemistry, (7) Surface charge, (8) Solubility, and (9) Adhesion, defined as the force by which the nanoparticles and it components are held together."
The Danish researchers then went ahead and combined their categorization framework from above with this list of properties to construct a hazard identification scheme.
"Clearly, not all the properties apply to all the different categories and the above table summarizes our evaluation of the relevance of each property for each category is" says Hansen. "We distinguish three cases: (i) The property is relevant in determining the hazard; (ii) The property does not apply to the category; for instance, it makes no sense to talk about the surface chemistry of a solid such as Ia; and (iii) The property can be determined, but it is not relevant in order to determine the hazard of the nanomaterial of that category, because it is unlikely that the surface charge of nanoparticles suspended in a solid plays any role in the overall toxicity given the limited exposure potential."
Having constructed their model, the researchers then went on to test its workability by reviewing the literature on the potential hazards of nanomaterials belonging to two of the proposed categories: Nanoparticles suspended in liquids (Category IIIb) and airborne nanoparticles (Category IIId). Altogether, they identified 428 relevant studies which, in total, reported the observed adverse effects of 965 tested nanoparticles of various compositions. Hansen and his colleagues then analyzed each one of the 428 studies in order to identify which inherent physical and chemical properties were reported and the main effects observed. Then they filled the information into their hazard identification scheme.
The above table shows that there is a large difference in which characteristics have been reported in the literature. The only information that all the studies report is the chemical composition of the nanoparticles tested.
"In general there is a lack of characterization of the nanoparticles tested in the identified studies and this hampers the potential of identifying causality between observed hazards and specific physical and chemical properties" says Hansen. "Furthermore, it is evident from our work that the information provided is 'all over the map' making it impossible to systematically analyze the studies for properties of the nanoparticles of importance for the observed effects."
Although the lack of characterization is troublesome, it is hardly surprising as nanotoxicology is a very new field. Hansen points out that a true understanding of the hazardous properties that materials begin to exhibit at the nanoscale requires a level of interdisciplinary research that has not yet been reached.
"In order to conduct and interpret scientific studies on the hazardous properties of nanomaterials that are relevant for future risk assessment of nanotechnology-based compounds and products, we need strong interdisciplinary collaborations between (eco)toxicologists, and nanoscientists such as physicists, chemists, and material engineers."