Identification of Risks in the Life Cycle of Nanotechnology-Based Products


Address correspondence to:
Dr. Nathan Swami
Department of Electrical & Computer Engineering
University of Virginia, PO Box 400743
Thornton Hall C218
Charlottesville, VA 22904-4743


In order to realize the projected market potential of nanotechnology, the environmental, health, and safety (EHS) uncertainties posed by a nano-product (i.e., a nanotechnology-enabled product) need to be characterized through the identification of risks and opportunities in early stages of product development. We present a methodology to identify risks from nano-products using a scenario analysis approach that allows for expert elicitation on a set of preidentified use and disposal scenarios and what we have labeled “risk triggers” to obtain scores on their likelihood of occurrence and severity. Use and disposal scenarios describe product life-cycle stages that could result in risk attributed to the nano-product, whereas risk triggers are particular to nanoparticle properties. These are potential risks, as the risk assessment community is currently debating the specific risks attributed to nanotechnology. Through such a framework, our goal is to identify which products pose greater risks, where these risks occur in the product life cycle, and the impacts of these environmental risks on society. The comparison of risk triggers across nano-products allows relative risk ranking on axes of exposure- and hazard-related risk triggers. For the specific case of air fresheners, areas of acute risks resulted from bioavailability of nanoparticles in air release and water entrainment exposure scenarios; catalytic activity of nanoparticles in inhalation and air release exposure scenarios; the harmful effects due to the antibacterial property on useful bacteria particularly in susceptible populations; and, finally, risks from the lack of nanoparticle coating stability in air release scenarios.


The prefix nano is derived from the Greek word for dwarf, and 1 nanometer (nm) is equal to 1 billionth of a meter (10−9 m). Motivated in part by the scientific scope of nanotechnology as laid out by Richard Feynman in his 1959 lecture “There's Plenty of Room at the Bottom” (Feynman 1959) and its technological scope for a broad range of applications, the National Nanotechnology Initiative (2007), which has the charge of promoting nanotechnology research, defines nanotechnology as “the understanding and control of matter at dimensions of roughly 1 to 100 nanometers, where unique phenomena enable novel applications.”

The definition has two parts. One is the part about engineering at dimensions of 1–100 nm, and the other is about properties of materials at the nanoscale that enable their use for novel applications. The size range that holds so much interest is typically from 100 nm down to the atomic level (approximately 0.2 nm), because it is in this range that materials have radically different properties from their bulk counterparts. The main reasons for this change in behavior are an increased relevance of surface and interfacial phenomena, a greater level of biological activity, and the dominance of size-related quantum effects. For instance, an increase in surface area results in corresponding increases in chemical reactivity, making some nanomaterials useful as catalysts in chemical reactions that may be applied within fuel cells. At less than 10 nm, quantum effects begin to change the optical, magnetic, and electronic properties. Nanotechnologies aim to exploit these size-related effects to create structures, devices, and systems with novel properties and functions.

Although some applications of novel nanostructures are currently in basic research phases, other nanotechnology applications have now moved beyond scientific journals into the marketplace. On one hand are products that replace bulk ingredients with nanoscale counterparts, such as improved sunscreens, and on the other hand are products enabled through novel nanoscale phenomena, such as field emission properties of carbon nanotubes for ultra-light flat-panel displays in cell phones (Project on Emerging Nanotechnologies 2007), or quantum-dot-based1 lighting. In 2005, the worldwide public and private investment in nanotechnology totaled approximately $9 billion (President's Council of Advisors on Science and Technology 2005), and hundreds of nanotechnology-enabled products (Project on Emerging Nanotechnologies 2007) are available in the United States alone, with projected worldwide revenues of $150 billion in 2008 and $3 trillion by 2014; the nano-product marketplace is expected to grow tremendously. In this context, early-stage identification of environmental impacts and risks from nano-products is vital to their large-scale acceptance by society, and this requires studies to distinguish real risks from assumed risks without substantiated data. Prior attempts at formal environmental risk assessment have been stymied by a paucity of toxicological data (Maynard et al. 2004; Shvedova et al. 2003), thereby greatly hampering risk management efforts for nanotechnology workers and risk communication efforts with the public (Maynard 2006). Hence, there is a need for frameworks that allow for prospective analysis of the risks and opportunities that nanotechnology presents.

We present here a framework that modifies traditional risk assessment methodologies in order to assess the impacts of novel nanomaterial properties and to identify risks from nanotechnology-enabled products in the marketplace (hence called nano-products). The article first introduces our objective: to apply risk analysis to nanotechnology, elucidating the special properties of nanomaterials that make nanotechnology risk assessment a challenge. Nanomaterials are the nanotechnology-based materials present in nano-products or components of nano-products. This framework is based on the analysis of nano-products with consideration of the nanomaterial present.2 The following sections present our risk identification methodology. The framework uses expert elicitation with an input of nano-product information, use and disposal scenarios, and what we have called risk triggers, which are inherent nanoparticle properties that trigger a higher level of risk, to identify the substantive risks and map the affected population. The article illustrates how the model works for one example product, namely air freshener sprays. Finally, we discuss some future directions, such as the inclusion of regulatory and knowledge gaps within the risk identification framework. Through such a framework, our objective is to identify which products pose greater risks, where these risks occur in the product life cycle, and the impact of these environmental risks on society. Because this framework allows risk identification in conjunction with regulatory gaps, we anticipate that it will be useful in developing approaches to the risk-based nanotechnology regulation.

Challenges of Risk Assessment as Applied to Nanotechnology

Risk is defined as a measure of probability and severity of adverse effects (Lowrance 1976). Risk is a functional combination of likelihood and severity of the effect, leading to the following functional definition: risk =f(hazard, exposure). Risk assessment methods use quantitative models based on empirical research and thus do not attempt to assess risk with absolute certainty. Most risk assessment methodologies employ models that can be fitted to the studied system, and for nanomaterials, models for similar bulk materials may be used. These mathematical models themselves come with a set of their own uncertainties due to model parameters (parametric uncertainty) and as a result of unclear or unidentified relationships between model variables (model uncertainty; Kandlikar et al. 2006). Thus, the aim of risk assessment is not to arrive at a single risk or adverse effect estimate but to allow a characterization of a range of possible consequences.

The maximum degrees of freedom for managing a new technology occur at the design and development phases. Earth systems engineering management (ESEM) is a set of principles that grew out of industrial ecology (Allenby 1999; Graedel and Allenby 2003). A core principle of ESEM is reversibility: New technological systems should be designed so that they can be modified or even shut down if unanticipated negative impacts emerge. Reversibility has to be designed in from the beginning. ESEM therefore calls for adaptive management of coupled human–technological–environmental systems. In this case, adaptive management means continuously monitoring not just the local but also the systemic impacts of emerging technologies, throughout their life cycles, and making adjustments to the system based both on the environmental impacts and on responses from human stakeholders who are part of the system (Allenby 1999). Risk assessment, in contrast, works best after a technology has been on the market long enough for its risks to be quantified. At this later stage of development, society may be locked into systems that are difficult to reverse. Reversibility may be nearly impossible, and other forms of adaptive management may also be extremely difficult and costly.

In this article, we present a methodology to assess the potential impacts and benefits of nanotechnology as nanomaterials are invented and the first products emerge. At this stage, ESEM principles can be applied, provided information on the nanomaterials and technologies in new products is made available and tests are conducted under controlled conditions using appropriate methodological standards. Identification of risks from nanotechnology is already a topic of much interest to industry and academia (Maynard 2006; Balbus et al. 2007).

Why is an early-stage analysis particularly important for nanotechnology? First, risk identification is essential to, and the first major step in, traditional risk assessment. This leads to risk filtering, ranking, and management (Haimes 2004). Second, in the United States, in spite of calls for increased Environmental Health and Safety (EHS) research budgets for nanotechnology (Balbus et al. 2006; Maynard 2006), the most recent requested budget is less than 10% of the total budget for the National Nanotechnology Initiative (Subcommittee of Nanoscale Science Engineering and Technology of Committee on Technology of the National Science and Technology Council 2008). Early identification of potential risks can assist in prioritizing research when budgets are constrained for such research. These potential risks can be given more attention relative to an exhaustive but undifferentiated list. Finally, as stated above, adaptive management is an element of being able to identify risks early and adapt the regulatory system to accommodate the needed changes. At the same time, early risk identification can lead to redesign of systems and reversal of product development. Identifying risks early on is one tool for implementing adaptive management in such governance systems.

As a society, we need tools for identifying potential impacts as technologies emerge, so, at the very least, data can be collected and system impacts monitored. This framework can be considered a first effort for assessment of the potential risks of nano-products. Such analysis can be fed into “design for the environment” programs for emerging technologies.

Some of the challenges that nanotechnology presents for formal risk assessment are summarized below, to illustrate the need for risk identification early in the product life cycle to begin the process as stated above:

  • 1Environment, health, and safety information on nanomaterials and nanotechnologies is scarce. The first issue is related to the hazard that nanomaterials may pose due to reactions at the cellular level, especially in light of their apparent similarity in size, shape, and chemical form to known offenders such as asbestos and polychlorinated biphenyls (PCBs). This hazard is further exacerbated by a paucity of toxicological data, especially for studies using standardized protocols (e.g., biological system to be studied) with technologically relevant nanomaterials (e.g., uniform nanomaterials in their product form with the necessary coatings and isolated from catalysts used in their synthesis) within their likely exposure pathways (e.g., ingestion of water-entrained nanoparticles from sunscreens; Friends of the Earth 2006). Formal risk assessment may have to wait until some of these hazards are better characterized.
  • 2There is no single index to measure the toxicity of nanomaterials. The interpretation of toxicology results is further complicated by the lack of one single index to comprehensively describe toxicity. Traditionally, dose–response curves served as that index for most chemicals. For nanomaterials, the response may not always increase predictably with dose. This is because chemical toxicity is not determined by one factor alone, such as size. Other factors, such as surface charge, shape, aggregation, structure, area, and stability of coatings, affect reactivity, uptake, and transport, thereby differentially affecting response over different dose ranges (Warheit et al. 2006). Early risk identification may assist in the formulation of potentially useful indexes for characterizing toxicity.
  • 3Nanotechnology nomenclature is still in development, with no shared understanding. Nanotechnology is composed of nanomaterial components that span various classes: organic, inorganic, ceramic, and even biological in some cases. This makes their classification, especially within existing regulatory frameworks such as the U.S. Toxic Substances Control Act (TSCA), exceedingly difficult, thereby confounding regulators (Wardak et al. 2007).
  • 4The ready environmental transport of nanomaterials increases chances of exposure. Nanomaterials can be easily distributed throughout an ecosystem due to their size and solubility and hence present an increased risk of exposure. Even in cases where they may be aggregated, they may represent a problem (Fortner et al. 2005). The higher mobility of nanomaterials in the environment implies a greater potential for exposure as they are dispersed over greater distances and their effective persistence in the environment increases (Wiesner et al. 2006).
  • 5Nanomaterials are not easily monitored in real time. Due to their size, nanomaterials cannot be easily detected with an optical microscope and are difficult to monitor in real time. Cheap and easy-to-use monitoring technology is not in place. For instance, technologies such as electron microscopy require extensive sample preparation, whereas spectrometric methods do not give sufficient 3D information at the nanoscale. Monitoring, characterizing, and measuring nanomaterials are critical to understanding human and environmental exposures to nanomaterials and are thus important for occupational health. At present, technologies for these purposes are not readily available, but they may become more available over time (U.S. Environmental Protection Agency 2007a).
  • 6Nanomaterials may have system-level human health and environmental risks, where attention to one set or system overlooks the greater impacts. Nanotechnology's societal effects are not localized to one community or industry. Nanotechnologies designed for application in one particular environment can easily be transmitted to other environments, where there may not be sufficient risk data and research. For instance, silver nanoparticles in air fresheners have antibacterial properties. They are intended for use in indoor air, and it would be easy to envision human inhalation. But it is not clear that they have been studied for digestive system effects. Risks from such system-level impacts on human health across organ systems could be more worrisome with active nanostructures in the near future. Similar system-level impacts are possible across the environment (Swiss Re 2004; Davies 2006).

Uncertainty can be addressed through early risk identification. The goal of risk identification is to identify possible hazards and to estimate the likelihood and magnitude of exposure. Hazards can be defined as those materials or processes that produce toxic effects in humans or the environment or that have potential adverse effects, such as causing fires or explosions. The characterization of hazard includes quantification of toxicity through testing and determination of specific biological response. Research on exposure must evaluate whether, and to what degree, exposure will occur for each nanomaterial at each stage of its life cycle (Subcommittee of Nanoscale Science Engineering and Technology of Committee on Technology of the National Science and Technology Council 2006). Extrapolations from bulk materials can probably be made for smaller particles of well-known materials, but even that should be done on a case-by-case basis (Lux Research 2006). Nanomaterials present new exposure scenarios, given that small particles reach areas of the human body and regions of the ecosystem that large particles cannot. With this uncertainty in mind, the following framework was developed to help identify and analyze risks with expert elicitation.

Research Methodology and Framework for Risk Identification

Data about particular risks attributed to nanotechnology are scarce for the application of traditional risk assessment methods; thus, this risk framework builds on expert opinion and existing research and data, as depicted in figure 1. We developed use and disposal scenarios and risk triggers on the basis of a broad literature review. We solicited expert opinion on these scenarios and triggers for a set of nano-products. We developed the best nano-product information on the basis of publicly available data. Experts scored these scenarios and triggers for each nano-product. Taking these data into account, we looked for exacerbation of these scenarios and triggers when they overlap. This resulted in a final priority list of potential high-risk topics, or “hot spots,” for each nano-product. In this article, we only review the example of the air freshener. Additionally, identification of the hot spots also provided for a relative risk ranking of products, whereby we could hypothesize that some products are of higher concern for risk than others.

Figure 1.

Risk identification framework in use in this article.

Scenario Analysis

We started with a set of nano-products, as depicted in table 1, available in the marketplace and obtained complete information about the product composition and use (Project on Emerging Nanotechnologies 2007). The nano-products were classified into three groups on the basis of the nanomaterial present in the nano-product—metals/metal-oxides, carbon nanotubes, and fullerenes. The nano-products could be further classified as passive, where the nanomaterial's properties are not turned on or off (e.g., sunscreen and disinfectants), or active, where the nanomaterial's optical, electronic, or quantum-level properties can be actively modulated in some manner.

Table 1.  Summary of nano-product information given to experts
Nanomaterial typeProducts/usageNanomaterial compositionNano-enabled functionNanomaterial production
  1. Note: UVA = ultraviolet light, type A; UVB = ultraviolet light, type B; CNT = carbon nanotube.

Metal/metal oxide (passive)Sunscreen/applied on skin40% TiO2 Mn doping Polymer matrix (proprietary)Absorbs UVA and UVB rays without the creation of free radicalsPhysical vapor synthesis
Toothpaste/applied onto teeth∼7 nm size colloidal Ag Polymer matrix (proprietary)Antibacterial 
Air freshener/sprayed20–50 nm Ag polymer matrix (proprietary)Antibacterial 
BatteryLi4Ti5O12Intercalates lithium ions 
Food supplementCa and Mg nano-powderBetter absorption of Ca and Mg in blood 
Carbon nanotube (passive)Tennis racquet0.7 nm–7 μm CNT polymer matrix (proprietary)Increased toughness and decreased weightArc discharge—separate CNT from the product
Field emission displayCNT, small quantities of catalyst (Fe, Ni, Y)High electric field emissionArc discharge chemical vapor deposition
Metallo-fullerenes (active)MRI contrast agentsC80 (fullerene) cage enclosing magnetic lanthanides (Gd, Ho, etc.)Nanoscale magnets that enhance MRI contrast and can be specifically targetedArc discharge with nitrogen to synthesize endohedrals

Our methodology was to develop questions for expert elicitation. These questions were refined iteratively on the basis of preliminary interviews. Scenarios through which human health or the environment may be impacted by the nano-products during their use and disposal are called use and disposal scenarios, as in table 2. Though the exposure scenarios in use and disposal are the same, they are considered in different product life-cycle stages because the environmental and human health implications can be different in each stage. For use, we focused on the intended use of the nano-products. Disposal scenarios are viewed as the intended end-of-life of the nano-products.

Table 2.  Use and disposal scenarios reviewed by experts
Use and disposalScenarioDescription
UseInhalationInhalation of nanoparticles in the product during use
Skin absorptionAbsorption of the nanoparticles into skin
IngestionAccidental ingestion of nanoparticles in the product during use
Water entrainmentEntrainment of nanoparticles in water system or the sewer during product use
Air releaseRelease of nanoparticles in the air during product use
DisposalInhalationInhalation of nanoparticles in the product when it is disposed
Skin absorptionAbsorption of the nanoparticles into skin
IngestionAccidental ingestion of nanoparticles in the product during disposal
Water entrainmentEntrainment of nanoparticles in water system or the sewer during product disposal
Air releaseRelease of nanoparticles in the air during product disposal

Nanomaterial characteristics can differ radically from those of their bulk counterparts, if such counterparts exist. Certain properties within the nano-product life cycle, as shown in table 3, can trigger a higher risk potential. Our methodology aims to enable a risk-based scoring of the triggers for each nano-product, and they are considered separately during the expert elicitation process. The particular nanomaterial properties that make them different from bulk counterparts are called risk triggers. For purposes of prospective risk identification across various nano-products, rather than using use and disposal scenarios alone, experts were more comfortable rating the role of risk triggers within each of the products. This approach was added due to the paucity of direct scientific data for use and disposal scenarios, whereas triggers were based on known nanomaterial properties within the products. The list is divided into exposure-related and hazard-related risk triggers. Certain nanomaterial properties can cause increased exposure to these substances, and certain nanomaterial properties can cause increased severity once exposure has occurred.

Table 3.  Hazard-related and exposure-related triggers reviewed by experts
Exposure-Related Risk Triggers
New Product– Is this a previously unavailable product application enabled through the nanomaterial? A new product application can potentially be more risky because of the unknown exposure scenarios or because it may potentially use more nanomaterial than an application that uses nanomaterials to enhance an existing application, thus increasing the exposure.
Coating Stability– Are there scenarios where the coating of the nanomaterial breaks down? If the coating for a nanomaterial breaks down during usage or through disposal, this may increase the likelihood of nanotechnology-related risks.
Media-Dependent Properties– Do the nanomaterials in the nano-product behave or act significantly different in different media (air, soil, water)? If the nano-product changes significantly in different media outside its intended application, then the exposure likelihood increases.
Used in Conjunction With Other Products– Synergistic effects that arise due to interaction of one product component with another might give rise to a higher risk. Is the nano-product used in conjunction with other products that may raise such exposure pathways?
Multiple Disposal Pathways– Is the product disposal pathway (e.g., recycling, trash, incineration) fixed? Multiple pathways to disposal create more risk scenarios and further consideration.
Particle Size Under 200 nm– Is the nanomaterial present in the nano-product under 200 nm in size? At sizes below 200 nm, surface phenomena dominate the exposure pathways due to high surface-to-volume ratios.
Dispersibility and Bioavailability– Does material become dispersed as free nanoparticles, or is there a coating that makes it less dispersible? If it is dispersed as a free nanoparticle, this increases its bioavailability.
High Aspect Ratio– Does the aspect ratio cause the material to be more easily transported in the environment? Fiberlike structures may increase their environmental mobility.
Hazard-Related Triggers
Exists Only in Nano Form– Is the material engineered only at the nanoscale? The implied assumption is that if this material exists only in the nanoscale form (e.g., carbon nanotubes, fullerenes) then less information is available on hazard and exposure pathways.
Aggregated Nanoparticles– Are properties of the free nanoparticle different from aggregated forms of the nanoparticle (e.g., solubility)? Is the aggregated form of the nanoparticle as hazardous as the free nanoparticle? This suggests that further research is needed into the nanomaterials' properties for aggregation.
Photocatalytic and Other Catalytic Activity– Do the nanoparticles in the product catalyze chemical or photochemical processes leading to adsorption, transport, or generation of hazardous substances (e.g., free radical generation in the presence of sunlight)?
Susceptible Population– Are there scenarios during product use where one population demographic is more negatively affected? The degree of this possibility will dictate further research.
Antibacterial Properties– Does the nano-product contain materials that can kill/harm useful bacteria in the environment or the human body? Although bacteria can cause infections, they can also be good for the environment.

Expert Elicitation

In risk assessment situations where information is lacking, expert elicitation is often used to fill in the information gaps (M. G. Morgan et al. 1990; Martin et al. 2004; Zilinskas et al. 2004; K. Morgan 2005). In this study, we interviewed a total of eight experts in five fields—environmental sciences, toxicology, chemistry, material sciences, and technology policy. These experts' interviews were scored, as explained below, and another set of ten expert interviews assisted in early development of the use and disposal scenarios and risk triggers. Each field contributed a specific skill-set toward the expert elicitation process. A cross-section of experts from the government, industry, and academia were included.

A survey was sent out to the experts in advance with the product information, the use and disposal scenarios, and the risk triggers for each product. Using the set of described triggers, experts were drawn into discussions on the relative importance of the triggers and how they may apply to particular nano-products. In this manner, the likelihood of scenarios, including bounds and range of values of important parameters within the scenario, was assessed. By avoiding a sole focus on scores, we were also able to understand the reasoning behind the experts' ratings (M. G. Morgan et al. 2002).

After the survey responses were collected from experts, the nano-products were scored in the following categories: scenarios and triggers. The scenario scores were obtained from the responses to table 2. Every scenario was given a rating of high, medium-high, medium, medium-low, or low, considering the probability and severity of the scenario, and these ratings were converted to scores from 5 to 1, respectively. This resulted in two different scores for every use and disposal scenario: hazard and exposure. This process provided us with a score for every scenario for each product.

From expert responses to table 3, the triggers were given a rating of high, medium-high, medium, medium-low, or low, considering the probability and severity for each trigger for each product. Once again, these ratings were converted to scores from 5 to 1, respectively. The risk triggers were averaged separately for all hazard-related risk triggers and all exposure-related risk triggers. The higher the hazard-related and exposure-related scores are, the higher is the potential risk from the nanomaterial present in the product.

In cases where the expert did not specifically offer a rating (high, medium-high, etc.) or a score (5, 4, etc.), we inferred his or her scores from the discussion. All experts were emailed a spreadsheet of their respective scores. If they disagreed with the scoring, they could edit their responses and email them back to us. It is important to note that this study only provides subjective judgment on the nature of risk, hazards, and exposure due to the nano-products. These judgments are not definitive, but they offer a first step in a risk assessment.

Risk Identification and Mapping

From the data entries to table 3, we can compare the expert responses and resulting scores of scenarios to the risk triggers. The scenarios identify use and disposal situations with potentially high or medium risks, and the risk triggers identify the particular properties leading to high or medium hazards and exposures. The expert responses that indicated a high or medium rating for a use and disposal scenario were used to identify potentially high-risk scenarios. The expert responses that indicated a high or medium rating for a risk trigger were used to identify potentially high-risk triggers. The intersections of these high-risk triggers and scenarios can be proposed as hot spots for future research, given that high-risk scenarios and high-risk triggers could have a multiplicative or exacerbating effect on each other. These hot spots would relate to particular nano-products and their associated nanomaterial.

The Case of the Air Freshener Spray

In this section we present a set of preliminary results from the application of our risk identification framework to air freshener spray, as an illustration of the methodology. For the use and disposal scenarios in table 2, the experts were asked to give scores for the exposure and hazard potential. As explained above, the scores corresponded with ratings: 5 = high, 4 = medium-high, 3 = medium, 2 = medium-low, and 1 = low. As explained above, the high-risk scenarios are identified as having both hazard and exposure scores of 3 or above.

For the air freshener spray, as shown in figure 2, the high-risk scenarios were identified as inhalation during use, skin absorption during use, air release during use, and water entrainment during disposal. Figure 2 is a visual depiction of these scores. For example, for air release during use, the experts' average rating was 4, or medium-high, for both exposure and hazard. This means the scenario has a medium-high risk rating overall. For inhalation during use, all experts judged both hazard and exposure as high.

Figure 2.

Scenario plot of hazard versus exposure for air freshener spray nano-product. Higher scores indicate greater risks.

Identical to the use and disposal scenarios, the risk triggers were scored from 1 = low to 5 = high. Each expert gave one composite score for each risk, considering both the probability and the severity of each trigger. The following high-risk triggers were identified as having a score of 4 or above: dispersibility and bioavailability, particle size under 200 nm, coating stability, antibacterial properties, photocatalytic and other catalytic activity, and susceptible population. These are explained in table 3.

Other triggers, such as effects of aggregated nanoparticles, media-dependent properties, and multiple disposal pathways, were ranked at medium risk only. Finally, triggers such as new product, used in conjunction with other products, high aspect ratio,3 and exists only in nano-form were judged as lower-level risks or did not apply directly to this product.

Next, it would be of interest to consider the intersection of high-risk scenarios and high-risk triggers for the same product. We can consider each high-risk trigger and see how each high-risk scenario is affected by it. For example, high bioavailability of the nanoparticles in the sprays could lead to a higher degree of inhalation during use. Taking this example further, we can now construct a matrix with high-risk triggers as rows and high-risk scenarios as columns, as in table 4. Each cell of the matrix can now be rated as a high, medium, or low degree of interactive effect. Particle size under 200 nm was not considered within this analysis, as it was deemed by experts to encompass a rather broad cross-section of nanoparticle properties. Because some of the risk triggers include the significant nanoparticle properties considered as risks for particle size under 200 nm, we are considering the exclusion of this trigger in future elicitations.

Table 4.  Evaluation of the intersection of high-risk triggers with high-risk scenarios for air freshener spray nano-product
High-risk trigger/scenarioUse – inhalationUse – skin absorptionUse – air releaseDisposal – water entrainment
  1. Note: M = medium; H = high; L = low.

Dispersibility and bioavailabilityMMHH
Catalytic propertyHMHM
Antibacterial propertiesHHMH
Susceptible populationHHLL
Coating stabilityMMHM

Each cell is the intersection of a scenario and a risk trigger, with a high-, medium-, or low-risk label. These indicated the degree of multiplicative or exacerbating effect on the intersection of the scenario and trigger. For the air freshener sprays, we can make the following observations about the intersection: Bioavailability enhanced the risk from air release and water entrainment. Similarly, catalytic activity of nanoparticles enhanced inhalation and air release risks during use. The material in the sprays is antibacterial and could kill or harm useful bacteria in the human body and the environment; thus, the risk from scenarios by which humans and the environment can be exposed to the material is enhanced. The major intersection hot spots involving the antibacterial risk trigger resulted during the use–inhalation, use–skin absorption, and disposal–water entrainment scenarios. Experts agree that people with weak respiratory systems can be affected by the spray particles; thus, risk from scenarios where the nanoparticles in the spray can be dispersed in air is enhanced. There was some uncertainty among the experts as to how risks from aggregated nanoparticles from air freshener sprays compared with those from dispersed nanoparticles; hence, this knowledge gap was deemed as possibly leading to high-risk scenarios.

Relative Risk Ranking of Nano-Products

Average hazard-related and exposure-related risk trigger scores across eight interviews for each product were then plotted as in figure 3.

Figure 3.

Potential risk plot for products under study. Higher scores indicate greater risks.

Preliminary analysis suggests that, in relative terms, potential risks are higher for sunscreens, toothpastes, and air fresheners, on the basis of scores for the hazard-related risk triggers and exposure-related risk triggers. Food supplements and MRI contrast agents pose significant hazards, but, assuming the governments regulate these products, experts judged a lower likelihood of effect. Conversely, experts judged a high degree of exposure due to nanoparticles present in the body. Field emission displays4 and racquets pose medium risk levels, and batteries pose the least risk, comparatively, to human health and the environment. Products like sunscreens and air fresheners pose a high risk because the nanoparticles in these products can easily break their nanomaterial matrices, whereas nanoparticles within displays, racquets, and batteries are bound within a matrix. However, these nanoparticles could become free during the disposal stage. Products with antibacterial properties pose a higher environmental risk, because of the sensitivity of the ecosystem to perturbations in bacterial life.5

Summary of Results

We present a methodology to identify risks posed by nano-products using an expert elicitation process to judge the likelihood and severity of risk triggers, determine nanoparticle properties that enhance risk, and identify the use and disposal scenarios where these risks occur. Furthermore, we delineate hot spots by taking into account the intersection of risk triggers with the use and disposal scenarios. For the air freshener spray, the significant risk triggers were judged to be dispersibility and bioavailability, particle size under 200 nm, coating stability, antibacterial properties, photocatalytic and other catalytic activity, and susceptible population. The significant use and disposal scenarios were judged to be inhalation during use, skin absorption during use, air release during use, and water entrainment during disposal. The risk hot spots were judged to arise from the intersection of (1) the bioavailability of the nanoparticles with air release during use and water entrainment during disposal scenarios, (2) the intersection of the catalytic activity trigger with the inhalation and air release during use scenarios, (3) the intersection of the antibacterial properties and susceptible population triggers with the inhalation and skin absorption during use scenarios, and, finally, (4) the intersection of the coating stability trigger with the air release during use scenarios. The air freshener nano-product was chosen as a proof of concept for the methodology explained here, and similar risk identification studies were also done with other products mentioned herein.

Discussion and Future Work

The intent of this expert elicitation methodology is to enable a comparison across nano-products of the intensity and relative importance of factors (triggers, scenarios, regulatory gaps, and knowledge gaps) that increased risk, on the basis of their nanomaterial content. In this article, we presented a proof-of-concept study of only the risk triggers and scenarios for the air freshener spray, though we have also studied the other nano-products presented. In the present methodology, the use and disposal scenarios and risk triggers are considered as contributing equally to the potential risk associated with each of the different nano-products. In the future, there should be a separate expert elicitation study comparing the relative significance of risk triggers across nano-products. A risk trigger that is rated high on three products might impact the composite risk profile differently for each.

The expert elicitation process should also be continuously widened to consider more scenarios and risk triggers and to include regulatory gaps and knowledge gaps in the process. As indicated earlier, nano-products are likely to fall through regulatory gaps due to classification and nomenclature issues. In previous work, we analyzed the nano-product life-cycle stages where particular regulatory agencies may be involved and possible regulatory gaps can occur (Wardak and Gorman 2006; Wardak et al. 2007).

In future work, we will aim to assess which regulatory gaps apply to which nano-products, to what extent, and where in the product life cycle. This will establish pathways to move from the current regime of inventory-based regulation to risk-based regulation with an adaptive management capacity. In this context, risk-based refers to the ability of regulations to identify and mitigate environmental problems with nanotechnology-based consumer products, where an inventory-based system collects information on the products without rigorous risk analysis. The gaps constitute places where it is especially important to apply ESEM by designing the product, so that it poses minimum risk and can be rapidly removed from the system if unanticipated problems emerge. Future work will also focus on the interaction between products and nanotechnologies that could pose great problems for the current regulatory system. Ultimately, we have to move from a reactive approach to anticipatory governance (Guston and Sarewitz 2002).

The inability of experts to judge certain scenarios or triggers represents an opportunity for future environmental, health, and safety research in these areas. As mentioned earlier, risks posed by aggregated nanoparticles in air freshener sprays could not even be subjectively judged due to a lack of knowledge. This risk identification and expert elicitation methodology provides a justification for prioritizing such studies. Furthermore, although the present study focused on a framework for the early evaluation of risks of nanotechnology-based products, the method could also be used to anticipate possible benefits, creating opportunities for new nanotechnology products (U.S. Environmental Protection Agency 2007b).


We thank our interdisciplinary panel of experts and Andrew D. Maynard of the Woodrow Wilson International Center for Scholars, Washington, DC, for discussions on part of this work. This work was supported through National Science Foundation SES Award #s 0508347 and 0708914.


  • 1

    A quantum dot is an ensemble of atoms of a particular material (metal or semiconductor) that is confined within a nanostructure in all three dimensions, to result in quantization of energy levels of the material, so that emission of monochromatic light of high intensity can occur upon appropriate excitation.

  • 2

    Risks related to other characteristics of the products and unrelated to nanomaterials are not included in this framework.

  • 3

    High aspect ratio usually refers to particles with one dimension (e.g., length) being far larger than two other dimensions (e.g., width and height). Given the similarity of this property to those of particles such as asbestos fibers, the suspected risk is that a particle with a high aspect ratio is likely more easily transported within the human body and the ecosystem.

  • 4

    Displays that use the property of enhanced electron emission due to high electric fields within nanomaterials are refered to as field emission displays.

  • 5

    Our research group is currently conducting an expert elicitation study of the potential risks of silver nanoparticles.

About the Authors

Ahson Wardak is a Ph.D. student in Systems & Information Engineering, Michael E. Gorman is a Professor of Science, Technology & Society, and Nathan S. Swami is an Assistant Professor of Electrical and Computer Engineering, all at the University of Virginia in Charlottesville, Virginia. Shilpa Deshpande graduated from the University of Virginia with an M.S. in Electrical and Computer Engineering in May 2007.