Control Banding Tools for Engineered Nanoparticles: What the Practitioner Needs to Know

Control Banding Tools for Engineered Nanoparticles: What the Practitioner Needs to Know Abstract Control banding (CB) has been widely recommended for the selection of exposure controls for engineered nanomaterials (ENMs) in the absence of ENM-specific occupational exposure limits (OELs). Several ENM-specific CB strategies have been developed but have not been systematically evaluated. In this article, we identify the data inputs and compare the guidance provided by eight CB tools, evaluated on six ENMs, and assuming a constant handling/use scenario. The ENMs evaluated include nanoscale silica, titanium dioxide, silver, carbon nanotubes, graphene, and cellulose. Several of the tools recommended the highest level of exposure control for each of the ENMs in the evaluation, which was driven largely by the hazard banding. Dustiness was a factor in determining the exposure band in many tools, although most tools did not provide explicit guidance on how to classify the dustiness (high, medium, low), and published data are limited on this topic. The CB tools that recommended more diverse control options based on ENM hazard and dustiness data appear to be better equipped to utilize the available information, although further validation is needed by comparison to exposure measurements and OELs for a variety of ENMs. In all CB tools, local exhaust ventilation was recommended at a minimum to control exposures to ENMs in the workplace. Generally, the same or more stringent control levels were recommended by these tools compared with the OELs proposed for these ENMs, suggesting that these CB tools would generally provide prudent exposure control guidance, including when data are limited. control banding, dustiness, engineered nanomaterials, hazard banding, occupational exposure banding, occupational exposure limits Introduction The introduction of engineered nanomaterials (ENMs) into the workplace has created a challenge in assuring that their development, manufacture, production, and use can be performed safely. Given the limited information about the health risks associated with occupational exposure to these ENMs, individual companies, trade associations, and government agencies have instituted various risk management strategies to protect the health of workers (Schulte et al., 2013). In the absence of specific information, precautionary approaches to exposure control are recommended to ensure worker health protection (BSI, 2007; Schulte and Salamanca-Buentello, 2007; NIOSH, 2009a, 2012, 2013b). The traditional approach to protecting worker health is to measure worker exposures to potentially hazardous agents, compare them with occupational exposure limits (OELs), and then determine if existing control measures provide adequate protection (NIOSH, 2009b). Reliance on this approach has become increasingly difficult because of the growing number of potentially hazardous materials in the workplace that do not have OELs (Garrod and Rajan-Sithamparanadarajah, 2003). Control banding (CB) strategies have been proposed to make engineering control decisions for general chemical substances without OELs (NIOSH, 2009b). Many ENMs and ENM-enabled compounds also lack specific OELs and may have little or no toxicity information, and thus CB strategies have been proposed for evaluating and controlling exposures to ENMs in the workplace. These strategies are evaluated in this article. Although regulatory OELs for ENMs are not available to date, various groups have derived OELs for a number of ENMs based on nanotoxicology data and using various derivation methods (Mihalache et al., 2017). These OELs provide a basis for comparison of the hazard and CB results based on the ENM CB tools for a set of ENMs. Early efforts to address the control of exposures to potentially toxic or biologically active materials with little or no toxicity information available were simultaneously developed in the pharmaceutical (Sargent and Kirk, 1988; Naumann et al., 1996) and chemical (Brooke, 1998; Henry and Schaper, 1990) industries. Gardner and Oldershaw (1991) proposed the use of pragmatic exposure control concentrations (PECC) for volatile organic compounds without OELs in response to classification, packaging, and labeling directives in Europe; the proposed PECC were set at the mean OELs for similar substances with both OELs and risk phrases. CB strategies have also been used for many years to support hazard communications and labeling and to provide practical approaches to hazard evaluation and exposure control for use in small businesses, including the Control of Substances Hazardous to Health (COSHH) Essentials (HSE, 2009); Global Harmonization System (GHS) (UNECE, 2011); and Occupational Safety and Health Administration (OSHA) guidance (OSHA, 2012). Typically, CB strategies consist of two main components: (i) hazard bands (HBs), and (ii) exposure (or emission potential) bands. These qualitative bands provide rankings of substances based on their hazardous properties and their production/use, which range from low to high levels of hazard and/or exposure potential (EP). The combination of the hazard and exposure bands is used to derive the control band and associated engineering control options for a given occupational scenario. HBs are typically derived from toxicological data of adverse responses associated with acute or chronic exposures to hazardous substances in experimental animal studies, as well as data in humans when available. The five hazard categories, ranging from minimal to severe, are related to the health hazard rating system proposed by Henry and Schaper (1990). In addition to qualitative descriptors of the toxic effects, some HBs include quantitative exposure concentration ranges. Some of the earliest ‘target airborne concentration ranges’ were proposed by Brooke (1998) and are included in the COSHH Essentials CB tool. A general term for these exposure concentration ranges is occupational exposure bands (OEBs), which are typically order-of-magnitude, 8-h time-weighted average (TWA) concentrations (McKernan and Seaton, 2014). OEBs are related to the severity of the hazard such that the more severe the hazard, the lower the OEB (Fig. 1). Figure 1. View largeDownload slide CB for nanomaterials. Adapted from Naumann et al. (1996); Brooke (1998); Ader et al. (2005); Zalk and Nelson (2008); HSE (2009); ANSES (2010); UNECE (2011); OSHA (2012); Kuempel et al. (2012); ISO (2014). Abbreviation: TWA: Time-weighted average. Figure 1. View largeDownload slide CB for nanomaterials. Adapted from Naumann et al. (1996); Brooke (1998); Ader et al. (2005); Zalk and Nelson (2008); HSE (2009); ANSES (2010); UNECE (2011); OSHA (2012); Kuempel et al. (2012); ISO (2014). Abbreviation: TWA: Time-weighted average. Exposure bands or emission potential bands are qualitative descriptors of potential exposure levels given the factors that influence exposure such as dustiness (propensity of the material to become airborne), type of process or task being performed, and amount of material being handled (ISO, 2014). The CB recommendations on exposure control options often include the following four main areas: (i) good occupational hygiene practices, including general ventilation and intermittent use of personal protective equipment; (ii) engineering controls, including local exhaust ventilation; (iii) containment systems; and (iv) the need to seek guidance from a specialist. Other CB schemes include five control bands and associated performance-based exposure control limits, as shown in Fig. 1. CB strategies have also been suggested as a pragmatic approach to manage the potential health risk resulting from exposure to nanomaterials (Maynard, 2007; Schulte et al., 2008; Kuempel et al., 2012). Selection of appropriate control bands is uncertain in the absence of specific toxicology and exposure data for many nanomaterials. Several of the proposed ENM-specific CB tools attempt to address this concern by (i) taking a precautionary approach in assigning higher HBs, and consequently assigning higher risk or control bands, when information is limited or lacking; (ii) identifying high-concern substances based on particle properties (e.g. fibrous structure); and (iii) identifying the most severe health endpoints (e.g. carcinogenicity) to drive the selection of the control band. Some ENM-specific CB tools [e.g. French Agency for Food, Environmental, and Occupational Safety (ANSES) and International Organization for Standardization (ISO)] recommend adding one or more bands when using bulk material information to assign a HB for the nanomaterial (ANSES, 2010; ISO, 2014). Currently available CB tools that are specific to ENMs include the following eight tools: the CB Nanotool (Paik et al., 2008; Zalk et al., 2009); ANSES (ANSES, 2010); Stoffenmanager Nano (Duuren-Stuurman et al., 2011); Precautionary Matrix (Höck et al., 2013); ISO (ISO, 2014); EC Guidance (European Commission, 2014); NanoSafer (v. 1.1 beta) (Jensen et al., 2013); and the GoodNanoGuide (Kulinowski and Jaffe, 2009). These strategies have both similarities and differences in their features, including their scope and applicability, parameters used in the hazard/severity banding, and exposure/probability/emission potential banding, and in the classification of risk or control bands (Brouwer, 2012; Sánchez Jiménez et al., 2016). Each strategy targets different users and applicability domains (e.g. laboratory versus small business). The amount and detail of information and professional knowledge required for implementing each strategy also vary. A recent article by Liguori et al. (2016) provides a detailed review of six of these CB tools and updates the overview by Brouwer (2012). Draft guidance on developing OEBs for chemical hazards was issued by the National Institute for Occupational Safety and Health (NIOSH), which includes ENMs when sufficient toxicity data are available for either the ENM or its parent material (NIOSH, 2017). The NIOSH (2017) process does not provide CB recommendations, and it is not considered further here. All of the CB strategies currently available for ENMs are evaluated in this article using a set of six ENMs and defined working conditions, and cross-tool comparisons of the inputs and outcomes are provided. The objectives of this articles are to utilize the available CB tools for ENMs on a pilot set of ENMs to (i) identify the types and sources of information required, as illustrated by assessing a diverse set of ENMs, (ii) compare and evaluate the specific guidance provided by each tool, including its utility and limitations, and (iii) identify important data gaps that hinder the effective use of these tools and suggest areas of research to improve the evidence basis needed for hazard and CB of ENMs. Methods Description of selected engineered nanomaterials Six ENMs were evaluated in this article, including nanoscale silicon dioxide (SiO2), titanium dioxide (TiO2), silver (Ag), single-walled carbon nanotubes (SWCNT), graphene, and cellulose. These materials were selected because they are commonly used nanomaterials worldwide (Future Markets Inc., 2013) and because they represent a range of information available for nanomaterials in terms of hazard and dustiness (Table 1). SiO2 nanoparticles are used in a wide variety of markets, including medical, transportation, building materials, electronics, energy, and food industries. TiO2 nanoparticles have been used extensively in cosmetics, pigments, paints, and coatings (Piccinno et al., 2012). Silver nanoparticles have been used in various applications such as jewelry, photography, and antibacterial products and are increasingly being used in medical and consumer products including electronics and textile coating because of their physicochemical properties at the nanoscale (Wijnhoven et al., 2009; Nowack et al., 2011). Carbon nanotubes (CNTs) consist of nanoscale cylinders of carbon that can be produced with very large aspect ratios and are used in many industrial applications including electronics, polymer composites, and coatings and in biomedical applications including enhanced electron-scanning microscopy imaging and biosensors (NIOSH, 2013a). Graphene is made of pure carbon with atoms arranged in a regular hexagonal pattern and in a flat one-atom thick sheet; its commercial applications utilize its properties such as mechanical stiffness, strength and elasticity, and very high electrical and thermal conductivity (Novoselov et al., 2012). Nanocellulose is one of the newest commercially available ENMs, which has high strength and thermal stability and is gaining attention within ‘green chemistry’ as a renewable and biodegradable material (Isogai, 2013). Table 1. Characteristics of ENMs evaluated in this article in the various CB tools. Chemical composition and form of ENM  Name or description  Manufacturer (SDS revision date)  CAS number  Description  Primary particle dimensions (nm)  Specific surface area (m2/g)a  Dustiness (%) respirable fractionb  Silicon dioxide (SiO2), amorphous  Aerosil 380 F  Evonik, Essen Germany (15 October 2013)  112945-52-5 [SiO2]  Fumed, nanoscale powder  nrc  380  5.5  Titanium dioxide (TiO2)  Aeroxide P25  Evonik, Essen Germany (9 February 2014)  13463-67-7 [TiO2]  Fumed, nanoscale powder  20 (diameter)a  50  7.2  Silver nanoparticles  nr  Quantum Sphere, Inc., Santa Ana, CA (24 May 2007)  7440-22-4 [Ag]  Nanoscale powder  20–40 (diameter)  nr  0.4  CNT  SWCNT  Unidym Inc., Sunnyvale, CA (8 February 2011)  nr  Nanoscale powder  0.8–1.2 (diameter); 100–1000 (length)  508  31.8  Graphene  nr  Angstron Materials, Inc, Dayton, OH (8 May 2013)  1034343-98-0 (Graphene)  Nanoscale powder with <3 graphene layers  <10 (diameter); 1 (thickness); length nr  nr  nrd  Nanocellulose  Nanofibrillated fiber  Engineered Fiber Technologies, Shelton, CT (21 June 2007)  68442-85-3  Cellulose nanofibrils  100–500 (diameter)  nr  nrd  Chemical composition and form of ENM  Name or description  Manufacturer (SDS revision date)  CAS number  Description  Primary particle dimensions (nm)  Specific surface area (m2/g)a  Dustiness (%) respirable fractionb  Silicon dioxide (SiO2), amorphous  Aerosil 380 F  Evonik, Essen Germany (15 October 2013)  112945-52-5 [SiO2]  Fumed, nanoscale powder  nrc  380  5.5  Titanium dioxide (TiO2)  Aeroxide P25  Evonik, Essen Germany (9 February 2014)  13463-67-7 [TiO2]  Fumed, nanoscale powder  20 (diameter)a  50  7.2  Silver nanoparticles  nr  Quantum Sphere, Inc., Santa Ana, CA (24 May 2007)  7440-22-4 [Ag]  Nanoscale powder  20–40 (diameter)  nr  0.4  CNT  SWCNT  Unidym Inc., Sunnyvale, CA (8 February 2011)  nr  Nanoscale powder  0.8–1.2 (diameter); 100–1000 (length)  508  31.8  Graphene  nr  Angstron Materials, Inc, Dayton, OH (8 May 2013)  1034343-98-0 (Graphene)  Nanoscale powder with <3 graphene layers  <10 (diameter); 1 (thickness); length nr  nr  nrd  Nanocellulose  Nanofibrillated fiber  Engineered Fiber Technologies, Shelton, CT (21 June 2007)  68442-85-3  Cellulose nanofibrils  100–500 (diameter)  nr  nrd  nr: not reported. aAs reported in Evans et al. (2012). bDustiness measured at 50% relative humidity (Evans et al., 2012). cNot reported in SDS. dLack of published test data. View Large Overview of CB tools examined The various CB tools have been reviewed in recent publications (Eastlake et al., 2016; Liguori et al., 2016; Sánchez Jiménez et al., 2016). Several of the tools (ANSES, ISO, EC Guidance) follow a decision tree approach where the user answers questions about the nanomaterial, such as material form (solid/liquid/powder form), process (e.g. high/low energy process), and quantity to derive an EP, and then uses material characteristics (such as solubility, shape, biopersistence, and availability of toxicological data) to derive HBs. The second primary type of CB tool follows a score-based approach, which assesses overall hazard and EPs using explicit numerical criteria. The score-based approach gives a range of scores based on characteristics (similar to those in the decision tree approach) of the nanomaterial or parent material. CB Nanotool is the only tool to utilize a score-based approach for both hazard and EP (Paik et al., 2008; Zalk et al., 2009). EP and hazard severity are scored on a potential total of 100 points (higher values indicate higher hazard/EP). Any unknown properties or information should be assigned as ‘unknown’ and scored as 75% of the maximum value for each category. This score-based approach in CB Nanotool results in a default recommendation of containment control when key information is missing. Stoffenmanager Nano is a tiered approach in which the risk prioritization score allows for the implementation of controls followed by further evaluation of hazard and EP. The exposure banding process in Stoffenmanager Nano is a score-based approach that utilizes a range of user inputs including type of task, room ventilation, and whether engineering controls or protective equipment is used. In contrast, the hazard banding process in Stoffenmanager Nano opts for a decision tree approach, which relies on classification and labeling of products in accordance with the European classification of chemicals scheme (Duuren-Stuurman et al., 2011; Duuren-Stuurman et al., 2012). NanoSafer focuses on nanomaterials in powder form. This tool uses physical data (particle size, density, and surface area) and toxicological data from the safety data sheet (SDS) along with process data to determine a HB score (Jensen et al., 2013). NanoSafer places materials in one of four HBs: HB1 (0–0.25); HB2 (0.26–0.50); HB3 (0.51–0.75); HB4 (0.76–1.00). The EP is calculated for both short (15-min) and longer (8-h) exposures and for workers near the process (near field) and further from the work area (far field). This scoring takes into account dustiness, handling energy, amount handled, work duration and process cycles, volume of the room and air exchange rate. The EP is placed into five bands: EP1 (<0.11); EP2 (0.11–0.25); EP3 (0.26–0.50); EP4 (0.51–1.00); and EP5 (>1.00). The final risk level (RL1–RL5) is based on a combination of the HB and EP scores. The output for most of the CB tools discussed in this article is a control band, which recommends an appropriate exposure control approach in four or five levels (e.g. general ventilation, local exhaust ventilation, containment or seek specialist advice). The two exceptions are Stoffenmanager Nano and the Precautionary Matrix. Stoffenmanager Nano combines the hazard and control bands into a risk matrix, which results in a three-level prioritization scheme (high, medium, and low priority). This approach allows the user to implement appropriate controls and then assess exposure or utilize the tool to reevaluate the process and material based on risk. The Precautionary Matrix is unique in that it is designed to help businesses address the need for nanospecific action based on factors that consider both human and environmental risks. The final output of this tool provides a score indicating precautionary need with respect to employees handling materials and/or environmental issues. Any score above 20 indicates a need for caution. Description of CB tool inputs The primary parameters for the hazard and exposure banding process for each tool are summarized in Tables 2 and 3, respectively, along with the main input values for each of the tools in this evaluation. For comparison of the various CB strategies, the handling/use scenario was kept constant (e.g. hours worked, quantity of material used). The assumptions in this scenario include (i) ENMs were used in a small-scale production setting (i.e. research and development) that would include a small number of employees (one to five workers); (ii) employees performed tasks associated with handling a dry powder form of the ENM of interest approximately less than or equal to 4 h per day and 5 d per week; and (iii) the quantity used was approximately 50 g per day, which is based on reported levels in several carbonaceous production and downstream plants showing typical use quantities between 5 and 100 g in a standard weighing task (Dahm et al., 2012). Table 2. Hazard banding inputs for each material and tool evaluated. CB tool    Hazard input parameters  SiO2  TiO2  CNT  Nanosilver  Graphene  Nanocellulose  CB Nanotool  Parent material  Lowest occupation exposure limit (mg/m3)  6  2.4  3.5  0.01  2.5  5    Carcinogen?  Yes  Yes  No  No  Yesa  Unknownb    Dermal hazard?  Unknownb  No  No  No  Unknownb  Unknownb    Asthmagen?  Unknownb  Unknownb  Yes  No  Yesa  Unknownb  Nanoscale material  Surface reactivity  Unknownb  Unknownb  Unknownb  Unknownb  Unknownb  Unknownb    Particle shape  Unknownb  Unknownb  Tubular or fibrous  Unknownb  Anisotropic  Tubular or fibrous    Particle diameter (nm)  Unknownb  Unknownb  1–10  11–40  1–10  41–100    Solubility  Insoluble  Insoluble  Insoluble  Insoluble  Insoluble  Insoluble    Carcinogen?  No  Yes  Unknownb  Unknownb  No  No    Reproductive hazard?  No  Yes  Unknownb  Unknownb  Unknownb  Unknownb    Mutagen?  No  Unknownb  Unknownb  Unknownb  Unknownb  Unknownb    Dermal hazard?  No  No  Unknownb  Unknownb  Unknownb  Unknownb    Asthmagen?  Unknownb  Unknownb  Unknownb  Unknownb  Unknownb  Unknownb  GoodNanoGuide    Hazard group  C  B  B  C  C  C  Potential for material release  Free/unbound  Free/unbound  Free/unbound  Free/unbound  Free/unbound  Free/unbound  ANSES  Preliminary question  Does the product contain nanomaterials?  Yes  Yes  Yes  Yes  Yes  Yes    Is the nanosubstance already classified by a relevant authority?  No  No  No  No  No  No    Is it a biopersistent fiber?  No, not a fiber  No, not a fiber  No, not a fiber  No, not a fiber  No, not a fiber  Yes, it is an insoluble fiber    Is there a preliminary HB for the bulk material or most toxic analogous?  Yesc  Yesc  —  Yesc  Yesc  —  Bulk material  Bulk material: substance dissolution time >1 h  Yes, insoluble in waterd  Yes, insoluble in waterd  —  Yes, insoluble in waterd  Yesa,d  —    Bulk material: evidence of higher reactivity than bulk/ analogous material?  Yesa  Yesa  —  Yesa  Yesa  —  ISO    OEL dust (8-h TWA)  A  B  A  C  A  A  Acute toxicity  Be  Ba,e  Be  Be  —  Be  LD50 oral route (mg/kg)  —  A  —  —  —  Ad  LD50 dermal route (mg/kg)  —  Aa  —  —  —  Ad  LC50 inhalation 4H (mg/l) aerosols/particles  —  —  —  —  —  Ad  Severity of acute (life-threatening) effects  Ce  Be  Be  Ce  Be  Be  Sensitization  —  Aa  —  Ce  —  —  Mutagenicity/genotoxicity  —  Aa  —  —  —  —  Irritant/corrosiveness  Ee  A  Ae  Ae  Ae  Ee  Carcinogenicity  Ea,e  Ea,d,e  Ce  —  —  —  Developmental/reproductive toxicity  —  —  De  —  —  —  EC Guidance  Concern category  Characteristics of the manufactured nanomaterial  Medium–high concern  Medium–low concern  High concern  Medium–high concern  Medium–high concern  Medium–low concern  Dustiness band  Dustiness  Highf  Highf  Highf  Highf  Highf  Highf  Precautionary Matrix    Size of primary particles (free, bound, aggregated, or agglomerates) (nm)  1–500  1–500  1–500  1–500  1–500  1–500  Nanorelevance  Do the nanoparticles/rods form agglomerates >500 nm?  Nog  Nog  Nog  Nog  Nog  Nog  Potential effect  Redox activity or catalytic activity of nanoparticles/rods present in the nanomaterial  Not knownb  Not knownb  Not knownb  Not knownb  Not knownb  Not knownb    Stability (half-life) of the nanomaterial in the body  Not knownb  Not knownb  Not knownb  Not knownb  Not knownb  Not knownb  Stoffenmanager Nano    Product appearance  Powder  Powder  Powder  Powder  Powder  Powder    Dustiness (mg/kg)  High (>150–500)h  High (>150–500)h  Very high (>500)h  Medium (50–150)h  Unknownb  Unknownb    Moisture  Dry product (<5% moisture content)  Dry product (<5% moisture content)  Dry product (<5% moisture content)  Dry product (<5% moisture content)  Dry product (<5% moisture content)  Dry product (<5% moisture content)    Concentration of the nanocomponent in the product (%)  100  100  50–99  99.90  50–99  50–99    Does the product contain fibers/ fiber-like particles?  No  No  Yes  No  No  Yes    Length: diameter of the fiber (aspect ratio)  No  No  Yes  No  No  No    Hazardous properties  Unknownb  Carcinogenic (not mutagenic), reprotoxic and/ or very toxic  Toxic, corrosive, and/or respiratory allergens  Unknownb  Unknownb  Unknownb  NanoSafer    CAS Number  112945-52-5  13463-67-7  —  7440-22-4  1034343-98-0  68442-85-3      Is the material coated? (yes/no)  No  No  No  No  No  No      Morphology  Unknownb   No  Tube  No  Flake/plate/tabular/clay  Fiber      Solubility in water (g/L)  Insoluble (<1)  Insoluble (<1)  Insoluble (<1)  Insoluble (<1)  Insoluble (<1)  Insoluble (<1)      Shortest dimension (nm)  —  —  0.8  20  1  50      Middle dimension (nm)  —  —  —  —  —  —      Longest dimension (nm)  —  —  1000  40  1000  500      Average size (nm)  —  —  —  —  —  —      Density (g/cm3)  2.2  4.1  1.6  0.25  2.2  1.5      Surface area (powder material) (m2/g)  380i  15–50i,j  144 or 508i,j  5–25j  800j,k  284j,l      Respirable dustiness index (mg/ kg)  187.5  187.5  937.5  7.5  937.5  937.5  CB tool    Hazard input parameters  SiO2  TiO2  CNT  Nanosilver  Graphene  Nanocellulose  CB Nanotool  Parent material  Lowest occupation exposure limit (mg/m3)  6  2.4  3.5  0.01  2.5  5    Carcinogen?  Yes  Yes  No  No  Yesa  Unknownb    Dermal hazard?  Unknownb  No  No  No  Unknownb  Unknownb    Asthmagen?  Unknownb  Unknownb  Yes  No  Yesa  Unknownb  Nanoscale material  Surface reactivity  Unknownb  Unknownb  Unknownb  Unknownb  Unknownb  Unknownb    Particle shape  Unknownb  Unknownb  Tubular or fibrous  Unknownb  Anisotropic  Tubular or fibrous    Particle diameter (nm)  Unknownb  Unknownb  1–10  11–40  1–10  41–100    Solubility  Insoluble  Insoluble  Insoluble  Insoluble  Insoluble  Insoluble    Carcinogen?  No  Yes  Unknownb  Unknownb  No  No    Reproductive hazard?  No  Yes  Unknownb  Unknownb  Unknownb  Unknownb    Mutagen?  No  Unknownb  Unknownb  Unknownb  Unknownb  Unknownb    Dermal hazard?  No  No  Unknownb  Unknownb  Unknownb  Unknownb    Asthmagen?  Unknownb  Unknownb  Unknownb  Unknownb  Unknownb  Unknownb  GoodNanoGuide    Hazard group  C  B  B  C  C  C  Potential for material release  Free/unbound  Free/unbound  Free/unbound  Free/unbound  Free/unbound  Free/unbound  ANSES  Preliminary question  Does the product contain nanomaterials?  Yes  Yes  Yes  Yes  Yes  Yes    Is the nanosubstance already classified by a relevant authority?  No  No  No  No  No  No    Is it a biopersistent fiber?  No, not a fiber  No, not a fiber  No, not a fiber  No, not a fiber  No, not a fiber  Yes, it is an insoluble fiber    Is there a preliminary HB for the bulk material or most toxic analogous?  Yesc  Yesc  —  Yesc  Yesc  —  Bulk material  Bulk material: substance dissolution time >1 h  Yes, insoluble in waterd  Yes, insoluble in waterd  —  Yes, insoluble in waterd  Yesa,d  —    Bulk material: evidence of higher reactivity than bulk/ analogous material?  Yesa  Yesa  —  Yesa  Yesa  —  ISO    OEL dust (8-h TWA)  A  B  A  C  A  A  Acute toxicity  Be  Ba,e  Be  Be  —  Be  LD50 oral route (mg/kg)  —  A  —  —  —  Ad  LD50 dermal route (mg/kg)  —  Aa  —  —  —  Ad  LC50 inhalation 4H (mg/l) aerosols/particles  —  —  —  —  —  Ad  Severity of acute (life-threatening) effects  Ce  Be  Be  Ce  Be  Be  Sensitization  —  Aa  —  Ce  —  —  Mutagenicity/genotoxicity  —  Aa  —  —  —  —  Irritant/corrosiveness  Ee  A  Ae  Ae  Ae  Ee  Carcinogenicity  Ea,e  Ea,d,e  Ce  —  —  —  Developmental/reproductive toxicity  —  —  De  —  —  —  EC Guidance  Concern category  Characteristics of the manufactured nanomaterial  Medium–high concern  Medium–low concern  High concern  Medium–high concern  Medium–high concern  Medium–low concern  Dustiness band  Dustiness  Highf  Highf  Highf  Highf  Highf  Highf  Precautionary Matrix    Size of primary particles (free, bound, aggregated, or agglomerates) (nm)  1–500  1–500  1–500  1–500  1–500  1–500  Nanorelevance  Do the nanoparticles/rods form agglomerates >500 nm?  Nog  Nog  Nog  Nog  Nog  Nog  Potential effect  Redox activity or catalytic activity of nanoparticles/rods present in the nanomaterial  Not knownb  Not knownb  Not knownb  Not knownb  Not knownb  Not knownb    Stability (half-life) of the nanomaterial in the body  Not knownb  Not knownb  Not knownb  Not knownb  Not knownb  Not knownb  Stoffenmanager Nano    Product appearance  Powder  Powder  Powder  Powder  Powder  Powder    Dustiness (mg/kg)  High (>150–500)h  High (>150–500)h  Very high (>500)h  Medium (50–150)h  Unknownb  Unknownb    Moisture  Dry product (<5% moisture content)  Dry product (<5% moisture content)  Dry product (<5% moisture content)  Dry product (<5% moisture content)  Dry product (<5% moisture content)  Dry product (<5% moisture content)    Concentration of the nanocomponent in the product (%)  100  100  50–99  99.90  50–99  50–99    Does the product contain fibers/ fiber-like particles?  No  No  Yes  No  No  Yes    Length: diameter of the fiber (aspect ratio)  No  No  Yes  No  No  No    Hazardous properties  Unknownb  Carcinogenic (not mutagenic), reprotoxic and/ or very toxic  Toxic, corrosive, and/or respiratory allergens  Unknownb  Unknownb  Unknownb  NanoSafer    CAS Number  112945-52-5  13463-67-7  —  7440-22-4  1034343-98-0  68442-85-3      Is the material coated? (yes/no)  No  No  No  No  No  No      Morphology  Unknownb   No  Tube  No  Flake/plate/tabular/clay  Fiber      Solubility in water (g/L)  Insoluble (<1)  Insoluble (<1)  Insoluble (<1)  Insoluble (<1)  Insoluble (<1)  Insoluble (<1)      Shortest dimension (nm)  —  —  0.8  20  1  50      Middle dimension (nm)  —  —  —  —  —  —      Longest dimension (nm)  —  —  1000  40  1000  500      Average size (nm)  —  —  —  —  —  —      Density (g/cm3)  2.2  4.1  1.6  0.25  2.2  1.5      Surface area (powder material) (m2/g)  380i  15–50i,j  144 or 508i,j  5–25j  800j,k  284j,l      Respirable dustiness index (mg/ kg)  187.5  187.5  937.5  7.5  937.5  937.5  LD50: Dose associated with 50% lethality; LC50: Concentration associated with 50% lethality. Dash (—) indicates that no information was found for this specific parameter. aTOXNET. bInterpreted as unknown as no proper option available. cCreated using Table 1 of ANSES. dGESTIS. eECHA C&L. fMethod provides no option for ‘unknown’ therefore ‘high’. gNot known if deagglomeration of agglomerates (or aggregates) to primary nanoparticles/rods or agglomerates <500 nm occurs in the body. hStoffenmanager Nano dustiness levels are medium, high and very high, and unknown. iReported in Evans et al. (2012). jMultiple numbers provided and larger number used. kTechnical data sheet/SDS for material. lSehaui, Zhou, Berglund (2011). View Large Table 3. Exposure band parameters for each tool and levels of each category. CB tool  Information/scenario    Materials    SiO2  TiO2  CNT  Nanosilver  Graphene  Nanocellulose  All tools  Substance emission potential/physical form  Dry powder  Activity emission potential/ amount handled per day  50 g  Task duration  1–4 h/d  Task frequency  5 d/wk  Volume of the working room  108 m3 per NanoSafer  CB Nanotool  Dustiness  5.5%a medium  7.2%a medium  31.8%a high  0.4%a low  Unknown  Unknown  Number of employees with similar exposure  ≤5 employees  Good Nano Guide  Exposure duration  Medium  ANSES  Emission potential (high/ moderate +1 band)  EP3, powder  EP3, powder  EP3, powder  EP3, powder  EP3, powder  EP3, powder  Manufacturing/handling process  Handling powder  ISO  Exposure band for dust generation/dustiness  EB2  EB2  EB2  EB2  EB2  EB2  Manufacturing/handling process  Material in powder form—manufacturing use and handling—amount used >0.1 g—Low potential of dust  EC Guidance  Level of Exposure  High  Stoffenmanager Nano  Task characterization  Handling of products in small amounts (up to 100 g) or in situations where only low quantities of products are likely to be released    Is the task carried out at the breathing zone of the employee (distance person product <1 m)?  Yes    Is there more than one employee carrying out the same task simultaneously?  Yes    Is the working room being cleaned daily?  Yes    Are inspections and maintenance of machines/ancillary equipment being done at least monthly to ensure good condition and proper functioning and performance?  Yes    Ventilation of the working room  Mechanical and/or natural    Local control measures at the source  No control measures    Is the employee situation in a cabin?  No    Is personal protective equipment applied?  No  NanoSafer  Energy level  H3 (0.50): moderate energy (e.g. pour 5–30 cm drop height, blending of powder in liquid medium)  Air exchanges  8/h  Mass handled per cycle  0.025 kg  Length x width x height of workroom (m)  6 x 6 x 3  Cycle duration  60 min  Time to perform work cycle  15 min  Amount of product used per work cycle  0.1 kg  How many times is the cycle repeated daily?  4  Activity level of room  Low quiet  CB tool  Information/scenario    Materials    SiO2  TiO2  CNT  Nanosilver  Graphene  Nanocellulose  All tools  Substance emission potential/physical form  Dry powder  Activity emission potential/ amount handled per day  50 g  Task duration  1–4 h/d  Task frequency  5 d/wk  Volume of the working room  108 m3 per NanoSafer  CB Nanotool  Dustiness  5.5%a medium  7.2%a medium  31.8%a high  0.4%a low  Unknown  Unknown  Number of employees with similar exposure  ≤5 employees  Good Nano Guide  Exposure duration  Medium  ANSES  Emission potential (high/ moderate +1 band)  EP3, powder  EP3, powder  EP3, powder  EP3, powder  EP3, powder  EP3, powder  Manufacturing/handling process  Handling powder  ISO  Exposure band for dust generation/dustiness  EB2  EB2  EB2  EB2  EB2  EB2  Manufacturing/handling process  Material in powder form—manufacturing use and handling—amount used >0.1 g—Low potential of dust  EC Guidance  Level of Exposure  High  Stoffenmanager Nano  Task characterization  Handling of products in small amounts (up to 100 g) or in situations where only low quantities of products are likely to be released    Is the task carried out at the breathing zone of the employee (distance person product <1 m)?  Yes    Is there more than one employee carrying out the same task simultaneously?  Yes    Is the working room being cleaned daily?  Yes    Are inspections and maintenance of machines/ancillary equipment being done at least monthly to ensure good condition and proper functioning and performance?  Yes    Ventilation of the working room  Mechanical and/or natural    Local control measures at the source  No control measures    Is the employee situation in a cabin?  No    Is personal protective equipment applied?  No  NanoSafer  Energy level  H3 (0.50): moderate energy (e.g. pour 5–30 cm drop height, blending of powder in liquid medium)  Air exchanges  8/h  Mass handled per cycle  0.025 kg  Length x width x height of workroom (m)  6 x 6 x 3  Cycle duration  60 min  Time to perform work cycle  15 min  Amount of product used per work cycle  0.1 kg  How many times is the cycle repeated daily?  4  Activity level of room  Low quiet  aReported in Evans et al. (2012); categories assigned here are based on judgment: 0.1–1%, low; 1–10%, medium; >10%, high. View Large It should be noted that the rates of production from TiO2 and silver may be much higher—in the range of 1–5 kg per day based on published data (Lee et al., 2011). However, the upper range of material quantity for scoring of EP of any of the CB tools evaluated herein is 1 kg, with most tools giving quantities of greater than 1 g the highest score in this category. The physical properties of the ENMs utilized in this evaluation were obtained from the manufacturer’s technical data sheets and/or SDSs. The dustiness of the materials was classified in this article (based on judgment) as low, medium, or high according to the following respirable fraction: 0.1–1% low, 1–10% medium, >10% high. This information was used in the tools requiring dustiness category inputs (Tables 2 and 3). The data on the ENM dustiness were taken from the results of dustiness characterization reported in Evans et al. (2012), because no other large-scale dustiness test dataset for fine and nanomaterials was available. These data were collected using a Venturi test procedure, which may not be applicable to all models. Specifically, NanoSafer and ANSES recommend the use of methods from the EN 15051 standard for dustiness testing, which employ less aggressive methods of dispersion (European Committee for Standardization (CEN), 2013). Thus, the values used in this evaluation may overestimate the relative dustiness of the materials and result in higher EP scores. For the hazard banding of these six ENMs, data were collected from a variety of sources including governmental sources, professional organizations, online databases, and published guidance/literature (Table 4). SDSs were consulted to obtain information specific to the properties of each ENM: the physical, health, and environmental health hazards; protective measures; and safety precautions for handling, storing, and transporting the material. If an SDS specific to that ENM was available, then that information was obtained and used. OELs for the bulk (non-nanometer sized) material most similar to each ENM in this study were used in the banding. The lowest authoritative OELs were used, which were not necessarily regulatory OELs. Information from NIOSH and other authoritative guidance documents was used to address questions regarding toxicity and health hazards associated with each substance. As most ENMs do not have guidance documents with extensive literature and data reviews, these data may be obtained from online databases. For this study, we used a German substance database (GESTIS), United States National Library of Medicine Toxicology Data Network (TOXNET), and the European Chemicals Agency (ECHA) Classification and Labeling Inventory. In general, surface reactivity for a given mass-based exposure to each ENM was assumed to be high because of both the unknown potential for functionalization and the higher surface area of most ENMs versus the parent (or bulk) material. Solubility was determined based on information provided in the SDS or database literature search. If more than one type of solubility (soluble and insoluble) was listed, then the ENM was considered insoluble. If the parent material was indicated to be carcinogenic, a dermal hazard, or an asthmagen, then the ENM was also assumed to have similar health effects. Otherwise, when information was not available, all ENM data were indicated or interpreted as unknown. Table 4. Sources of information for CB tools and model inputs. Source information  Source  Content description  Website address  OEL guidance  US OSHA  Permissible exposure limits  https://www.osha.gov/  US NIOSH  Recommended exposure limits  http://www.cdc.gov/niosh/  American Conference of Governmental Industrial Hygienists (ACGIH)  Threshold limit values  http://www.acgih.org/  Online databases  Institute for Occupational Safety and Health of the German Social Accident Insurance  Substance database (GESTIS)  http://gestis-en.itrust.de/nxt/ gateway.dll/gestis_en/000000. xml?f=templates$fn=default.htm$3.0  US National Library of Medicine  Toxicology data network (TOXNET)  http://toxnet.nlm.nih.gov/  European Chemicals Agency  Classification & Labeling Inventory  http://echa.europa.eu/ information-on-chemicals  CB methods  Lawrence Livermore National Laboratory  CB Nanotool  http://controlbanding.net/    US NIOSH Oregon Nanoscience and Microtechnologies Institute (OMAMI) Oregon State University (OSU)  GoodNanoGuide  https://nanohub.org/groups/gng    French Agency for Food, Environmental and Occupational Health & Safety (ANSES)  Development of a specific CB tool for nanomaterials  https://www.anses.fr/sites/default/files/ documents/AP2008sa0407RaEN.pdf    ISO  Nanotechnologies— Occupational risk management applied to engineered nanomaterials—Part 2: Use of the CB approach TS 12901–2:2014  http://www.iso.org/iso/catalogue_detail. htm?csnumber=53375    National Research Centre for the Working Environment, Copenhagen, Denmark  NanoSafer  http://www.nanosafer.org/    Schweizerische Eidgenossenschaft— Federal office of Public Health  Precautionary Matrix  http://www.bag.admin.ch/nanotechnologie/ 12171/12174/12175/index. html?webgrab_path=aHR0cDovL3d3dy5iYWctYW53LmFkbWluLmNoL25hbm9yYXN0ZXIvcG9ydGFsX2VuLnBocD9tb2Q9YSZsYW5nPWVu&lang=en    Dutch Ministry of Social Affairs and Employment (SAE), TNO, Arbo Unie, BECO(EY)  Stoffenmanager Nano  https://nano.stoffenmanager.nl/    European Agency for Safety and Health at Work (EU-OSHA)  Guidance on the protection of the health and safety of workers from the potential risks related to nanomaterials at work  https://osha.europa.eu/en/news/ eu-safe-use-of-nanomaterials-commission- publishes-guidance-for-employers- and-workers  Guidance/ literature  US NIOSH  Occupational exposure to CNTs and nanofibers  http://www.cdc.gov/niosh/docs/2013–145/pdfs/ 2013–145.pdf  US NIOSH  Occupational exposure to titanium dioxide  http://www.cdc.gov/niosh/docs/2011–160/pdfs/ 2011–160.pdf  Varies depending on material  Material SDS  Specific to the material used  Source information  Source  Content description  Website address  OEL guidance  US OSHA  Permissible exposure limits  https://www.osha.gov/  US NIOSH  Recommended exposure limits  http://www.cdc.gov/niosh/  American Conference of Governmental Industrial Hygienists (ACGIH)  Threshold limit values  http://www.acgih.org/  Online databases  Institute for Occupational Safety and Health of the German Social Accident Insurance  Substance database (GESTIS)  http://gestis-en.itrust.de/nxt/ gateway.dll/gestis_en/000000. xml?f=templates$fn=default.htm$3.0  US National Library of Medicine  Toxicology data network (TOXNET)  http://toxnet.nlm.nih.gov/  European Chemicals Agency  Classification & Labeling Inventory  http://echa.europa.eu/ information-on-chemicals  CB methods  Lawrence Livermore National Laboratory  CB Nanotool  http://controlbanding.net/    US NIOSH Oregon Nanoscience and Microtechnologies Institute (OMAMI) Oregon State University (OSU)  GoodNanoGuide  https://nanohub.org/groups/gng    French Agency for Food, Environmental and Occupational Health & Safety (ANSES)  Development of a specific CB tool for nanomaterials  https://www.anses.fr/sites/default/files/ documents/AP2008sa0407RaEN.pdf    ISO  Nanotechnologies— Occupational risk management applied to engineered nanomaterials—Part 2: Use of the CB approach TS 12901–2:2014  http://www.iso.org/iso/catalogue_detail. htm?csnumber=53375    National Research Centre for the Working Environment, Copenhagen, Denmark  NanoSafer  http://www.nanosafer.org/    Schweizerische Eidgenossenschaft— Federal office of Public Health  Precautionary Matrix  http://www.bag.admin.ch/nanotechnologie/ 12171/12174/12175/index. html?webgrab_path=aHR0cDovL3d3dy5iYWctYW53LmFkbWluLmNoL25hbm9yYXN0ZXIvcG9ydGFsX2VuLnBocD9tb2Q9YSZsYW5nPWVu&lang=en    Dutch Ministry of Social Affairs and Employment (SAE), TNO, Arbo Unie, BECO(EY)  Stoffenmanager Nano  https://nano.stoffenmanager.nl/    European Agency for Safety and Health at Work (EU-OSHA)  Guidance on the protection of the health and safety of workers from the potential risks related to nanomaterials at work  https://osha.europa.eu/en/news/ eu-safe-use-of-nanomaterials-commission- publishes-guidance-for-employers- and-workers  Guidance/ literature  US NIOSH  Occupational exposure to CNTs and nanofibers  http://www.cdc.gov/niosh/docs/2013–145/pdfs/ 2013–145.pdf  US NIOSH  Occupational exposure to titanium dioxide  http://www.cdc.gov/niosh/docs/2011–160/pdfs/ 2011–160.pdf  Varies depending on material  Material SDS  Specific to the material used  View Large Hazard data on the adverse effects from repeated exposure to these nanomaterials in animals were also evaluated given the relevance to potential worker exposures for up to a working lifetime. Rat is the rodent species used in the criteria for specific target organ toxicity–repeated exposure (STOT-RE) in many of the hazard banding schemes. Therefore, subchronic inhalation studies in rats were identified from literature searches in Pubmed, using the search terms ‘nanomaterial name’ and ‘rat’ and ‘inhalation’. The adverse effect levels from the identified rat studies are compared with the effect levels in the ANSES and GHS hazard banding schemes for STOT-RE. OELs that have been proposed for nanomaterials (Table S1, available at Annals of Occupational Hygiene online) are used in comparisons with the CB results in this evaluation. OELs are typically based on a more in-depth analysis of the data, although different data, methods, and assumptions may have been used in deriving those OELs. The steps involved in selecting and using the evaluated CB tools are shown in Fig. 2. This figure references the process and data sources, which are used in conducting the analyses described in this article. Figure 2. View largeDownload slide Steps for selecting and using the CB tools. Figure 2. View largeDownload slide Steps for selecting and using the CB tools. Results A summary of results of the recommended risk/control bands for each CB strategy and for all six ENMs evaluated is shown in Table 5. The results of both the exposure and HBs are presented, when applicable. This table shows that the output for each tool is unique. For example, the Precautionary Matrix is different than the other tools discussed, in that the process does not result in the determination of a control band. Rather the Precautionary Matrix specifies whether precautionary, nanospecific safety measures are needed or not based on a calculated score. The ANSES and ISO tools are very similar in nature and include five CB levels: 1, general ventilation; 2, local exhaust ventilation (exterior hood, table hood); 3, enclosed ventilation (fume hood, ventilated booth); 4, full containment; or 5, full containment and review by specialist. NanoSafer also has five RLs, which correspond to control recommendations including the following: RL1, local exhaust ventilation (LEV)/fume hood; RL2, LEV/fume hood potentially with respirator; RL3, LEV/fume hood with respirator; RL4, fume hood/enclosure/glovebox with respirator; RL5, fume hood/enclosure/glovebox with Supplied Air Respirator. CB Nanotool, GoodNanoGuide, and the EC Guidance have four control bands: 1, general ventilation; 2, local exhaust ventilation/engineering controls; 3, containment; or 4, seek specialist advice. In contrast, Stoffenmanager Nano assigns one of three risk priority bands (1, high priority; 2, medium priority; or 3, low priority). Table 5. Summary of CB tool results.     Nanomaterial      SiO2  TiO2  CNT  Nanosilver  Graphene  Nanocellulose  CB Nanotool  Severity score  45.5  59  60  56  60  53  Exposure probability score  65  65  80  57.5  72.5  72.5  Control band  RL2, fume hoods or local exhaust ventilation  RL3, containment  RL4, seek specialist advice  RL3, containment  RL3, containment  RL3, containment  Good Nano Guide    Hazard group C, limited data CB, 3 seek specialist advice  Hazard group B, NIOSH CIB TiO2 CB, 3 containment  Hazard group B, NIOSH CIB CNT/F CB, 3 containment  Hazard group C, limited data CB, 4 seek specialist advice  Hazard group C, limited data CB, 4 seek specialist advice  Hazard group C, limited data CB, 4 seek specialist advice  ANSES    HB5 + emission potential 4 = control level 5: full containment and review by a specialist required: seek expert advice  HB5 + emission potential 4 = control level 5: full containment and review by a specialist required: seek expert advice  HB5 + emission potential 4 = control level 5: full containment and review by a specialist required: seek expert advice  HB5 + emission potential 3 = control level 5: full containment and review by a specialist required: seek expert advice  HB5 + emission potential 4 = control level 5: full containment and review by a specialist required: seek expert advice  HB5 + emission potential 4 = control level 5: full containment and review by a specialist required: seek expert advice  ISO    HB E, severe hazard + exposure band 2 = control band 5 Full containment and review by a specialist: seek expert advice  HB E, severe hazard + exposure band 2 = control band 5 Full containment and review by a specialist: seek expert advice  HB E, severe hazard + exposure band 2 = control band 5 Full containment and review by a specialist: seek expert advice  HB E, severe hazard + exposure band 2 = control band 5 Full containment and review by a specialist: seek expert advice  HB E, severe hazard + exposure band 2 = control band 5 Full containment and review by a specialist: seek expert advice  HB E, severe hazard + exposure band 2 = control band 5 Full containment and review by a specialist: seek expert advice  EC Guidance    Medium–high concern + high level of exposure = RL4 Risk assessment performed by an expert + it is essential that measures specifically designed for the processes in question to be adopted  Medium–low concern + high level of exposure = RL3 Risk assessment performed by an expert + closed systems or containment must be used  High concern category + high level of exposure = RL4 Risk assessment performed by an expert + it is essential that measures specifically designed for the processes in question to be adopted  Medium–high concern + high level of exposure = RL4 Risk assessment performed by an expert + it is essential that measures specifically designed for the processes in question to be adopted  Medium–high concern + high level of exposure = RL4 Risk assessment performed by an expert + it is essential that measures specifically designed for the processes in question to be adopted  Medium–low concern + high level of exposure = RL3 Risk assessment performed by an expert + closed systems or containment must be used  Precautionary Matrix    All numbers > 20 Precautionary need warranted  All numbers > 20 Precautionary need warranted  All numbers > 20 Precautionary need warranted  All numbers > 20 Precautionary need warranted  All numbers > 20 Precautionary need warranted  All numbers > 20 Precautionary need warranted  Stoffenmanager Nano  Task weighted  E, extreme hazard class 3, high EP I, high-risk priority  D, very high hazard class 3, high EP I, high-risk priority  E, extreme hazard class 3, high EP I, high-risk priority  D, very high hazard class 2, average EP II, medium-risk priority  E, extreme hazard class 3, high EP I, high-risk priority  E, extreme hazard class 3, high EP I, high-risk priority    Time and frequency weighted  E, extreme hazard class 2, average EP I, high-risk priority  D, very high hazard class 3, high EP II, medium-risk priority  E, extreme hazard class 2, average EP I, high-risk priority  D, very high hazard class 2, average EP II, medium-risk priority  E, extreme hazard class 2, average EP I, high-risk priority  E, extreme hazard class 2, average EP I, high-risk priority  NanoSafer  Estimated hazard level  0.59  1  0.59  0.761  0.539  0.488    Near field  Acute: 0.1267 RL2: low toxicity/low EP EB2: low EP  Acute: 0.0777 RL4: high toxicity/high EP EB1: very low EP  Acute: 1.056 RL5: very high toxicity/moderate- to-very high EP EB5: very high EP  Acute: 0.0227 RL4: high toxicity/high EP EB1: very low EP  Acute: 3.2 RL5: very high toxicity/moderate- to-very high EP EB5: very high EP  Acute: 0.3873 RL4: high toxicity/high EP EB3: moderate EP    Daily: 0.1275 RL2: low toxicity/low EP EB2: low EP  Daily: 0.0782 RL4: high toxicity/high EP EB1: very low EP  Daily: 1.063 RL5: very high toxicity/moderate- to-very high EP EB5: very high EP  Daily: 0.0229 RL4: high toxicity/high EP EB1: very low EP  Daily: 3.22 RL5: very high toxicity/moderate- to-very high EP EB5: very high EP  Daily: 0.3899 RL4: high toxicity/high EP EB3: moderate EP      Nanomaterial      SiO2  TiO2  CNT  Nanosilver  Graphene  Nanocellulose  CB Nanotool  Severity score  45.5  59  60  56  60  53  Exposure probability score  65  65  80  57.5  72.5  72.5  Control band  RL2, fume hoods or local exhaust ventilation  RL3, containment  RL4, seek specialist advice  RL3, containment  RL3, containment  RL3, containment  Good Nano Guide    Hazard group C, limited data CB, 3 seek specialist advice  Hazard group B, NIOSH CIB TiO2 CB, 3 containment  Hazard group B, NIOSH CIB CNT/F CB, 3 containment  Hazard group C, limited data CB, 4 seek specialist advice  Hazard group C, limited data CB, 4 seek specialist advice  Hazard group C, limited data CB, 4 seek specialist advice  ANSES    HB5 + emission potential 4 = control level 5: full containment and review by a specialist required: seek expert advice  HB5 + emission potential 4 = control level 5: full containment and review by a specialist required: seek expert advice  HB5 + emission potential 4 = control level 5: full containment and review by a specialist required: seek expert advice  HB5 + emission potential 3 = control level 5: full containment and review by a specialist required: seek expert advice  HB5 + emission potential 4 = control level 5: full containment and review by a specialist required: seek expert advice  HB5 + emission potential 4 = control level 5: full containment and review by a specialist required: seek expert advice  ISO    HB E, severe hazard + exposure band 2 = control band 5 Full containment and review by a specialist: seek expert advice  HB E, severe hazard + exposure band 2 = control band 5 Full containment and review by a specialist: seek expert advice  HB E, severe hazard + exposure band 2 = control band 5 Full containment and review by a specialist: seek expert advice  HB E, severe hazard + exposure band 2 = control band 5 Full containment and review by a specialist: seek expert advice  HB E, severe hazard + exposure band 2 = control band 5 Full containment and review by a specialist: seek expert advice  HB E, severe hazard + exposure band 2 = control band 5 Full containment and review by a specialist: seek expert advice  EC Guidance    Medium–high concern + high level of exposure = RL4 Risk assessment performed by an expert + it is essential that measures specifically designed for the processes in question to be adopted  Medium–low concern + high level of exposure = RL3 Risk assessment performed by an expert + closed systems or containment must be used  High concern category + high level of exposure = RL4 Risk assessment performed by an expert + it is essential that measures specifically designed for the processes in question to be adopted  Medium–high concern + high level of exposure = RL4 Risk assessment performed by an expert + it is essential that measures specifically designed for the processes in question to be adopted  Medium–high concern + high level of exposure = RL4 Risk assessment performed by an expert + it is essential that measures specifically designed for the processes in question to be adopted  Medium–low concern + high level of exposure = RL3 Risk assessment performed by an expert + closed systems or containment must be used  Precautionary Matrix    All numbers > 20 Precautionary need warranted  All numbers > 20 Precautionary need warranted  All numbers > 20 Precautionary need warranted  All numbers > 20 Precautionary need warranted  All numbers > 20 Precautionary need warranted  All numbers > 20 Precautionary need warranted  Stoffenmanager Nano  Task weighted  E, extreme hazard class 3, high EP I, high-risk priority  D, very high hazard class 3, high EP I, high-risk priority  E, extreme hazard class 3, high EP I, high-risk priority  D, very high hazard class 2, average EP II, medium-risk priority  E, extreme hazard class 3, high EP I, high-risk priority  E, extreme hazard class 3, high EP I, high-risk priority    Time and frequency weighted  E, extreme hazard class 2, average EP I, high-risk priority  D, very high hazard class 3, high EP II, medium-risk priority  E, extreme hazard class 2, average EP I, high-risk priority  D, very high hazard class 2, average EP II, medium-risk priority  E, extreme hazard class 2, average EP I, high-risk priority  E, extreme hazard class 2, average EP I, high-risk priority  NanoSafer  Estimated hazard level  0.59  1  0.59  0.761  0.539  0.488    Near field  Acute: 0.1267 RL2: low toxicity/low EP EB2: low EP  Acute: 0.0777 RL4: high toxicity/high EP EB1: very low EP  Acute: 1.056 RL5: very high toxicity/moderate- to-very high EP EB5: very high EP  Acute: 0.0227 RL4: high toxicity/high EP EB1: very low EP  Acute: 3.2 RL5: very high toxicity/moderate- to-very high EP EB5: very high EP  Acute: 0.3873 RL4: high toxicity/high EP EB3: moderate EP    Daily: 0.1275 RL2: low toxicity/low EP EB2: low EP  Daily: 0.0782 RL4: high toxicity/high EP EB1: very low EP  Daily: 1.063 RL5: very high toxicity/moderate- to-very high EP EB5: very high EP  Daily: 0.0229 RL4: high toxicity/high EP EB1: very low EP  Daily: 3.22 RL5: very high toxicity/moderate- to-very high EP EB5: very high EP  Daily: 0.3899 RL4: high toxicity/high EP EB3: moderate EP  View Large Reviewing the results of the evaluation shown in Table 5 illustrates the differences between both the ENMs and the CB strategies evaluated by keeping the handling/use scenarios constant, as discussed in Methods. For the CB Nanotool, the RL ranged from RL4—seek specialist advice for CNTs to RL3 for titanium dioxide, nanoscale silver, graphene, and nanocellulose to RL2 for silicon dioxide. For these ENMs, the lowest hazard (severity) score was for SiO2, while the highest was for CNTs. The primary difference that resulted in the differing RL bands was the severity (HB) score, which placed all ENMs except for SiO2 in the high-severity category. Stoffenmanager Nano indicated that SiO2, CNTs, graphene, and nanocellulose are overall a high-risk priority. TiO2 and nanosilver were both considered a high-risk priority when task-weighted but considered to be a medium-risk priority when the time and frequency of handling were taken into account indicating a lower overall risk. For the Precautionary Matrix, evaluation indicated that a risk is present for both workers and the environment based on a final calculated score of over 20 for all ENMs evaluated. For both the ANSES tool and ISO guidance, all ENMs fell into the same hazard and exposure bands resulting in similar control band—CB5—full containment and requiring expert advice. For the EC Guidance, nanosilver, CNTs, silica, and graphene fell into the highest RL resulting in the recommendation to adopt process-based control measures. TiO2 and nanocellulose were at the next lowest level, which recommended the use of closed systems or containment of the process. For the GoodNanoGuide, SiO2, graphene, nanocellulose, and nanoscale silver were put into the highest hazard grouping because of lack of information on the health effects associated with these ENMs. TiO2 and CNTs were placed into a lower hazard group because of the availability of hazard data (NIOSH, 2011). Finally, for NanoSafer, CNTs and graphene fell into the highest RL resulting in the recommendation for a fume hood/enclosure/glovebox with supplied air respirator. Nanocellulose, silica (amorphous), nanosilver, and TiO2 were at the next lowest level, which recommended the use of highly efficient local exhaust ventilation, fume hood, or glovebox along with a respirator. The evaluation of the repeated exposure data in rodents for those ENMs with these data showed that the lowest observed adverse effect levels were all <20 mg/m3, which is the level of concern for chronic adverse effects (STOT-RE). Based on these results, the HB would be either ‘category D—serious hazard’ according to the ANSES and ISO or ‘category 1—health hazard—danger’ based on the GHS and US OSHA hazard banding strategies for nanoscale amorphous silica, TiO2, silver, and multiwalled CNTs. These HBs are similar to or lower than the equivalent concentrations at the OELs (Table S1, available at Annals of Occupational Hygiene online). Discussion Relatively limited evaluation and validation have been performed on the available CB tools for ENMs. This study adds to the current scientific literature by providing a systematic evaluation and application of all eight of the currently available CB tools for ENMs, using six different types of ENMs of varying dustiness level, for a fixed exposure and use scenario in the workplace. Outcomes are examined across the CB tools and compared with the proposed OELs for these ENMs. Data gaps in the key inputs to these CB tools are identified. Finally, the drivers for these outcomes are identified, and research needs are suggested to improve the information available and the utility of these CB tools for making workplace exposure control decisions. Recent articles by Eastlake et al., Ligouri et al., and Sanchez Jiménez et al. are complementary with this article but also differ in both approach and scope (Eastlake et al., 2016; Liguori et al., 2016; Sánchez Jiménez et al., 2016). Eastlake et al. (2016) provide a systematic review of the ENM-specific CB tools and conclude that few of these tools have been validated with regard to their effectiveness in controlling exposure. Ligouri et al. provide an update of the earlier review by Brouwer (Brouwer, 2012), including a more in-depth description of those tools. Ligouri et al. review six of the eight CB tools examined in this article (which also includes GoodNanoGuide and ISO/TS 12901–2). Ligouri et al. also describes the different inputs and possible outputs of the CB tools, but they did not conduct any actual evaluations on ENMs as performed in this article on a set of six ENMs. These evaluations show that differences in the particle properties can influence the outcomes of the different CB tools, depending on how a particular property is treated in the various hazard and exposure banding approaches. The Sanchez Jiménez et al. article provides a broad evaluation of four of these CB tools, including sensitivity analyses of the tool inputs and limited exposure validation testing using airborne number concentration data on one ENM (cloisite) and three processes. The Sanchez Jiménez et al. article focused on assessment of the tools geared more to researchers, whereas this article provides step-by-step information and examples that may be useful to the practitioner in selecting CB tools, gathering the input information, and assessing the usefulness of the results. Evaluation of CB tool outcomes The findings of this current evaluation show that the ANSES and ISO tools recommended the highest level of exposure control for the majority of ENMs in this use scenario (Table 6). CB Nanotool, EC Guidance, NanoSafer, and GoodNanoGuide recommended lower levels of control by ENM. The CB resulted in either the same or higher levels of exposure control to those suggested by the proposed OELs for nanoscale TiO2 and CNTs (Table 6). CNTs were generally in the most protective band ‘seek expert advice’ with a controls performance level of <1 µg/m3. In contrast, the recommended control bands for silica and graphene differ widely between CB Nanotool and EC Guidance, i.e. either level 2 or level 4, respectively, in this evaluation. The proposed OELs for CNTs also vary over two or more orders of magnitude and control bands. However, these OELs for ENMs are all lower on a mass basis than their bulk counterparts (Table S1, available at Annals of Occupational Hygiene online). The EC Guidance and NanoSafer recommended a similar or higher level of exposure control to that based on the OELs proposed for CNTs, TiO2, and silver, whereas CB Nanotool recommendations were either higher (TiO2 and CNTs) or lower (silver) in this scenario compared with the proposed OELs (Table 6). The ISO and ANSES tools required the most complete hazard data and yielded the highest level of exposure control. It is useful to the practitioner to understand how the input data can influence the CB findings, which factors are most influential on these results, and how these findings compare to existing OELs. Table 6. CB recommendations for the nanomaterials evaluated, compared with recommendations that align with proposed OELs. CB tool  Recommended control bands and performance-based exposure rangesa    Seek specialist advice/ adopt special measures  Containment  Engineering controls (fume hoods or LEV)    <1 µg/m3  1–10 µg/m3  10–1000 µg/m3  Recommended control approaches by CB tool  CB Nanotool  CNTs  Graphene Nanocellulose Silver Titanium dioxide  Silica (amorphous)  GoodNanoGuide  Graphene Nanocellulose Silica (amorphous) Silver  CNTs Titanium dioxide    ANSES  CNTs Graphene Nanocellulose Silica (amorphous) Silver Titanium dioxide      ISO  CNTs Graphene Nanocellulose Silica (amorphous) Silver Titanium dioxide      EC Guidance  CNTs Graphene Silica (amorphous) Silver  Nanocellulose Titanium dioxide    NanoSafer  CNTs Graphene  Nanocellulose Silver Titanium dioxide  Silica (amorphous)  Recommended control approaches that align with OELsb    Silver  CNTs  CNTs Silica (amorphous) Titanium dioxide  CB tool  Recommended control bands and performance-based exposure rangesa    Seek specialist advice/ adopt special measures  Containment  Engineering controls (fume hoods or LEV)    <1 µg/m3  1–10 µg/m3  10–1000 µg/m3  Recommended control approaches by CB tool  CB Nanotool  CNTs  Graphene Nanocellulose Silver Titanium dioxide  Silica (amorphous)  GoodNanoGuide  Graphene Nanocellulose Silica (amorphous) Silver  CNTs Titanium dioxide    ANSES  CNTs Graphene Nanocellulose Silica (amorphous) Silver Titanium dioxide      ISO  CNTs Graphene Nanocellulose Silica (amorphous) Silver Titanium dioxide      EC Guidance  CNTs Graphene Silica (amorphous) Silver  Nanocellulose Titanium dioxide    NanoSafer  CNTs Graphene  Nanocellulose Silver Titanium dioxide  Silica (amorphous)  Recommended control approaches that align with OELsb    Silver  CNTs  CNTs Silica (amorphous) Titanium dioxide  aEstimated from CB approaches shown in Fig. 1; these correspond to the OEL concentration ranges (also called OEBs) associated with the hazard categories in the ANSES (2010) and ISO (2014) CB tools. Note that some control band categories (Control Level 2 and Control Level 3) have been combined for ANSES and ISO to make the control bands’ results consistent between tools. And the control recommendations provided by NanoSafer differ from categories provided here (e.g. LEV, containment, special precautions) and include recommendations on respiratory protection (see Fig. 2). bBased on proposed OELs (Table S1, available at Annals of Occupational Hygiene online) and corresponding performance-based exposure ranges shown in this table. View Large The primary drivers for the control bands were the hazard scores in this small-scale production scenario. The hazards scoring approach used by the CB Nanotool resulted in the ranking of several of these ENMs to lower overall control bands than other decision tree tools (ISO, ANSES, Stoffenmanager Nano). The CB Nanotool approach combines scores for all hazards for the ENM and parent material into a total composite score, so positive research findings in any hazard category (carcinogenicity, mutagenicity, reproductive toxicity) do not automatically drive the ENM to the highest control band like the decision tree tools. For instance, for inhaled TiO2, the International Agency for Research on Cancer (IARC) has classified this chemical agent as a 2B (possibly carcinogenic to humans), which automatically places it in the highest hazard class for ISO, ANSES, NanoSafer, and Stoffenmanager Nano. However, when considered with all of the other hazard categories, TiO2 was scored as high severity (band 3 of 4) in CB Nanotool resulting in the containment control band. It is difficult to determine whether a more precautionary approach (as provided by ANSES, ISO, NanoSafer, and Stoffenmanager Nano) is a better choice given the lack of full hazard data on these ENMs. The best assessment that can be made at this point is to compare these tools to published risk assessments, which derive OELs based on a more thorough hazard analysis. However, variability in the proposed OELs for these ENMs also results in uncertainty in the appropriate level of exposure control (as shown in Table 6 and Table S1, available at Annals of Occupational Hygiene online). The primary driver for the different control bands by the EC Guidance tool was also the differences in hazard assessment. All of the ENMs evaluated were insoluble in water (based on the SDS or technical data sheets), but for CNTs and nanocellulose, these differences were due to the fibrous geometry/shape of these materials. And finally, the lower control bands for CNTs and TiO2 by the GoodNanoGuide were driven by availability of information on these ENMs. The GoodNanoGuide tool categorizes hazard groups by ‘known to be inert’ (hazard group A), ‘understand reactivity/function’ (hazard group B), or ‘unknown hazard’ (hazard group C). So an ENM such as TiO2 would be a group B because information is available, including that it has been classified as possibly carcinogenic by IARC and NIOSH, whereas other ENMs would be at a higher level because there is little or no information available on their hazard. That feature of GoodNanoGuide is mainly driven by EP (based on material form and task duration) and does not consider hazard potential in depth. The minimal assessment of hazard may limit the utility of the tool. Despite some differences in approach, most tools gave similar exposure bands (typically medium–high EP) primarily because the use scenario was consistent among these six ENMs evaluated. In general, exposure bands are driven by three primary factors: (i) material form; (ii) amount of material used; and (iii) process/task. In addition, all models except the Precautionary Matrix utilize dustiness as a factor in determining EP. Stoffenmanager Nano and NanoSafer, however, use much more detailed exposure models utilizing parameters such as process energy, volume, and ventilation rate of the work room, as well as frequency and duration of the evaluated task. With the CB Nanotool, the exposure probability score differed primarily on the dustiness determination of the ENM used and ranged from ‘probable exposure’ (CNT, high dustiness) to ‘likely exposure’ (TiO2, SiO2, medium dustiness; graphene, nanocellulose, unknown dustiness; and silver, low dustiness). Although dustiness can be a differentiating factor in these tools, another important factor is the amount of ENM used (by mass), and in this area, the tools differ considerably. For ISO, the highest exposure factor related to amount handled is applied when using >1 kg of powdered nanomaterials. However, with CB Nanotool, EC Guidance, and Precautionary Matrix, the highest material quantity category is bounded at a much lower level, i.e. less than 1 g of the nanomaterial. An evaluation of a few of these CB tools on a different set of nanoscale and microscale particles also showed a range of hazard and exposure outputs across tools and concluded that some of the recommendations may be excessive in some situations (Sánchez Jiménez et al., 2016). In general, the more specific and complete the input information, the more accurate and useful the CB tool outcomes would be expected, although the structure and flexibility of the tools to utilize specific parameter data (e.g. dustiness) also differ across tools. Such evaluations provide useful insights into the performance of these tools for the practitioner to gain an understanding of the utility and limitations of these various tools. Existing OELs for ENMs examined in CB tools One way to evaluate the utility and validity of the outcomes of these CB tools is to compare their recommended controls and associated performance levels with the OELs that have been proposed for these same or similar ENMs (as discussed above). OELs proposed by nonregulatory governmental agencies or by nongovernmental organizations (Table S1, available at Annals of Occupational Hygiene online) include nanoscale titanium dioxide, silica, silver, CNTs, and cellulose, which are all examined in this article. No published OELs for graphene were found in the literature or reported in a recent systematic review of ENM OELs (Mihalache et al., 2017). The OELs for nanoscale particles are typically lower airborne mass concentrations than the closest applicable regulatory OELs (Table S1, available at Annals of Occupational Hygiene online). For example, 5 mg/m3 is the OSHA permissible exposure limit for either graphite (synthetic), particles not otherwise regulated, or cellulose (respirable fraction, 8-h TWA concentration) (OSHA, 1983). This exposure concentration has been used in some nanotoxicology studies (e.g. for SWCNTs) (Shvedova et al., 2008). An OEL of 5 mg/m3 (i.e. 5000 µg/m3) is approximately one to three orders of magnitude greater than the proposed OELs for carbonaceous, metal, or metal oxide nanoparticles (Table S1, available at Annals of Occupational Hygiene online). Differences in both the toxicity of the substance and the data and methods used to derive the OELs could contribute to these differences. Some of the existing OELs may have included information on nanoscale particle exposures (although possibly not defined as such). For example, high combustion processes such as silver refining can produce airborne nanoscale particles (Miller et al., 2010). NIOSH recommended separate mass-based OELs for titanium dioxide by particle size (nanoscale/ultrafine and microscale/fine) (Table S1, available at Annals of Occupational Hygiene online) (NIOSH, 2011). The pulmonary toxicity of titanium dioxide and other poorly soluble particles is correlated with the total particle surface area, which is greater for an equal mass of smaller particles (NIOSH, 2011). Ease of use of CB tools for ENMs During the course of this study, several observations emerged regarding the user-friendly nature of the various tools. In particular, the level of information required and the complexity in completing the assessments differ among these tools. For quick, high-level assessments, the GoodNanoGuide, EC Guidance, and Precautionary Matrix provide results with minimal data. These tools were the easiest to complete given the minimal level of information required for the evaluation. EC Guidance tool categorizes nanomaterial hazard solely based on the physicochemical properties of biopersistence and particle/fiber shape, whereas GoodNanoGuide includes three simple bands for physicochemical properties, known to be inert, reactivity/function known, or unknown properties. CB Nanotool utilizes an intermediate level of information on both hazard and EP, which is at a level that would generally be available in a well-documented SDS. The hazard scoring approach of CB Nanotool is relatively easy to use by answering yes, no, or unknown to the hazard questions and assigning a score. CB Nanotool quantitatively addresses a lack of information by including ‘unknown’ as a choice, which defaults to a containment recommendation when no data are available. The transparent scoring approach in CB Nanotool allows the user to easily assess the drivers of the control band results to explore where changes to materials or use parameters (quantity, material form, etc.) could impact the control band. ISO, ANSES, Stoffenmanager Nano, and NanoSafer require more detailed information, and each of the sources shown in Table 4 was used to complete these assessments (to the extent that data were available). The ANSES and ISO tools are similar to each other and use the GHS system to provide a ready basis for standardization of inputs to hazard banding, which is useful but also may require more toxicology expertise [e.g. identifying lethal dose (LD50) and other endpoints] than does CB Nanotool, which includes yes/no options for the main endpoints. The ANSES and ISO tools address lack of information in the hazard banding by defaulting to the highest HB, which results in recommendations for higher levels of exposure control. The Stoffenmanager Nano and Precautionary Matrix are different in scope compared with the other tools because they address risk prioritization and do not lead to a control band. The Stoffenmanager Nano and NanoSafer tools utilized the most complex exposure banding approach requiring the most information from the user, including amount of material, process duration and frequency, work room volume, and ventilation rate among other parameters. The Precautionary Matrix assesses hazard potential through two primary physicochemical parameters: redox/catalytic activity and stability (half-life) in the body/environment. It provides a table of reactivity information for 12 nanomaterials. Stoffenmanager Nano provides guidance on hazard banding for 19 commonly used nanomaterials (Duuren-Stuurman et al., 2011, 2012). However, for those nanomaterials not included in the table, the HB is derived from an assessment of hazards based on the parent material. If the HB of the parent material is not known (or the material is not characterized according to carcinogenicity, mutagenicity, reproductive, and/or developmental effects), the tool defaults to the highest HB. Finally, several of the tools provide online or downloadable spreadsheets to help guide the user through the process. CB Nanotool provides a downloadable score-based spreadsheet with examples to help guide the user through the process. Stoffenmanager Nano, NanoSafer, and Precautionary Matrix have online tools to help facilitate the process. The various parameters and inputs to these tools, including those used in these assessments, are summarized in Tables 2 and 3. These input parameters are valuable information that are needed to use these CB tools to arrive at the recommended control bands. The parameters that were found to be drivers of the control band findings (e.g. availability of dustiness or specific health effects data), as discussed in this article, could be considered essential to obtaining more useful and reliable results from these CB tools. Evidence available for evaluating CB tools for ENMs Only a few types of ENMs have undergone relatively extensive toxicological evaluation, e.g. TiO2 and CNTs. Even for these ENMs, significant data gaps remain, especially for chronic adverse health effects. The limited hazard and dustiness data make it challenging to provide relevant information for the SDS. In addition, SDSs are not uniform and provide variable inconsistent amounts of information (Eastlake et al., 2012). A useful addition to SDSs would be a standardized format for CB tool input factors, which would provide the practitioner with more readily available information for applying CB methods to specific ENMs. In particular, the inclusion of standard information needed in CB tools would be useful information in SDSs. Current toxicity data, where available, would be especially useful in the SDSs, including the adverse effect levels in rodent studies, to evaluate severity and potency. In the future, the development of default HBs or OEBs for ENMs-based physicochemical properties and limited toxicology data would help facilitate the determination of appropriate control bands (Kuempel et al., 2012). In general, regardless of the CB strategy used, the uncertainty of the potential health risks of ENMs tends to result in a higher level of exposure control than would be used based on the ENM-specific OELs. These higher levels of exposure control appear to be due to the limited data on ENMs for many of the inputs in the CB tools, resulting in the default to the more protective categories in the absence of specific information. Indeed, a utility of these CB tools is that they generally recommend a high level of exposure control in the absence of specific information, which is a protective default. This approach is consistent with using greater precaution in the absence of data (Schulte and Salamanca-Buentello, 2007). Such strategies also encourage research to provide the more specific data needed to replace default assumptions. On the other hand, CB tools that do not discern among the hazards based on available data may not be sufficient for decision-making. This analysis has shown that certain factors that drive the CB decisions (e.g. default toxicity assumptions; dustiness levels) would be useful priorities for future research in order to improve the evidence basis for the application of these CB tools for ENMs. Dustiness data would be also useful in future validation studies as well as research studies correlating exposure with dustiness levels of ENMs by job task. Possible limitations in this analysis include the limited number of ENMs evaluated (six). OELs have been proposed for five of these ENMs (Table S1, available at Annals of Occupational Hygiene online). Using the proposed OELs for comparison to the HBs/OEBs is an uncertain criterion because the proposed OELs can vary widely and none are regulatory limits. Because the workplace use scenario was kept constant in this analysis, the findings may not apply to other use scenarios. Finally, the performance-based exposure concentrations (Table 6) have not been fully validated for the specific engineering control options, and comparison of the recommendations can be challenging because of the overlapping control bands across the CB tools. It should be noted that all of the CB tools evaluated recommended at a minimum the use of local exhaust ventilation for each of these ENMs (Table 6) in this exposure scenario (dry powder handling of small quantities for 4 h or fewer per day). For those ENMs with proposed OELs (Table 6), the associated performance-based exposure concentrations would also necessitate the use of local exhaust ventilation or a higher level of control. The most comprehensive validation studies performed to date have been on the CB Nanotool (Paik et al., 2008; Zalk et al., 2009), as discussed in an earlier systematic review (Eastlake et al., 2016). In a study of 32 job activities and nanomaterial combinations, the exposure control recommendations from CB Nanotool were reported to be at the same or higher level to those recommended independently by an experienced industrial hygienist for 28 (~88%) of the job activities. Roughly similar results were seen in this study, in which the control band recommendations from CB Nanotool were the same or lower than three of the four (75%) of the ENM OELs (Table 6). By comparison, the CB recommendations of EC Guidance, NanoSafer, and GoodNanoGuide were all the same or lower than the ENM OELs, whereas the ANSES and ISO tools recommended the lowest exposure level for each of the ENMs (Table 6). Sánchez Jiménez et al. (2016) provided some limited validation testing of the hazard and exposure results for three of these CB tools. They reported various differences in both the hazard and exposure results of the CB tools compared with reported toxicity and exposure measurement data. For example, the measured airborne number concentrations for cloisite (the only ENM in that evaluation) were lower for a weighing task, but higher for an extrusion task, compared with the results from the three CB tools (Sánchez Jiménez et al., 2016). A limitation in the validation studies to date is either the lack of data or the limited data on airborne exposure concentrations of ENMs associated with job activities and exposure controls; these data are needed to verify that the recommended controls achieved the expected results. Verification of CB tool recommendations with field-based measurements across jobs/tasks and working conditions has been previously recommended for general chemicals in industry (Jones and Nicas, 2006a, b). In addition, the lack of OELs for many ENMs does not permit verification that the recommended exposure control levels would be protective of workers’ health. An evaluation of CB tool recommendations with ENM-specific OELs, as illustrated in this article (Table 6), could be extended to additional ENMs as more toxicology data and OELs become available. Key findings This study demonstrated the use of the eight CB tools for ENMs currently available, showed what input data are needed, suggested several useful sources and websites to search for the information needed, and demonstrated the application and outcomes in a case study of six different ENMs, most of which have proposed OELs. A fixed workplace exposure scenario allowed focus on the role of the properties of the ENMs themselves, both biological and physicochemical. The key biological input parameters include qualitative hazard information and quantitative effect levels of the ENMs or bulk material [OELs or No Observed Adverse Effect Levels (NOAELs)]. The key physicochemical input parameters include dustiness, surface activity, shape (fibrous or not), and solubility. Understanding the information needed to utilize these tools and comparing the findings across these tools for a set of ENMs and fixed workplace exposure conditions helps the practitioner better understand how to select and use these tools. The purpose of this article is not to recommend the use of any specific tool but to illustrate and compare the inputs and findings of each tool under the same ENM and exposure scenarios. The findings of this study provide further input into the key drivers for the findings of each of these tools. Ultimately the selection of a tool depends on the purpose of the evaluation (e.g. risk prioritization or exposure control selection) as well as the availability of the input information. In using these tools, the practitioner will find different levels of information needed and complexity in completing the assessments. For quick, high-level assessments, the GoodNanoGuide and EC Guidance provide results with minimal data. Precautionary Guidance is the most basic indicating the need for caution. CB Nanotool utilizes a moderate level of hazard and EP information, while being implemented through an easy to understand tool. ISO, ANSES, Stoffenmanager Nano, and NanoSafer use the most information and require more effort in collecting data and completing the assessments. Regardless of which tool is selected, the user should record the sources of information and the input parameters selected in the application of any of these tools. This practice is consistent with good recordkeeping of the information used to arrive at CB findings and also facilitates further evaluation when new information becomes available. Fig. 2 presents an overall approach for using and evaluating the tools, whereas Tables 2 and 3 provide a template of the key information needed for each of these tools. Likewise, Table 4 provides a guide to online databases where input information on hazard inputs can be gathered. The findings of this study provided limited validation testing of CB tool results compared with OELs proposed for four of the ENMs evaluated in this study. These findings confirm those of other studies (Eastlake et al., 2016; Sánchez Jiménez et al., 2016) that more information is needed to validate these CB tools in order to determine whether the use of CB can adequately reduce nanomaterial worker exposures to safe levels. Moreover, the inclusion of the basic information needed in CB reported in a standard format on the SDSs would be especially useful in the application of these tools. Research to provide basic toxicity data is required to fill those data gaps. The following data gaps were identified in this study, which if filled would reduce uncertainty and improve confidence in the reliability of CB tool findings: The amount of information required differs across tools; yet most of the tools recommended higher levels of exposure control for each of the ENMs in this evaluation compared with the proposed OELs, primarily because of the limited hazard data. The default for ISO, ANSES, and Stoffenmanager Nano to the highest RL based on the highest individual hazard category (carcinogenicity, reproductive toxicity, mutagenicity) and for unknown hazards resulted in the highest priority or highest level of exposure control for each of the ENMs in this assessment. The composite hazard scoring approach by CB Nanotool resulted in lower levels of exposure control for some of the ENMs in this assessment. EC Guidance, NanoSafer, and GoodNanoGuide also recommended some diversity in the CB recommendations across ENMs. All tools recommended the use of local exhaust ventilation, at a minimum, for working with any of these ENMs. Research needs The lack of available data for the main inputs into these tools significantly reduces their utility, at this time. Key information that drives the hazard, exposure, and CB recommendations would be the most useful to reducing uncertainty and increasing confidence in the application of these tools, including: Quantities of ENMs currently produced and used in various applications by job/task needs to be updated and made available to researchers and the practitioner for use in emission potential scoring in CB. Correlation of ENM quantity, dustiness, and process with exposure should be assessed and validated with laboratory and workplace data. Airborne exposure measurement data for specific job activities and ENMs could be used in validation testing of the CB recommendations. More specific information is needed to help classify ENMs according to the hazard and exposure parameters. For instance, relatively minimal information on the surface reactivity and dustiness of ENMs would be useful to classify these as low, medium, or high, as requested in several of these tools (some do not include dustiness while Stoffenmanager Nano and NanoSafer have quantitative dustiness categories). Further evaluation and refinement of the hazard categories for ENMs are needed to reduce uncertainty given the limited toxicity data for ENMs. Several research efforts are underway in the USA and other countries to group ENMs by hazard potential and could ultimately provide default HBs or OEBs using physicochemical properties and limited toxicology data. In the meantime, the practitioner needs to be aware that while these CB tools can be useful in decision-making about exposure control options when working with ENMs, it is also important when selecting a tool to consider the objective, the information needed, and level of validation. Ideally, more than one tool should be selected for comparison of findings and to better inform decision-making. In addition, comparison of the CB results with any ENM-specific OELs would also be useful. Updating the initial evaluations as new data or tools become available will provide continued improvement in the CB of ENMs. Conclusions The several CB tools that have been developed for nanomaterials represent a good first step in developing approaches to control worker exposure given the paucity of data on many ENMs in use. Findings of this study showed that the ISO, ANSES, and Stoffenmanager Nano tools recommended the highest level of risk or exposure control for each of the ENMs in this assessment, whereas CB Nanotool, EC Guidance, NanoSafer, and GoodNanoGuide recommended more diverse CB recommendations across ENMs. Further validation of these tools is needed, including by comparing the performance-based exposure ranges of control approaches to the measurements of airborne exposure concentration of ENMs in a worker’s breathing zone during typical job tasks. Research towards characterizing dustiness of more ENMs will help improve the utility of these tools. Efforts should continue to synthesize data from workplace studies to gain a better understanding of how well factors such as dustiness represent worker exposures. In addition, as more health hazard data become available, for ENMs individually or within similar physicochemical groups, the ability to provide more constructive exposure control guidance on the range of ENMs seen in the workplace will improve. The findings from this study show that significant data gaps remain, resulting in uncertainty about the optimal selection of controls to protect workers producing and handling ENMs. Research that focuses on providing the key data inputs for these CB tools and including standard information on SDSs would facilitate the utility of these tools. In most of the evaluated CB tools, uncertainty in the available data is managed by the selection of higher RLs and more protective exposure control options. An important finding of this evaluation is that local exhaust ventilation was recommended at a minimum to control exposures to ENMs in the workplace. More stringent controls, such as process containment, may come at a higher installation or maintenance cost, and it may not be certain whether these are necessary given the unknown risks. However, these CB tools generally appear to be providing prudent exposure control guidance in the face of uncertainty. Supplementary Data Supplementary data are available at Annals of Work Exposures and Health online. Declaration The authors declare no conflict of interest relating to the material presented in this article. The findings and conclusions in this article are those of the author(s) and do not necessarily represent the official position of the National Institute for Occupational Safety and Health, Centers for Disease Control and Prevention. Mention of a specific product or company does not constitute endorsement by the Centers for Disease Control and Prevention. Acknowledgements We would like to acknowledge the assistance of D. Hammond, T.J. Lentz, J.L. Topmiller, and M. Gressel for their careful review of this article. We also thank Vanessa Williams for assistance in preparing the figures. This research was supported through the NIOSH Nanotechnology Research Center, the Division of Applied Research and Technology, and the Education and Information Division. We would like to thank the anonymous reviewers for their excellent comments and helpful suggestions on earlier drafts of this article. References Ader AW, Farris JP, Ku RH. ( 2005) Occupational health categorization and compound handling practice systems—roots, application and future. Chem Health Saf ; 12: 20– 26. Google Scholar CrossRef Search ADS   ANSES. ( 2010) Development of a specific control banding tool for nanomaterials. Maisons-Alfort Cedex: French Agency for Food, Environmental and Occupational Health & Safety (ANSES). Request no. 2008-SA-0407. Available at https://www.anses.fr/en/system/files/AP2008sa0407RaEN.pdf. Accessed 4 April 2016. Brooke IM. ( 1998) A UK scheme to help small firms control health risks from chemicals: toxicological considerations. Ann Occup Hyg ; 42: 377– 90. Google Scholar CrossRef Search ADS PubMed  Brouwer DH. ( 2012) Control banding approaches for nanomaterials. Ann Occup Hyg ; 56: 506– 14. Google Scholar PubMed  BSI. ( 2007) Nanotechnologies—Part 2: Guide to safe handling and disposal of manufactured nanomaterials. Book Nanotechnologies—Part 2: Guide to safe handling and disposal of manufactured nanomaterials . London: British Standards Institution. Dahm MM, Evans DE, Schubauer-Berigan MKet al.  ( 2012) Occupational exposure assessment in carbon nanotube and nanofiber primary and secondary manufacturers. Ann Occup Hyg ; 56: 542– 56. Google Scholar PubMed  Duuren-Stuurman B, Vink S, Brouwer Det al.   ( 2011) Stoffenmanager Nano: description of the conceptual control banding model . Zeist, Netherlands: Netherlands Organisation for Applied Scientific Research (TNO). Duuren-Stuurman B, Vink SR, Verbist KJet al.  ( 2012) Stoffenmanager Nano version 1.0: a web-based tool for risk prioritization of airborne manufactured nano objects. Ann Occup Hyg ; 56: 525– 41. Google Scholar PubMed  Eastlake A, Hodson L, Geraci Cet al.  ( 2012) A critical evaluation of material safety data sheets (MSDSs) for engineered nanomaterials. Chem Health Saf ; 19: 1– 8. Google Scholar CrossRef Search ADS PubMed  Eastlake A, Zumwalde R, Geraci C. ( 2016) Can control banding be useful for the safe handling of nanomaterials? A systematic review. J Nanopart Res ; 18: 1– 24. Google Scholar CrossRef Search ADS   European Commission. ( 2014) Guidance on the protection of the health and safety of workers from the potential risks related to nanomaterials at work: guidance for employers and health and safety practitioners . Belgium: European Commission, Employment Social Affairs & Inclusion. Available at http://ec.europa.eu/social/home.jsp?langId=en. European Committee for Standardization (CEN). ( 2013) Workplace exposure. Measurement of the dustiness of bulk materials. Requirements and choice of test methods (EN 15051-1:2013) . Brussels: CEN. Evans DE, Turkevich LA, Roettgers CTet al.  ( 2012) Dustiness of fine and nanoscale powders. Ann Occup Hyg ; 57: 261– 77. Google Scholar PubMed  Future Markets Inc.©( 2013) The global nanotechnology and nanomaterials industry . Future Markets, Inc., Technology Report No. 68. Available at https://futuremarketsinc.com/. Gardner RJ, Oldershaw PJ. ( 1991) Development of pragmatic exposure-control concentrations based on packaging regulation risk phrases. Ann Occup Hyg ; 35: 51– 9. Google Scholar PubMed  Garrod AN, Rajan-Sithamparanadarajah R. ( 2003) Developing COSHH Essentials: dermal exposure, personal protective equipment and first aid. Ann Occup Hyg ; 47: 577– 88. Google Scholar PubMed  Kulinowski KM, Jaffe MP. (2009) The goodnanoguide: a novel approach for developing good practices for handling engineered nanomaterials in an occupational setting. Nanotech. L. & Bus.; 6: 37–44. Henry BJ, Schaper KL. ( 1990) PPG’s Safety and Health Index System: a 10-year update of an in-plant Hazardous Materials Identification System and its relationship to finished product labeling, industrial hygiene, and medical programs. Am Ind Hyg Assoc J ; 51: 475– 84. Google Scholar CrossRef Search ADS PubMed  Höck J, Epprecht T, Furer Eet al.   ( 2013) Guidelines on the precautionary matrix for synthetic nanomaterials, version 3.0 . Berne: Federal Office for Public Health and Federal Office for the Environment. HSE. ( 2009) COSHH Essentials . Liverpool, England: Health and Safety Executive. ISO. ( 2014) ISO/TS 12901–2:2014 Nanotechnologies—occupational risk management applied to engineered nanomaterials—Part 2: Use of the control banding approach . Geneva, Switzerland: International Organization for Standardization. Isogai A. ( 2013) Wood nanocelluloses: fundamentals and applications as new bio-based nanomaterials. J Wood Sci .; 59: 449– 59. Google Scholar CrossRef Search ADS   Jensen KA, Saber AT, Kristensen HV, Koponen IK, Liguori B, Wallin H. (2013) NanoSafer vs. 1.1-nanomaterial risk assessment using first order modeling. In 6th International Symposium on Nanotechnology, Occupational and Environmental Health 2013 Oct 28 (Vol. 120). Jones RM, Nicas M. ( 2006a) Evaluation of COSHH Essentials for vapor degreasing and bag filling operations. Ann Occup Hyg ; 50: 137– 47. Jones RM, Nicas M. ( 2006b) Margins of safety provided by COSHH Essentials and the ILO Chemical Control Toolkit. Ann Occup Hyg ; 50: 149– 56. Kuempel ED, Castranova V, Geraci CLet al.  ( 2012) Development of risk-based nanomaterial groups for occupational exposure control. J Nanopart Res ; 14: 1029. Google Scholar CrossRef Search ADS PubMed  Lee JH, Kwon M, Ji JHet al.  ( 2011) Exposure assessment of workplaces manufacturing nanosized TiO2 and silver. Inhal Toxicol ; 23: 226– 36. Google Scholar CrossRef Search ADS PubMed  Liguori B, Hansen SF, Baun Aet al.   ( 2016) Control banding tools for occupational exposure assessment of nanomaterials—ready for use in a regulatory context? NanoImpact ; 2: 1– 17. Google Scholar CrossRef Search ADS   Maynard AD. ( 2007) Nanotechnology: the next big thing, or much ado about nothing? Ann Occup Hyg ; 51: 1– 12. Google Scholar PubMed  McKernan LT, Seaton M. ( 2014) The banding marches on: NIOSH proposes a new process for occupational exposure banding. The Synergist® ; 25: 44– 46. Mihalache R, Verbeek J, Graczyk Het al.  ( 2017) Occupational exposure limits for manufactured nanomaterials, a systematic review. Nanotoxicology ; 11: 7– 19. Google Scholar CrossRef Search ADS PubMed  Miller A, Drake PL, Hintz Pet al.  ( 2010) Characterizing exposures to airborne metals and nanoparticle emissions in a refinery. Ann Occup Hyg ; 54: 504– 13. Google Scholar PubMed  Naumann BD, Sargent EV, Starkman BSet al.   ( 1996) Performance-based exposure control limits for pharmaceutical active ingredients. Am Ind Hyg Assoc J ; 57: 33– 42. Google Scholar CrossRef Search ADS PubMed  NIOSH. ( 2009a) Approaches to safe nanotechnology: managing the health and safety concerns associated with engineered nanomaterials . Cincinnati, OH: U.S. Department of Health and Human Services, Public Health Service, Centers for Disease Control and Prevention, National Institute for Occupational Safety and Health, DHHS (NIOSH). Publication No. 2009–125. NIOSH. ( 2009b) Quantitative risk characterization and management of occupational hazards: control banding (CB)—a literature review and critical analysis . Cincinnati, OH: U.S. Department of Health and Human Services, Public Health Service, Centers for Disease Control and Prevention, National Institute for Occupational Safety and Health, DHHS (NIOSH). Publication 2009–152. NIOSH. ( 2011) Current Intelligence Bulletin 63: occupational exposure to titanium dioxide . Cincinnati, OH: U.S. Department of Health and Human Services, Public Health Service, Centers for Disease Control and Prevention, National Institute for Occupational Safety and Health, DHHS (NIOSH). Publication No. 2011–160. NIOSH. ( 2012) General safe practices for working with engineered nanomaterials in research laboratories . Cincinnati, OH: U.S. Department of Health and Human Services, Centers for Disease Control and Prevention, National Institute for Occupational Safety and Health, DHHS (NIOSH). Publication No. 2012–147. NIOSH. ( 2013a) Current Intelligence Bulletin 65: occupational exposure to carbon nanotubes and nanofibers . Cincinnati, OH: U.S. Department of Health and Human Services, Public Health Service, Centers for Disease Control and Prevention, National Institute for Occupational Safety and Health, DHHS (NIOSH). Publication No. 2013–145. NIOSH. ( 2013b) Current strategies for engineering controls in nanomaterial production and downstream handling processes . Cincinnati, OH: U.S. Department of Health and Human Services, Public Health Service, Centers for Disease Control and Prevention, National Institute for Occupational Safety and Health, DHHS (NIOSH). Publication No. 2014–102. NIOSH. ( 2017) External review draft—Current Intelligence Bulletin: the occupational exposure banding process: guidance for the evaluation of chemical hazards . Cincinnati, OH: U.S. Department of Health and Human Services, Public Health Service, Centers for Disease Control and Prevention, National Institute for Occupational Safety and Health. 8 March. Novoselov KS, Fal’ko VI, Colombo Let al.  ( 2012) A roadmap for graphene. Nature ; 490: 192– 200. Google Scholar CrossRef Search ADS PubMed  Nowack B, Krug HF, Height M. ( 2011) 120 years of nanosilver history: implications for policy makers. Environ Sci Technol ; 45: 1177– 83. Google Scholar CrossRef Search ADS PubMed  OSHA. ( 1983) OSHA safety and health standards. 29 CFR 1910.1000 . Washington, D.C.: U.S. Department of Labor, Occupational Safety and Health Administration (OSHA). 29 CFR 1910.1000. OSHA. ( 2012) Appendix A to §1910.1200—health hazard criteria. Fed Reg ; 77: 17574– 896. Paik SY, Zalk DM, Swuste P. ( 2008) Application of a pilot control banding tool for risk level assessment and control of nanoparticle exposures. Ann Occup Hyg ; 52: 419– 28. Google Scholar PubMed  Piccinno F, Gottschalk F, Seeger Set al.   ( 2012) Industrial production quantities and uses of ten engineered nanomaterials in Europe and the world. J Nanopart Res ; 14: 1– 11. Google Scholar CrossRef Search ADS PubMed  Sánchez Jiménez A, Varet J, Poland Cet al.  ( 2016) A comparison of control banding tools for nanomaterials. J Occup Environ Hyg ; 13: 936– 49. Google Scholar CrossRef Search ADS PubMed  Sargent EV, Kirk GD. ( 1988) Establishing airborne exposure control limits in the pharmaceutical industry. Am Ind Hyg Assoc J ; 49: 309– 13. Google Scholar CrossRef Search ADS PubMed  Schulte P, Geraci C, Hodson Let al.   ( 2013) Overview of risk management for engineered nanomaterials. J. Phys. Conf. Ser . 429. Schulte P, Geraci C, Zumwalde Ret al.  ( 2008) Occupational risk management of engineered nanoparticles. J Occup Environ Hyg ; 5: 239– 49. Google Scholar CrossRef Search ADS PubMed  Schulte PA, Salamanca-Buentello F. ( 2007) Ethical and scientific issues of nanotechnology in the workplace. Cien Saude Colet ; 12: 1319– 32. Google Scholar CrossRef Search ADS PubMed  Shvedova AA, Kisin E, Murray ARet al.  ( 2008) Inhalation vs. aspiration of single-walled carbon nanotubes in C57BL/6 mice: inflammation, fibrosis, oxidative stress, and mutagenesis. Am J Physiol Lung Cell Mol Physiol ; 295: L552– 65. Google Scholar CrossRef Search ADS PubMed  UNECE. ( 2011) Globally harmonized system of classification and labelling of chemicals (GHS) . 4th rev. edn. Geneva, Switzerland: United Nations Economic Commission for Europe. http://www.unece.org/trans/danger/publi/ghs/ghs_rev04/04files_e.html. Accessed 4 April 2016. PubMed PubMed  Wijnhoven SW, Peijnenburg WJ, Herberts CAet al.   ( 2009) Nano-silver-a review of available data and knowledge gaps in human and environmental risk assessment. Nanotoxicology ; 3: 109– 38. Google Scholar CrossRef Search ADS   Zalk DM, Nelson DI. ( 2008) History and evolution of control banding: a review. J Occup Environ Hyg ; 5: 330– 46. Google Scholar CrossRef Search ADS PubMed  Zalk DM, Paik SY, Swuste P. ( 2009) Evaluating the control banding nanotool: a qualitative risk assessment method for controlling nanoparticle exposures. J Nanopart Res ; 11: 1685– 704. Google Scholar CrossRef Search ADS   Published by Oxford University Press on behalf of The British Occupational Hygiene Society 2018. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Annals of Work Exposures and Health (formerly Annals Of Occupational Hygiene) Oxford University Press

Control Banding Tools for Engineered Nanoparticles: What the Practitioner Needs to Know

Loading next page...
 
/lp/ou_press/control-banding-tools-for-engineered-nanoparticles-what-the-cy4Zrh55zJ
Publisher
British Occupational Hygiene Society
Copyright
Published by Oxford University Press on behalf of The British Occupational Hygiene Society 2018.
ISSN
2398-7308
eISSN
2398-7316
D.O.I.
10.1093/annweh/wxy002
Publisher site
See Article on Publisher Site

Abstract

Abstract Control banding (CB) has been widely recommended for the selection of exposure controls for engineered nanomaterials (ENMs) in the absence of ENM-specific occupational exposure limits (OELs). Several ENM-specific CB strategies have been developed but have not been systematically evaluated. In this article, we identify the data inputs and compare the guidance provided by eight CB tools, evaluated on six ENMs, and assuming a constant handling/use scenario. The ENMs evaluated include nanoscale silica, titanium dioxide, silver, carbon nanotubes, graphene, and cellulose. Several of the tools recommended the highest level of exposure control for each of the ENMs in the evaluation, which was driven largely by the hazard banding. Dustiness was a factor in determining the exposure band in many tools, although most tools did not provide explicit guidance on how to classify the dustiness (high, medium, low), and published data are limited on this topic. The CB tools that recommended more diverse control options based on ENM hazard and dustiness data appear to be better equipped to utilize the available information, although further validation is needed by comparison to exposure measurements and OELs for a variety of ENMs. In all CB tools, local exhaust ventilation was recommended at a minimum to control exposures to ENMs in the workplace. Generally, the same or more stringent control levels were recommended by these tools compared with the OELs proposed for these ENMs, suggesting that these CB tools would generally provide prudent exposure control guidance, including when data are limited. control banding, dustiness, engineered nanomaterials, hazard banding, occupational exposure banding, occupational exposure limits Introduction The introduction of engineered nanomaterials (ENMs) into the workplace has created a challenge in assuring that their development, manufacture, production, and use can be performed safely. Given the limited information about the health risks associated with occupational exposure to these ENMs, individual companies, trade associations, and government agencies have instituted various risk management strategies to protect the health of workers (Schulte et al., 2013). In the absence of specific information, precautionary approaches to exposure control are recommended to ensure worker health protection (BSI, 2007; Schulte and Salamanca-Buentello, 2007; NIOSH, 2009a, 2012, 2013b). The traditional approach to protecting worker health is to measure worker exposures to potentially hazardous agents, compare them with occupational exposure limits (OELs), and then determine if existing control measures provide adequate protection (NIOSH, 2009b). Reliance on this approach has become increasingly difficult because of the growing number of potentially hazardous materials in the workplace that do not have OELs (Garrod and Rajan-Sithamparanadarajah, 2003). Control banding (CB) strategies have been proposed to make engineering control decisions for general chemical substances without OELs (NIOSH, 2009b). Many ENMs and ENM-enabled compounds also lack specific OELs and may have little or no toxicity information, and thus CB strategies have been proposed for evaluating and controlling exposures to ENMs in the workplace. These strategies are evaluated in this article. Although regulatory OELs for ENMs are not available to date, various groups have derived OELs for a number of ENMs based on nanotoxicology data and using various derivation methods (Mihalache et al., 2017). These OELs provide a basis for comparison of the hazard and CB results based on the ENM CB tools for a set of ENMs. Early efforts to address the control of exposures to potentially toxic or biologically active materials with little or no toxicity information available were simultaneously developed in the pharmaceutical (Sargent and Kirk, 1988; Naumann et al., 1996) and chemical (Brooke, 1998; Henry and Schaper, 1990) industries. Gardner and Oldershaw (1991) proposed the use of pragmatic exposure control concentrations (PECC) for volatile organic compounds without OELs in response to classification, packaging, and labeling directives in Europe; the proposed PECC were set at the mean OELs for similar substances with both OELs and risk phrases. CB strategies have also been used for many years to support hazard communications and labeling and to provide practical approaches to hazard evaluation and exposure control for use in small businesses, including the Control of Substances Hazardous to Health (COSHH) Essentials (HSE, 2009); Global Harmonization System (GHS) (UNECE, 2011); and Occupational Safety and Health Administration (OSHA) guidance (OSHA, 2012). Typically, CB strategies consist of two main components: (i) hazard bands (HBs), and (ii) exposure (or emission potential) bands. These qualitative bands provide rankings of substances based on their hazardous properties and their production/use, which range from low to high levels of hazard and/or exposure potential (EP). The combination of the hazard and exposure bands is used to derive the control band and associated engineering control options for a given occupational scenario. HBs are typically derived from toxicological data of adverse responses associated with acute or chronic exposures to hazardous substances in experimental animal studies, as well as data in humans when available. The five hazard categories, ranging from minimal to severe, are related to the health hazard rating system proposed by Henry and Schaper (1990). In addition to qualitative descriptors of the toxic effects, some HBs include quantitative exposure concentration ranges. Some of the earliest ‘target airborne concentration ranges’ were proposed by Brooke (1998) and are included in the COSHH Essentials CB tool. A general term for these exposure concentration ranges is occupational exposure bands (OEBs), which are typically order-of-magnitude, 8-h time-weighted average (TWA) concentrations (McKernan and Seaton, 2014). OEBs are related to the severity of the hazard such that the more severe the hazard, the lower the OEB (Fig. 1). Figure 1. View largeDownload slide CB for nanomaterials. Adapted from Naumann et al. (1996); Brooke (1998); Ader et al. (2005); Zalk and Nelson (2008); HSE (2009); ANSES (2010); UNECE (2011); OSHA (2012); Kuempel et al. (2012); ISO (2014). Abbreviation: TWA: Time-weighted average. Figure 1. View largeDownload slide CB for nanomaterials. Adapted from Naumann et al. (1996); Brooke (1998); Ader et al. (2005); Zalk and Nelson (2008); HSE (2009); ANSES (2010); UNECE (2011); OSHA (2012); Kuempel et al. (2012); ISO (2014). Abbreviation: TWA: Time-weighted average. Exposure bands or emission potential bands are qualitative descriptors of potential exposure levels given the factors that influence exposure such as dustiness (propensity of the material to become airborne), type of process or task being performed, and amount of material being handled (ISO, 2014). The CB recommendations on exposure control options often include the following four main areas: (i) good occupational hygiene practices, including general ventilation and intermittent use of personal protective equipment; (ii) engineering controls, including local exhaust ventilation; (iii) containment systems; and (iv) the need to seek guidance from a specialist. Other CB schemes include five control bands and associated performance-based exposure control limits, as shown in Fig. 1. CB strategies have also been suggested as a pragmatic approach to manage the potential health risk resulting from exposure to nanomaterials (Maynard, 2007; Schulte et al., 2008; Kuempel et al., 2012). Selection of appropriate control bands is uncertain in the absence of specific toxicology and exposure data for many nanomaterials. Several of the proposed ENM-specific CB tools attempt to address this concern by (i) taking a precautionary approach in assigning higher HBs, and consequently assigning higher risk or control bands, when information is limited or lacking; (ii) identifying high-concern substances based on particle properties (e.g. fibrous structure); and (iii) identifying the most severe health endpoints (e.g. carcinogenicity) to drive the selection of the control band. Some ENM-specific CB tools [e.g. French Agency for Food, Environmental, and Occupational Safety (ANSES) and International Organization for Standardization (ISO)] recommend adding one or more bands when using bulk material information to assign a HB for the nanomaterial (ANSES, 2010; ISO, 2014). Currently available CB tools that are specific to ENMs include the following eight tools: the CB Nanotool (Paik et al., 2008; Zalk et al., 2009); ANSES (ANSES, 2010); Stoffenmanager Nano (Duuren-Stuurman et al., 2011); Precautionary Matrix (Höck et al., 2013); ISO (ISO, 2014); EC Guidance (European Commission, 2014); NanoSafer (v. 1.1 beta) (Jensen et al., 2013); and the GoodNanoGuide (Kulinowski and Jaffe, 2009). These strategies have both similarities and differences in their features, including their scope and applicability, parameters used in the hazard/severity banding, and exposure/probability/emission potential banding, and in the classification of risk or control bands (Brouwer, 2012; Sánchez Jiménez et al., 2016). Each strategy targets different users and applicability domains (e.g. laboratory versus small business). The amount and detail of information and professional knowledge required for implementing each strategy also vary. A recent article by Liguori et al. (2016) provides a detailed review of six of these CB tools and updates the overview by Brouwer (2012). Draft guidance on developing OEBs for chemical hazards was issued by the National Institute for Occupational Safety and Health (NIOSH), which includes ENMs when sufficient toxicity data are available for either the ENM or its parent material (NIOSH, 2017). The NIOSH (2017) process does not provide CB recommendations, and it is not considered further here. All of the CB strategies currently available for ENMs are evaluated in this article using a set of six ENMs and defined working conditions, and cross-tool comparisons of the inputs and outcomes are provided. The objectives of this articles are to utilize the available CB tools for ENMs on a pilot set of ENMs to (i) identify the types and sources of information required, as illustrated by assessing a diverse set of ENMs, (ii) compare and evaluate the specific guidance provided by each tool, including its utility and limitations, and (iii) identify important data gaps that hinder the effective use of these tools and suggest areas of research to improve the evidence basis needed for hazard and CB of ENMs. Methods Description of selected engineered nanomaterials Six ENMs were evaluated in this article, including nanoscale silicon dioxide (SiO2), titanium dioxide (TiO2), silver (Ag), single-walled carbon nanotubes (SWCNT), graphene, and cellulose. These materials were selected because they are commonly used nanomaterials worldwide (Future Markets Inc., 2013) and because they represent a range of information available for nanomaterials in terms of hazard and dustiness (Table 1). SiO2 nanoparticles are used in a wide variety of markets, including medical, transportation, building materials, electronics, energy, and food industries. TiO2 nanoparticles have been used extensively in cosmetics, pigments, paints, and coatings (Piccinno et al., 2012). Silver nanoparticles have been used in various applications such as jewelry, photography, and antibacterial products and are increasingly being used in medical and consumer products including electronics and textile coating because of their physicochemical properties at the nanoscale (Wijnhoven et al., 2009; Nowack et al., 2011). Carbon nanotubes (CNTs) consist of nanoscale cylinders of carbon that can be produced with very large aspect ratios and are used in many industrial applications including electronics, polymer composites, and coatings and in biomedical applications including enhanced electron-scanning microscopy imaging and biosensors (NIOSH, 2013a). Graphene is made of pure carbon with atoms arranged in a regular hexagonal pattern and in a flat one-atom thick sheet; its commercial applications utilize its properties such as mechanical stiffness, strength and elasticity, and very high electrical and thermal conductivity (Novoselov et al., 2012). Nanocellulose is one of the newest commercially available ENMs, which has high strength and thermal stability and is gaining attention within ‘green chemistry’ as a renewable and biodegradable material (Isogai, 2013). Table 1. Characteristics of ENMs evaluated in this article in the various CB tools. Chemical composition and form of ENM  Name or description  Manufacturer (SDS revision date)  CAS number  Description  Primary particle dimensions (nm)  Specific surface area (m2/g)a  Dustiness (%) respirable fractionb  Silicon dioxide (SiO2), amorphous  Aerosil 380 F  Evonik, Essen Germany (15 October 2013)  112945-52-5 [SiO2]  Fumed, nanoscale powder  nrc  380  5.5  Titanium dioxide (TiO2)  Aeroxide P25  Evonik, Essen Germany (9 February 2014)  13463-67-7 [TiO2]  Fumed, nanoscale powder  20 (diameter)a  50  7.2  Silver nanoparticles  nr  Quantum Sphere, Inc., Santa Ana, CA (24 May 2007)  7440-22-4 [Ag]  Nanoscale powder  20–40 (diameter)  nr  0.4  CNT  SWCNT  Unidym Inc., Sunnyvale, CA (8 February 2011)  nr  Nanoscale powder  0.8–1.2 (diameter); 100–1000 (length)  508  31.8  Graphene  nr  Angstron Materials, Inc, Dayton, OH (8 May 2013)  1034343-98-0 (Graphene)  Nanoscale powder with <3 graphene layers  <10 (diameter); 1 (thickness); length nr  nr  nrd  Nanocellulose  Nanofibrillated fiber  Engineered Fiber Technologies, Shelton, CT (21 June 2007)  68442-85-3  Cellulose nanofibrils  100–500 (diameter)  nr  nrd  Chemical composition and form of ENM  Name or description  Manufacturer (SDS revision date)  CAS number  Description  Primary particle dimensions (nm)  Specific surface area (m2/g)a  Dustiness (%) respirable fractionb  Silicon dioxide (SiO2), amorphous  Aerosil 380 F  Evonik, Essen Germany (15 October 2013)  112945-52-5 [SiO2]  Fumed, nanoscale powder  nrc  380  5.5  Titanium dioxide (TiO2)  Aeroxide P25  Evonik, Essen Germany (9 February 2014)  13463-67-7 [TiO2]  Fumed, nanoscale powder  20 (diameter)a  50  7.2  Silver nanoparticles  nr  Quantum Sphere, Inc., Santa Ana, CA (24 May 2007)  7440-22-4 [Ag]  Nanoscale powder  20–40 (diameter)  nr  0.4  CNT  SWCNT  Unidym Inc., Sunnyvale, CA (8 February 2011)  nr  Nanoscale powder  0.8–1.2 (diameter); 100–1000 (length)  508  31.8  Graphene  nr  Angstron Materials, Inc, Dayton, OH (8 May 2013)  1034343-98-0 (Graphene)  Nanoscale powder with <3 graphene layers  <10 (diameter); 1 (thickness); length nr  nr  nrd  Nanocellulose  Nanofibrillated fiber  Engineered Fiber Technologies, Shelton, CT (21 June 2007)  68442-85-3  Cellulose nanofibrils  100–500 (diameter)  nr  nrd  nr: not reported. aAs reported in Evans et al. (2012). bDustiness measured at 50% relative humidity (Evans et al., 2012). cNot reported in SDS. dLack of published test data. View Large Overview of CB tools examined The various CB tools have been reviewed in recent publications (Eastlake et al., 2016; Liguori et al., 2016; Sánchez Jiménez et al., 2016). Several of the tools (ANSES, ISO, EC Guidance) follow a decision tree approach where the user answers questions about the nanomaterial, such as material form (solid/liquid/powder form), process (e.g. high/low energy process), and quantity to derive an EP, and then uses material characteristics (such as solubility, shape, biopersistence, and availability of toxicological data) to derive HBs. The second primary type of CB tool follows a score-based approach, which assesses overall hazard and EPs using explicit numerical criteria. The score-based approach gives a range of scores based on characteristics (similar to those in the decision tree approach) of the nanomaterial or parent material. CB Nanotool is the only tool to utilize a score-based approach for both hazard and EP (Paik et al., 2008; Zalk et al., 2009). EP and hazard severity are scored on a potential total of 100 points (higher values indicate higher hazard/EP). Any unknown properties or information should be assigned as ‘unknown’ and scored as 75% of the maximum value for each category. This score-based approach in CB Nanotool results in a default recommendation of containment control when key information is missing. Stoffenmanager Nano is a tiered approach in which the risk prioritization score allows for the implementation of controls followed by further evaluation of hazard and EP. The exposure banding process in Stoffenmanager Nano is a score-based approach that utilizes a range of user inputs including type of task, room ventilation, and whether engineering controls or protective equipment is used. In contrast, the hazard banding process in Stoffenmanager Nano opts for a decision tree approach, which relies on classification and labeling of products in accordance with the European classification of chemicals scheme (Duuren-Stuurman et al., 2011; Duuren-Stuurman et al., 2012). NanoSafer focuses on nanomaterials in powder form. This tool uses physical data (particle size, density, and surface area) and toxicological data from the safety data sheet (SDS) along with process data to determine a HB score (Jensen et al., 2013). NanoSafer places materials in one of four HBs: HB1 (0–0.25); HB2 (0.26–0.50); HB3 (0.51–0.75); HB4 (0.76–1.00). The EP is calculated for both short (15-min) and longer (8-h) exposures and for workers near the process (near field) and further from the work area (far field). This scoring takes into account dustiness, handling energy, amount handled, work duration and process cycles, volume of the room and air exchange rate. The EP is placed into five bands: EP1 (<0.11); EP2 (0.11–0.25); EP3 (0.26–0.50); EP4 (0.51–1.00); and EP5 (>1.00). The final risk level (RL1–RL5) is based on a combination of the HB and EP scores. The output for most of the CB tools discussed in this article is a control band, which recommends an appropriate exposure control approach in four or five levels (e.g. general ventilation, local exhaust ventilation, containment or seek specialist advice). The two exceptions are Stoffenmanager Nano and the Precautionary Matrix. Stoffenmanager Nano combines the hazard and control bands into a risk matrix, which results in a three-level prioritization scheme (high, medium, and low priority). This approach allows the user to implement appropriate controls and then assess exposure or utilize the tool to reevaluate the process and material based on risk. The Precautionary Matrix is unique in that it is designed to help businesses address the need for nanospecific action based on factors that consider both human and environmental risks. The final output of this tool provides a score indicating precautionary need with respect to employees handling materials and/or environmental issues. Any score above 20 indicates a need for caution. Description of CB tool inputs The primary parameters for the hazard and exposure banding process for each tool are summarized in Tables 2 and 3, respectively, along with the main input values for each of the tools in this evaluation. For comparison of the various CB strategies, the handling/use scenario was kept constant (e.g. hours worked, quantity of material used). The assumptions in this scenario include (i) ENMs were used in a small-scale production setting (i.e. research and development) that would include a small number of employees (one to five workers); (ii) employees performed tasks associated with handling a dry powder form of the ENM of interest approximately less than or equal to 4 h per day and 5 d per week; and (iii) the quantity used was approximately 50 g per day, which is based on reported levels in several carbonaceous production and downstream plants showing typical use quantities between 5 and 100 g in a standard weighing task (Dahm et al., 2012). Table 2. Hazard banding inputs for each material and tool evaluated. CB tool    Hazard input parameters  SiO2  TiO2  CNT  Nanosilver  Graphene  Nanocellulose  CB Nanotool  Parent material  Lowest occupation exposure limit (mg/m3)  6  2.4  3.5  0.01  2.5  5    Carcinogen?  Yes  Yes  No  No  Yesa  Unknownb    Dermal hazard?  Unknownb  No  No  No  Unknownb  Unknownb    Asthmagen?  Unknownb  Unknownb  Yes  No  Yesa  Unknownb  Nanoscale material  Surface reactivity  Unknownb  Unknownb  Unknownb  Unknownb  Unknownb  Unknownb    Particle shape  Unknownb  Unknownb  Tubular or fibrous  Unknownb  Anisotropic  Tubular or fibrous    Particle diameter (nm)  Unknownb  Unknownb  1–10  11–40  1–10  41–100    Solubility  Insoluble  Insoluble  Insoluble  Insoluble  Insoluble  Insoluble    Carcinogen?  No  Yes  Unknownb  Unknownb  No  No    Reproductive hazard?  No  Yes  Unknownb  Unknownb  Unknownb  Unknownb    Mutagen?  No  Unknownb  Unknownb  Unknownb  Unknownb  Unknownb    Dermal hazard?  No  No  Unknownb  Unknownb  Unknownb  Unknownb    Asthmagen?  Unknownb  Unknownb  Unknownb  Unknownb  Unknownb  Unknownb  GoodNanoGuide    Hazard group  C  B  B  C  C  C  Potential for material release  Free/unbound  Free/unbound  Free/unbound  Free/unbound  Free/unbound  Free/unbound  ANSES  Preliminary question  Does the product contain nanomaterials?  Yes  Yes  Yes  Yes  Yes  Yes    Is the nanosubstance already classified by a relevant authority?  No  No  No  No  No  No    Is it a biopersistent fiber?  No, not a fiber  No, not a fiber  No, not a fiber  No, not a fiber  No, not a fiber  Yes, it is an insoluble fiber    Is there a preliminary HB for the bulk material or most toxic analogous?  Yesc  Yesc  —  Yesc  Yesc  —  Bulk material  Bulk material: substance dissolution time >1 h  Yes, insoluble in waterd  Yes, insoluble in waterd  —  Yes, insoluble in waterd  Yesa,d  —    Bulk material: evidence of higher reactivity than bulk/ analogous material?  Yesa  Yesa  —  Yesa  Yesa  —  ISO    OEL dust (8-h TWA)  A  B  A  C  A  A  Acute toxicity  Be  Ba,e  Be  Be  —  Be  LD50 oral route (mg/kg)  —  A  —  —  —  Ad  LD50 dermal route (mg/kg)  —  Aa  —  —  —  Ad  LC50 inhalation 4H (mg/l) aerosols/particles  —  —  —  —  —  Ad  Severity of acute (life-threatening) effects  Ce  Be  Be  Ce  Be  Be  Sensitization  —  Aa  —  Ce  —  —  Mutagenicity/genotoxicity  —  Aa  —  —  —  —  Irritant/corrosiveness  Ee  A  Ae  Ae  Ae  Ee  Carcinogenicity  Ea,e  Ea,d,e  Ce  —  —  —  Developmental/reproductive toxicity  —  —  De  —  —  —  EC Guidance  Concern category  Characteristics of the manufactured nanomaterial  Medium–high concern  Medium–low concern  High concern  Medium–high concern  Medium–high concern  Medium–low concern  Dustiness band  Dustiness  Highf  Highf  Highf  Highf  Highf  Highf  Precautionary Matrix    Size of primary particles (free, bound, aggregated, or agglomerates) (nm)  1–500  1–500  1–500  1–500  1–500  1–500  Nanorelevance  Do the nanoparticles/rods form agglomerates >500 nm?  Nog  Nog  Nog  Nog  Nog  Nog  Potential effect  Redox activity or catalytic activity of nanoparticles/rods present in the nanomaterial  Not knownb  Not knownb  Not knownb  Not knownb  Not knownb  Not knownb    Stability (half-life) of the nanomaterial in the body  Not knownb  Not knownb  Not knownb  Not knownb  Not knownb  Not knownb  Stoffenmanager Nano    Product appearance  Powder  Powder  Powder  Powder  Powder  Powder    Dustiness (mg/kg)  High (>150–500)h  High (>150–500)h  Very high (>500)h  Medium (50–150)h  Unknownb  Unknownb    Moisture  Dry product (<5% moisture content)  Dry product (<5% moisture content)  Dry product (<5% moisture content)  Dry product (<5% moisture content)  Dry product (<5% moisture content)  Dry product (<5% moisture content)    Concentration of the nanocomponent in the product (%)  100  100  50–99  99.90  50–99  50–99    Does the product contain fibers/ fiber-like particles?  No  No  Yes  No  No  Yes    Length: diameter of the fiber (aspect ratio)  No  No  Yes  No  No  No    Hazardous properties  Unknownb  Carcinogenic (not mutagenic), reprotoxic and/ or very toxic  Toxic, corrosive, and/or respiratory allergens  Unknownb  Unknownb  Unknownb  NanoSafer    CAS Number  112945-52-5  13463-67-7  —  7440-22-4  1034343-98-0  68442-85-3      Is the material coated? (yes/no)  No  No  No  No  No  No      Morphology  Unknownb   No  Tube  No  Flake/plate/tabular/clay  Fiber      Solubility in water (g/L)  Insoluble (<1)  Insoluble (<1)  Insoluble (<1)  Insoluble (<1)  Insoluble (<1)  Insoluble (<1)      Shortest dimension (nm)  —  —  0.8  20  1  50      Middle dimension (nm)  —  —  —  —  —  —      Longest dimension (nm)  —  —  1000  40  1000  500      Average size (nm)  —  —  —  —  —  —      Density (g/cm3)  2.2  4.1  1.6  0.25  2.2  1.5      Surface area (powder material) (m2/g)  380i  15–50i,j  144 or 508i,j  5–25j  800j,k  284j,l      Respirable dustiness index (mg/ kg)  187.5  187.5  937.5  7.5  937.5  937.5  CB tool    Hazard input parameters  SiO2  TiO2  CNT  Nanosilver  Graphene  Nanocellulose  CB Nanotool  Parent material  Lowest occupation exposure limit (mg/m3)  6  2.4  3.5  0.01  2.5  5    Carcinogen?  Yes  Yes  No  No  Yesa  Unknownb    Dermal hazard?  Unknownb  No  No  No  Unknownb  Unknownb    Asthmagen?  Unknownb  Unknownb  Yes  No  Yesa  Unknownb  Nanoscale material  Surface reactivity  Unknownb  Unknownb  Unknownb  Unknownb  Unknownb  Unknownb    Particle shape  Unknownb  Unknownb  Tubular or fibrous  Unknownb  Anisotropic  Tubular or fibrous    Particle diameter (nm)  Unknownb  Unknownb  1–10  11–40  1–10  41–100    Solubility  Insoluble  Insoluble  Insoluble  Insoluble  Insoluble  Insoluble    Carcinogen?  No  Yes  Unknownb  Unknownb  No  No    Reproductive hazard?  No  Yes  Unknownb  Unknownb  Unknownb  Unknownb    Mutagen?  No  Unknownb  Unknownb  Unknownb  Unknownb  Unknownb    Dermal hazard?  No  No  Unknownb  Unknownb  Unknownb  Unknownb    Asthmagen?  Unknownb  Unknownb  Unknownb  Unknownb  Unknownb  Unknownb  GoodNanoGuide    Hazard group  C  B  B  C  C  C  Potential for material release  Free/unbound  Free/unbound  Free/unbound  Free/unbound  Free/unbound  Free/unbound  ANSES  Preliminary question  Does the product contain nanomaterials?  Yes  Yes  Yes  Yes  Yes  Yes    Is the nanosubstance already classified by a relevant authority?  No  No  No  No  No  No    Is it a biopersistent fiber?  No, not a fiber  No, not a fiber  No, not a fiber  No, not a fiber  No, not a fiber  Yes, it is an insoluble fiber    Is there a preliminary HB for the bulk material or most toxic analogous?  Yesc  Yesc  —  Yesc  Yesc  —  Bulk material  Bulk material: substance dissolution time >1 h  Yes, insoluble in waterd  Yes, insoluble in waterd  —  Yes, insoluble in waterd  Yesa,d  —    Bulk material: evidence of higher reactivity than bulk/ analogous material?  Yesa  Yesa  —  Yesa  Yesa  —  ISO    OEL dust (8-h TWA)  A  B  A  C  A  A  Acute toxicity  Be  Ba,e  Be  Be  —  Be  LD50 oral route (mg/kg)  —  A  —  —  —  Ad  LD50 dermal route (mg/kg)  —  Aa  —  —  —  Ad  LC50 inhalation 4H (mg/l) aerosols/particles  —  —  —  —  —  Ad  Severity of acute (life-threatening) effects  Ce  Be  Be  Ce  Be  Be  Sensitization  —  Aa  —  Ce  —  —  Mutagenicity/genotoxicity  —  Aa  —  —  —  —  Irritant/corrosiveness  Ee  A  Ae  Ae  Ae  Ee  Carcinogenicity  Ea,e  Ea,d,e  Ce  —  —  —  Developmental/reproductive toxicity  —  —  De  —  —  —  EC Guidance  Concern category  Characteristics of the manufactured nanomaterial  Medium–high concern  Medium–low concern  High concern  Medium–high concern  Medium–high concern  Medium–low concern  Dustiness band  Dustiness  Highf  Highf  Highf  Highf  Highf  Highf  Precautionary Matrix    Size of primary particles (free, bound, aggregated, or agglomerates) (nm)  1–500  1–500  1–500  1–500  1–500  1–500  Nanorelevance  Do the nanoparticles/rods form agglomerates >500 nm?  Nog  Nog  Nog  Nog  Nog  Nog  Potential effect  Redox activity or catalytic activity of nanoparticles/rods present in the nanomaterial  Not knownb  Not knownb  Not knownb  Not knownb  Not knownb  Not knownb    Stability (half-life) of the nanomaterial in the body  Not knownb  Not knownb  Not knownb  Not knownb  Not knownb  Not knownb  Stoffenmanager Nano    Product appearance  Powder  Powder  Powder  Powder  Powder  Powder    Dustiness (mg/kg)  High (>150–500)h  High (>150–500)h  Very high (>500)h  Medium (50–150)h  Unknownb  Unknownb    Moisture  Dry product (<5% moisture content)  Dry product (<5% moisture content)  Dry product (<5% moisture content)  Dry product (<5% moisture content)  Dry product (<5% moisture content)  Dry product (<5% moisture content)    Concentration of the nanocomponent in the product (%)  100  100  50–99  99.90  50–99  50–99    Does the product contain fibers/ fiber-like particles?  No  No  Yes  No  No  Yes    Length: diameter of the fiber (aspect ratio)  No  No  Yes  No  No  No    Hazardous properties  Unknownb  Carcinogenic (not mutagenic), reprotoxic and/ or very toxic  Toxic, corrosive, and/or respiratory allergens  Unknownb  Unknownb  Unknownb  NanoSafer    CAS Number  112945-52-5  13463-67-7  —  7440-22-4  1034343-98-0  68442-85-3      Is the material coated? (yes/no)  No  No  No  No  No  No      Morphology  Unknownb   No  Tube  No  Flake/plate/tabular/clay  Fiber      Solubility in water (g/L)  Insoluble (<1)  Insoluble (<1)  Insoluble (<1)  Insoluble (<1)  Insoluble (<1)  Insoluble (<1)      Shortest dimension (nm)  —  —  0.8  20  1  50      Middle dimension (nm)  —  —  —  —  —  —      Longest dimension (nm)  —  —  1000  40  1000  500      Average size (nm)  —  —  —  —  —  —      Density (g/cm3)  2.2  4.1  1.6  0.25  2.2  1.5      Surface area (powder material) (m2/g)  380i  15–50i,j  144 or 508i,j  5–25j  800j,k  284j,l      Respirable dustiness index (mg/ kg)  187.5  187.5  937.5  7.5  937.5  937.5  LD50: Dose associated with 50% lethality; LC50: Concentration associated with 50% lethality. Dash (—) indicates that no information was found for this specific parameter. aTOXNET. bInterpreted as unknown as no proper option available. cCreated using Table 1 of ANSES. dGESTIS. eECHA C&L. fMethod provides no option for ‘unknown’ therefore ‘high’. gNot known if deagglomeration of agglomerates (or aggregates) to primary nanoparticles/rods or agglomerates <500 nm occurs in the body. hStoffenmanager Nano dustiness levels are medium, high and very high, and unknown. iReported in Evans et al. (2012). jMultiple numbers provided and larger number used. kTechnical data sheet/SDS for material. lSehaui, Zhou, Berglund (2011). View Large Table 3. Exposure band parameters for each tool and levels of each category. CB tool  Information/scenario    Materials    SiO2  TiO2  CNT  Nanosilver  Graphene  Nanocellulose  All tools  Substance emission potential/physical form  Dry powder  Activity emission potential/ amount handled per day  50 g  Task duration  1–4 h/d  Task frequency  5 d/wk  Volume of the working room  108 m3 per NanoSafer  CB Nanotool  Dustiness  5.5%a medium  7.2%a medium  31.8%a high  0.4%a low  Unknown  Unknown  Number of employees with similar exposure  ≤5 employees  Good Nano Guide  Exposure duration  Medium  ANSES  Emission potential (high/ moderate +1 band)  EP3, powder  EP3, powder  EP3, powder  EP3, powder  EP3, powder  EP3, powder  Manufacturing/handling process  Handling powder  ISO  Exposure band for dust generation/dustiness  EB2  EB2  EB2  EB2  EB2  EB2  Manufacturing/handling process  Material in powder form—manufacturing use and handling—amount used >0.1 g—Low potential of dust  EC Guidance  Level of Exposure  High  Stoffenmanager Nano  Task characterization  Handling of products in small amounts (up to 100 g) or in situations where only low quantities of products are likely to be released    Is the task carried out at the breathing zone of the employee (distance person product <1 m)?  Yes    Is there more than one employee carrying out the same task simultaneously?  Yes    Is the working room being cleaned daily?  Yes    Are inspections and maintenance of machines/ancillary equipment being done at least monthly to ensure good condition and proper functioning and performance?  Yes    Ventilation of the working room  Mechanical and/or natural    Local control measures at the source  No control measures    Is the employee situation in a cabin?  No    Is personal protective equipment applied?  No  NanoSafer  Energy level  H3 (0.50): moderate energy (e.g. pour 5–30 cm drop height, blending of powder in liquid medium)  Air exchanges  8/h  Mass handled per cycle  0.025 kg  Length x width x height of workroom (m)  6 x 6 x 3  Cycle duration  60 min  Time to perform work cycle  15 min  Amount of product used per work cycle  0.1 kg  How many times is the cycle repeated daily?  4  Activity level of room  Low quiet  CB tool  Information/scenario    Materials    SiO2  TiO2  CNT  Nanosilver  Graphene  Nanocellulose  All tools  Substance emission potential/physical form  Dry powder  Activity emission potential/ amount handled per day  50 g  Task duration  1–4 h/d  Task frequency  5 d/wk  Volume of the working room  108 m3 per NanoSafer  CB Nanotool  Dustiness  5.5%a medium  7.2%a medium  31.8%a high  0.4%a low  Unknown  Unknown  Number of employees with similar exposure  ≤5 employees  Good Nano Guide  Exposure duration  Medium  ANSES  Emission potential (high/ moderate +1 band)  EP3, powder  EP3, powder  EP3, powder  EP3, powder  EP3, powder  EP3, powder  Manufacturing/handling process  Handling powder  ISO  Exposure band for dust generation/dustiness  EB2  EB2  EB2  EB2  EB2  EB2  Manufacturing/handling process  Material in powder form—manufacturing use and handling—amount used >0.1 g—Low potential of dust  EC Guidance  Level of Exposure  High  Stoffenmanager Nano  Task characterization  Handling of products in small amounts (up to 100 g) or in situations where only low quantities of products are likely to be released    Is the task carried out at the breathing zone of the employee (distance person product <1 m)?  Yes    Is there more than one employee carrying out the same task simultaneously?  Yes    Is the working room being cleaned daily?  Yes    Are inspections and maintenance of machines/ancillary equipment being done at least monthly to ensure good condition and proper functioning and performance?  Yes    Ventilation of the working room  Mechanical and/or natural    Local control measures at the source  No control measures    Is the employee situation in a cabin?  No    Is personal protective equipment applied?  No  NanoSafer  Energy level  H3 (0.50): moderate energy (e.g. pour 5–30 cm drop height, blending of powder in liquid medium)  Air exchanges  8/h  Mass handled per cycle  0.025 kg  Length x width x height of workroom (m)  6 x 6 x 3  Cycle duration  60 min  Time to perform work cycle  15 min  Amount of product used per work cycle  0.1 kg  How many times is the cycle repeated daily?  4  Activity level of room  Low quiet  aReported in Evans et al. (2012); categories assigned here are based on judgment: 0.1–1%, low; 1–10%, medium; >10%, high. View Large It should be noted that the rates of production from TiO2 and silver may be much higher—in the range of 1–5 kg per day based on published data (Lee et al., 2011). However, the upper range of material quantity for scoring of EP of any of the CB tools evaluated herein is 1 kg, with most tools giving quantities of greater than 1 g the highest score in this category. The physical properties of the ENMs utilized in this evaluation were obtained from the manufacturer’s technical data sheets and/or SDSs. The dustiness of the materials was classified in this article (based on judgment) as low, medium, or high according to the following respirable fraction: 0.1–1% low, 1–10% medium, >10% high. This information was used in the tools requiring dustiness category inputs (Tables 2 and 3). The data on the ENM dustiness were taken from the results of dustiness characterization reported in Evans et al. (2012), because no other large-scale dustiness test dataset for fine and nanomaterials was available. These data were collected using a Venturi test procedure, which may not be applicable to all models. Specifically, NanoSafer and ANSES recommend the use of methods from the EN 15051 standard for dustiness testing, which employ less aggressive methods of dispersion (European Committee for Standardization (CEN), 2013). Thus, the values used in this evaluation may overestimate the relative dustiness of the materials and result in higher EP scores. For the hazard banding of these six ENMs, data were collected from a variety of sources including governmental sources, professional organizations, online databases, and published guidance/literature (Table 4). SDSs were consulted to obtain information specific to the properties of each ENM: the physical, health, and environmental health hazards; protective measures; and safety precautions for handling, storing, and transporting the material. If an SDS specific to that ENM was available, then that information was obtained and used. OELs for the bulk (non-nanometer sized) material most similar to each ENM in this study were used in the banding. The lowest authoritative OELs were used, which were not necessarily regulatory OELs. Information from NIOSH and other authoritative guidance documents was used to address questions regarding toxicity and health hazards associated with each substance. As most ENMs do not have guidance documents with extensive literature and data reviews, these data may be obtained from online databases. For this study, we used a German substance database (GESTIS), United States National Library of Medicine Toxicology Data Network (TOXNET), and the European Chemicals Agency (ECHA) Classification and Labeling Inventory. In general, surface reactivity for a given mass-based exposure to each ENM was assumed to be high because of both the unknown potential for functionalization and the higher surface area of most ENMs versus the parent (or bulk) material. Solubility was determined based on information provided in the SDS or database literature search. If more than one type of solubility (soluble and insoluble) was listed, then the ENM was considered insoluble. If the parent material was indicated to be carcinogenic, a dermal hazard, or an asthmagen, then the ENM was also assumed to have similar health effects. Otherwise, when information was not available, all ENM data were indicated or interpreted as unknown. Table 4. Sources of information for CB tools and model inputs. Source information  Source  Content description  Website address  OEL guidance  US OSHA  Permissible exposure limits  https://www.osha.gov/  US NIOSH  Recommended exposure limits  http://www.cdc.gov/niosh/  American Conference of Governmental Industrial Hygienists (ACGIH)  Threshold limit values  http://www.acgih.org/  Online databases  Institute for Occupational Safety and Health of the German Social Accident Insurance  Substance database (GESTIS)  http://gestis-en.itrust.de/nxt/ gateway.dll/gestis_en/000000. xml?f=templates$fn=default.htm$3.0  US National Library of Medicine  Toxicology data network (TOXNET)  http://toxnet.nlm.nih.gov/  European Chemicals Agency  Classification & Labeling Inventory  http://echa.europa.eu/ information-on-chemicals  CB methods  Lawrence Livermore National Laboratory  CB Nanotool  http://controlbanding.net/    US NIOSH Oregon Nanoscience and Microtechnologies Institute (OMAMI) Oregon State University (OSU)  GoodNanoGuide  https://nanohub.org/groups/gng    French Agency for Food, Environmental and Occupational Health & Safety (ANSES)  Development of a specific CB tool for nanomaterials  https://www.anses.fr/sites/default/files/ documents/AP2008sa0407RaEN.pdf    ISO  Nanotechnologies— Occupational risk management applied to engineered nanomaterials—Part 2: Use of the CB approach TS 12901–2:2014  http://www.iso.org/iso/catalogue_detail. htm?csnumber=53375    National Research Centre for the Working Environment, Copenhagen, Denmark  NanoSafer  http://www.nanosafer.org/    Schweizerische Eidgenossenschaft— Federal office of Public Health  Precautionary Matrix  http://www.bag.admin.ch/nanotechnologie/ 12171/12174/12175/index. html?webgrab_path=aHR0cDovL3d3dy5iYWctYW53LmFkbWluLmNoL25hbm9yYXN0ZXIvcG9ydGFsX2VuLnBocD9tb2Q9YSZsYW5nPWVu&lang=en    Dutch Ministry of Social Affairs and Employment (SAE), TNO, Arbo Unie, BECO(EY)  Stoffenmanager Nano  https://nano.stoffenmanager.nl/    European Agency for Safety and Health at Work (EU-OSHA)  Guidance on the protection of the health and safety of workers from the potential risks related to nanomaterials at work  https://osha.europa.eu/en/news/ eu-safe-use-of-nanomaterials-commission- publishes-guidance-for-employers- and-workers  Guidance/ literature  US NIOSH  Occupational exposure to CNTs and nanofibers  http://www.cdc.gov/niosh/docs/2013–145/pdfs/ 2013–145.pdf  US NIOSH  Occupational exposure to titanium dioxide  http://www.cdc.gov/niosh/docs/2011–160/pdfs/ 2011–160.pdf  Varies depending on material  Material SDS  Specific to the material used  Source information  Source  Content description  Website address  OEL guidance  US OSHA  Permissible exposure limits  https://www.osha.gov/  US NIOSH  Recommended exposure limits  http://www.cdc.gov/niosh/  American Conference of Governmental Industrial Hygienists (ACGIH)  Threshold limit values  http://www.acgih.org/  Online databases  Institute for Occupational Safety and Health of the German Social Accident Insurance  Substance database (GESTIS)  http://gestis-en.itrust.de/nxt/ gateway.dll/gestis_en/000000. xml?f=templates$fn=default.htm$3.0  US National Library of Medicine  Toxicology data network (TOXNET)  http://toxnet.nlm.nih.gov/  European Chemicals Agency  Classification & Labeling Inventory  http://echa.europa.eu/ information-on-chemicals  CB methods  Lawrence Livermore National Laboratory  CB Nanotool  http://controlbanding.net/    US NIOSH Oregon Nanoscience and Microtechnologies Institute (OMAMI) Oregon State University (OSU)  GoodNanoGuide  https://nanohub.org/groups/gng    French Agency for Food, Environmental and Occupational Health & Safety (ANSES)  Development of a specific CB tool for nanomaterials  https://www.anses.fr/sites/default/files/ documents/AP2008sa0407RaEN.pdf    ISO  Nanotechnologies— Occupational risk management applied to engineered nanomaterials—Part 2: Use of the CB approach TS 12901–2:2014  http://www.iso.org/iso/catalogue_detail. htm?csnumber=53375    National Research Centre for the Working Environment, Copenhagen, Denmark  NanoSafer  http://www.nanosafer.org/    Schweizerische Eidgenossenschaft— Federal office of Public Health  Precautionary Matrix  http://www.bag.admin.ch/nanotechnologie/ 12171/12174/12175/index. html?webgrab_path=aHR0cDovL3d3dy5iYWctYW53LmFkbWluLmNoL25hbm9yYXN0ZXIvcG9ydGFsX2VuLnBocD9tb2Q9YSZsYW5nPWVu&lang=en    Dutch Ministry of Social Affairs and Employment (SAE), TNO, Arbo Unie, BECO(EY)  Stoffenmanager Nano  https://nano.stoffenmanager.nl/    European Agency for Safety and Health at Work (EU-OSHA)  Guidance on the protection of the health and safety of workers from the potential risks related to nanomaterials at work  https://osha.europa.eu/en/news/ eu-safe-use-of-nanomaterials-commission- publishes-guidance-for-employers- and-workers  Guidance/ literature  US NIOSH  Occupational exposure to CNTs and nanofibers  http://www.cdc.gov/niosh/docs/2013–145/pdfs/ 2013–145.pdf  US NIOSH  Occupational exposure to titanium dioxide  http://www.cdc.gov/niosh/docs/2011–160/pdfs/ 2011–160.pdf  Varies depending on material  Material SDS  Specific to the material used  View Large Hazard data on the adverse effects from repeated exposure to these nanomaterials in animals were also evaluated given the relevance to potential worker exposures for up to a working lifetime. Rat is the rodent species used in the criteria for specific target organ toxicity–repeated exposure (STOT-RE) in many of the hazard banding schemes. Therefore, subchronic inhalation studies in rats were identified from literature searches in Pubmed, using the search terms ‘nanomaterial name’ and ‘rat’ and ‘inhalation’. The adverse effect levels from the identified rat studies are compared with the effect levels in the ANSES and GHS hazard banding schemes for STOT-RE. OELs that have been proposed for nanomaterials (Table S1, available at Annals of Occupational Hygiene online) are used in comparisons with the CB results in this evaluation. OELs are typically based on a more in-depth analysis of the data, although different data, methods, and assumptions may have been used in deriving those OELs. The steps involved in selecting and using the evaluated CB tools are shown in Fig. 2. This figure references the process and data sources, which are used in conducting the analyses described in this article. Figure 2. View largeDownload slide Steps for selecting and using the CB tools. Figure 2. View largeDownload slide Steps for selecting and using the CB tools. Results A summary of results of the recommended risk/control bands for each CB strategy and for all six ENMs evaluated is shown in Table 5. The results of both the exposure and HBs are presented, when applicable. This table shows that the output for each tool is unique. For example, the Precautionary Matrix is different than the other tools discussed, in that the process does not result in the determination of a control band. Rather the Precautionary Matrix specifies whether precautionary, nanospecific safety measures are needed or not based on a calculated score. The ANSES and ISO tools are very similar in nature and include five CB levels: 1, general ventilation; 2, local exhaust ventilation (exterior hood, table hood); 3, enclosed ventilation (fume hood, ventilated booth); 4, full containment; or 5, full containment and review by specialist. NanoSafer also has five RLs, which correspond to control recommendations including the following: RL1, local exhaust ventilation (LEV)/fume hood; RL2, LEV/fume hood potentially with respirator; RL3, LEV/fume hood with respirator; RL4, fume hood/enclosure/glovebox with respirator; RL5, fume hood/enclosure/glovebox with Supplied Air Respirator. CB Nanotool, GoodNanoGuide, and the EC Guidance have four control bands: 1, general ventilation; 2, local exhaust ventilation/engineering controls; 3, containment; or 4, seek specialist advice. In contrast, Stoffenmanager Nano assigns one of three risk priority bands (1, high priority; 2, medium priority; or 3, low priority). Table 5. Summary of CB tool results.     Nanomaterial      SiO2  TiO2  CNT  Nanosilver  Graphene  Nanocellulose  CB Nanotool  Severity score  45.5  59  60  56  60  53  Exposure probability score  65  65  80  57.5  72.5  72.5  Control band  RL2, fume hoods or local exhaust ventilation  RL3, containment  RL4, seek specialist advice  RL3, containment  RL3, containment  RL3, containment  Good Nano Guide    Hazard group C, limited data CB, 3 seek specialist advice  Hazard group B, NIOSH CIB TiO2 CB, 3 containment  Hazard group B, NIOSH CIB CNT/F CB, 3 containment  Hazard group C, limited data CB, 4 seek specialist advice  Hazard group C, limited data CB, 4 seek specialist advice  Hazard group C, limited data CB, 4 seek specialist advice  ANSES    HB5 + emission potential 4 = control level 5: full containment and review by a specialist required: seek expert advice  HB5 + emission potential 4 = control level 5: full containment and review by a specialist required: seek expert advice  HB5 + emission potential 4 = control level 5: full containment and review by a specialist required: seek expert advice  HB5 + emission potential 3 = control level 5: full containment and review by a specialist required: seek expert advice  HB5 + emission potential 4 = control level 5: full containment and review by a specialist required: seek expert advice  HB5 + emission potential 4 = control level 5: full containment and review by a specialist required: seek expert advice  ISO    HB E, severe hazard + exposure band 2 = control band 5 Full containment and review by a specialist: seek expert advice  HB E, severe hazard + exposure band 2 = control band 5 Full containment and review by a specialist: seek expert advice  HB E, severe hazard + exposure band 2 = control band 5 Full containment and review by a specialist: seek expert advice  HB E, severe hazard + exposure band 2 = control band 5 Full containment and review by a specialist: seek expert advice  HB E, severe hazard + exposure band 2 = control band 5 Full containment and review by a specialist: seek expert advice  HB E, severe hazard + exposure band 2 = control band 5 Full containment and review by a specialist: seek expert advice  EC Guidance    Medium–high concern + high level of exposure = RL4 Risk assessment performed by an expert + it is essential that measures specifically designed for the processes in question to be adopted  Medium–low concern + high level of exposure = RL3 Risk assessment performed by an expert + closed systems or containment must be used  High concern category + high level of exposure = RL4 Risk assessment performed by an expert + it is essential that measures specifically designed for the processes in question to be adopted  Medium–high concern + high level of exposure = RL4 Risk assessment performed by an expert + it is essential that measures specifically designed for the processes in question to be adopted  Medium–high concern + high level of exposure = RL4 Risk assessment performed by an expert + it is essential that measures specifically designed for the processes in question to be adopted  Medium–low concern + high level of exposure = RL3 Risk assessment performed by an expert + closed systems or containment must be used  Precautionary Matrix    All numbers > 20 Precautionary need warranted  All numbers > 20 Precautionary need warranted  All numbers > 20 Precautionary need warranted  All numbers > 20 Precautionary need warranted  All numbers > 20 Precautionary need warranted  All numbers > 20 Precautionary need warranted  Stoffenmanager Nano  Task weighted  E, extreme hazard class 3, high EP I, high-risk priority  D, very high hazard class 3, high EP I, high-risk priority  E, extreme hazard class 3, high EP I, high-risk priority  D, very high hazard class 2, average EP II, medium-risk priority  E, extreme hazard class 3, high EP I, high-risk priority  E, extreme hazard class 3, high EP I, high-risk priority    Time and frequency weighted  E, extreme hazard class 2, average EP I, high-risk priority  D, very high hazard class 3, high EP II, medium-risk priority  E, extreme hazard class 2, average EP I, high-risk priority  D, very high hazard class 2, average EP II, medium-risk priority  E, extreme hazard class 2, average EP I, high-risk priority  E, extreme hazard class 2, average EP I, high-risk priority  NanoSafer  Estimated hazard level  0.59  1  0.59  0.761  0.539  0.488    Near field  Acute: 0.1267 RL2: low toxicity/low EP EB2: low EP  Acute: 0.0777 RL4: high toxicity/high EP EB1: very low EP  Acute: 1.056 RL5: very high toxicity/moderate- to-very high EP EB5: very high EP  Acute: 0.0227 RL4: high toxicity/high EP EB1: very low EP  Acute: 3.2 RL5: very high toxicity/moderate- to-very high EP EB5: very high EP  Acute: 0.3873 RL4: high toxicity/high EP EB3: moderate EP    Daily: 0.1275 RL2: low toxicity/low EP EB2: low EP  Daily: 0.0782 RL4: high toxicity/high EP EB1: very low EP  Daily: 1.063 RL5: very high toxicity/moderate- to-very high EP EB5: very high EP  Daily: 0.0229 RL4: high toxicity/high EP EB1: very low EP  Daily: 3.22 RL5: very high toxicity/moderate- to-very high EP EB5: very high EP  Daily: 0.3899 RL4: high toxicity/high EP EB3: moderate EP      Nanomaterial      SiO2  TiO2  CNT  Nanosilver  Graphene  Nanocellulose  CB Nanotool  Severity score  45.5  59  60  56  60  53  Exposure probability score  65  65  80  57.5  72.5  72.5  Control band  RL2, fume hoods or local exhaust ventilation  RL3, containment  RL4, seek specialist advice  RL3, containment  RL3, containment  RL3, containment  Good Nano Guide    Hazard group C, limited data CB, 3 seek specialist advice  Hazard group B, NIOSH CIB TiO2 CB, 3 containment  Hazard group B, NIOSH CIB CNT/F CB, 3 containment  Hazard group C, limited data CB, 4 seek specialist advice  Hazard group C, limited data CB, 4 seek specialist advice  Hazard group C, limited data CB, 4 seek specialist advice  ANSES    HB5 + emission potential 4 = control level 5: full containment and review by a specialist required: seek expert advice  HB5 + emission potential 4 = control level 5: full containment and review by a specialist required: seek expert advice  HB5 + emission potential 4 = control level 5: full containment and review by a specialist required: seek expert advice  HB5 + emission potential 3 = control level 5: full containment and review by a specialist required: seek expert advice  HB5 + emission potential 4 = control level 5: full containment and review by a specialist required: seek expert advice  HB5 + emission potential 4 = control level 5: full containment and review by a specialist required: seek expert advice  ISO    HB E, severe hazard + exposure band 2 = control band 5 Full containment and review by a specialist: seek expert advice  HB E, severe hazard + exposure band 2 = control band 5 Full containment and review by a specialist: seek expert advice  HB E, severe hazard + exposure band 2 = control band 5 Full containment and review by a specialist: seek expert advice  HB E, severe hazard + exposure band 2 = control band 5 Full containment and review by a specialist: seek expert advice  HB E, severe hazard + exposure band 2 = control band 5 Full containment and review by a specialist: seek expert advice  HB E, severe hazard + exposure band 2 = control band 5 Full containment and review by a specialist: seek expert advice  EC Guidance    Medium–high concern + high level of exposure = RL4 Risk assessment performed by an expert + it is essential that measures specifically designed for the processes in question to be adopted  Medium–low concern + high level of exposure = RL3 Risk assessment performed by an expert + closed systems or containment must be used  High concern category + high level of exposure = RL4 Risk assessment performed by an expert + it is essential that measures specifically designed for the processes in question to be adopted  Medium–high concern + high level of exposure = RL4 Risk assessment performed by an expert + it is essential that measures specifically designed for the processes in question to be adopted  Medium–high concern + high level of exposure = RL4 Risk assessment performed by an expert + it is essential that measures specifically designed for the processes in question to be adopted  Medium–low concern + high level of exposure = RL3 Risk assessment performed by an expert + closed systems or containment must be used  Precautionary Matrix    All numbers > 20 Precautionary need warranted  All numbers > 20 Precautionary need warranted  All numbers > 20 Precautionary need warranted  All numbers > 20 Precautionary need warranted  All numbers > 20 Precautionary need warranted  All numbers > 20 Precautionary need warranted  Stoffenmanager Nano  Task weighted  E, extreme hazard class 3, high EP I, high-risk priority  D, very high hazard class 3, high EP I, high-risk priority  E, extreme hazard class 3, high EP I, high-risk priority  D, very high hazard class 2, average EP II, medium-risk priority  E, extreme hazard class 3, high EP I, high-risk priority  E, extreme hazard class 3, high EP I, high-risk priority    Time and frequency weighted  E, extreme hazard class 2, average EP I, high-risk priority  D, very high hazard class 3, high EP II, medium-risk priority  E, extreme hazard class 2, average EP I, high-risk priority  D, very high hazard class 2, average EP II, medium-risk priority  E, extreme hazard class 2, average EP I, high-risk priority  E, extreme hazard class 2, average EP I, high-risk priority  NanoSafer  Estimated hazard level  0.59  1  0.59  0.761  0.539  0.488    Near field  Acute: 0.1267 RL2: low toxicity/low EP EB2: low EP  Acute: 0.0777 RL4: high toxicity/high EP EB1: very low EP  Acute: 1.056 RL5: very high toxicity/moderate- to-very high EP EB5: very high EP  Acute: 0.0227 RL4: high toxicity/high EP EB1: very low EP  Acute: 3.2 RL5: very high toxicity/moderate- to-very high EP EB5: very high EP  Acute: 0.3873 RL4: high toxicity/high EP EB3: moderate EP    Daily: 0.1275 RL2: low toxicity/low EP EB2: low EP  Daily: 0.0782 RL4: high toxicity/high EP EB1: very low EP  Daily: 1.063 RL5: very high toxicity/moderate- to-very high EP EB5: very high EP  Daily: 0.0229 RL4: high toxicity/high EP EB1: very low EP  Daily: 3.22 RL5: very high toxicity/moderate- to-very high EP EB5: very high EP  Daily: 0.3899 RL4: high toxicity/high EP EB3: moderate EP  View Large Reviewing the results of the evaluation shown in Table 5 illustrates the differences between both the ENMs and the CB strategies evaluated by keeping the handling/use scenarios constant, as discussed in Methods. For the CB Nanotool, the RL ranged from RL4—seek specialist advice for CNTs to RL3 for titanium dioxide, nanoscale silver, graphene, and nanocellulose to RL2 for silicon dioxide. For these ENMs, the lowest hazard (severity) score was for SiO2, while the highest was for CNTs. The primary difference that resulted in the differing RL bands was the severity (HB) score, which placed all ENMs except for SiO2 in the high-severity category. Stoffenmanager Nano indicated that SiO2, CNTs, graphene, and nanocellulose are overall a high-risk priority. TiO2 and nanosilver were both considered a high-risk priority when task-weighted but considered to be a medium-risk priority when the time and frequency of handling were taken into account indicating a lower overall risk. For the Precautionary Matrix, evaluation indicated that a risk is present for both workers and the environment based on a final calculated score of over 20 for all ENMs evaluated. For both the ANSES tool and ISO guidance, all ENMs fell into the same hazard and exposure bands resulting in similar control band—CB5—full containment and requiring expert advice. For the EC Guidance, nanosilver, CNTs, silica, and graphene fell into the highest RL resulting in the recommendation to adopt process-based control measures. TiO2 and nanocellulose were at the next lowest level, which recommended the use of closed systems or containment of the process. For the GoodNanoGuide, SiO2, graphene, nanocellulose, and nanoscale silver were put into the highest hazard grouping because of lack of information on the health effects associated with these ENMs. TiO2 and CNTs were placed into a lower hazard group because of the availability of hazard data (NIOSH, 2011). Finally, for NanoSafer, CNTs and graphene fell into the highest RL resulting in the recommendation for a fume hood/enclosure/glovebox with supplied air respirator. Nanocellulose, silica (amorphous), nanosilver, and TiO2 were at the next lowest level, which recommended the use of highly efficient local exhaust ventilation, fume hood, or glovebox along with a respirator. The evaluation of the repeated exposure data in rodents for those ENMs with these data showed that the lowest observed adverse effect levels were all <20 mg/m3, which is the level of concern for chronic adverse effects (STOT-RE). Based on these results, the HB would be either ‘category D—serious hazard’ according to the ANSES and ISO or ‘category 1—health hazard—danger’ based on the GHS and US OSHA hazard banding strategies for nanoscale amorphous silica, TiO2, silver, and multiwalled CNTs. These HBs are similar to or lower than the equivalent concentrations at the OELs (Table S1, available at Annals of Occupational Hygiene online). Discussion Relatively limited evaluation and validation have been performed on the available CB tools for ENMs. This study adds to the current scientific literature by providing a systematic evaluation and application of all eight of the currently available CB tools for ENMs, using six different types of ENMs of varying dustiness level, for a fixed exposure and use scenario in the workplace. Outcomes are examined across the CB tools and compared with the proposed OELs for these ENMs. Data gaps in the key inputs to these CB tools are identified. Finally, the drivers for these outcomes are identified, and research needs are suggested to improve the information available and the utility of these CB tools for making workplace exposure control decisions. Recent articles by Eastlake et al., Ligouri et al., and Sanchez Jiménez et al. are complementary with this article but also differ in both approach and scope (Eastlake et al., 2016; Liguori et al., 2016; Sánchez Jiménez et al., 2016). Eastlake et al. (2016) provide a systematic review of the ENM-specific CB tools and conclude that few of these tools have been validated with regard to their effectiveness in controlling exposure. Ligouri et al. provide an update of the earlier review by Brouwer (Brouwer, 2012), including a more in-depth description of those tools. Ligouri et al. review six of the eight CB tools examined in this article (which also includes GoodNanoGuide and ISO/TS 12901–2). Ligouri et al. also describes the different inputs and possible outputs of the CB tools, but they did not conduct any actual evaluations on ENMs as performed in this article on a set of six ENMs. These evaluations show that differences in the particle properties can influence the outcomes of the different CB tools, depending on how a particular property is treated in the various hazard and exposure banding approaches. The Sanchez Jiménez et al. article provides a broad evaluation of four of these CB tools, including sensitivity analyses of the tool inputs and limited exposure validation testing using airborne number concentration data on one ENM (cloisite) and three processes. The Sanchez Jiménez et al. article focused on assessment of the tools geared more to researchers, whereas this article provides step-by-step information and examples that may be useful to the practitioner in selecting CB tools, gathering the input information, and assessing the usefulness of the results. Evaluation of CB tool outcomes The findings of this current evaluation show that the ANSES and ISO tools recommended the highest level of exposure control for the majority of ENMs in this use scenario (Table 6). CB Nanotool, EC Guidance, NanoSafer, and GoodNanoGuide recommended lower levels of control by ENM. The CB resulted in either the same or higher levels of exposure control to those suggested by the proposed OELs for nanoscale TiO2 and CNTs (Table 6). CNTs were generally in the most protective band ‘seek expert advice’ with a controls performance level of <1 µg/m3. In contrast, the recommended control bands for silica and graphene differ widely between CB Nanotool and EC Guidance, i.e. either level 2 or level 4, respectively, in this evaluation. The proposed OELs for CNTs also vary over two or more orders of magnitude and control bands. However, these OELs for ENMs are all lower on a mass basis than their bulk counterparts (Table S1, available at Annals of Occupational Hygiene online). The EC Guidance and NanoSafer recommended a similar or higher level of exposure control to that based on the OELs proposed for CNTs, TiO2, and silver, whereas CB Nanotool recommendations were either higher (TiO2 and CNTs) or lower (silver) in this scenario compared with the proposed OELs (Table 6). The ISO and ANSES tools required the most complete hazard data and yielded the highest level of exposure control. It is useful to the practitioner to understand how the input data can influence the CB findings, which factors are most influential on these results, and how these findings compare to existing OELs. Table 6. CB recommendations for the nanomaterials evaluated, compared with recommendations that align with proposed OELs. CB tool  Recommended control bands and performance-based exposure rangesa    Seek specialist advice/ adopt special measures  Containment  Engineering controls (fume hoods or LEV)    <1 µg/m3  1–10 µg/m3  10–1000 µg/m3  Recommended control approaches by CB tool  CB Nanotool  CNTs  Graphene Nanocellulose Silver Titanium dioxide  Silica (amorphous)  GoodNanoGuide  Graphene Nanocellulose Silica (amorphous) Silver  CNTs Titanium dioxide    ANSES  CNTs Graphene Nanocellulose Silica (amorphous) Silver Titanium dioxide      ISO  CNTs Graphene Nanocellulose Silica (amorphous) Silver Titanium dioxide      EC Guidance  CNTs Graphene Silica (amorphous) Silver  Nanocellulose Titanium dioxide    NanoSafer  CNTs Graphene  Nanocellulose Silver Titanium dioxide  Silica (amorphous)  Recommended control approaches that align with OELsb    Silver  CNTs  CNTs Silica (amorphous) Titanium dioxide  CB tool  Recommended control bands and performance-based exposure rangesa    Seek specialist advice/ adopt special measures  Containment  Engineering controls (fume hoods or LEV)    <1 µg/m3  1–10 µg/m3  10–1000 µg/m3  Recommended control approaches by CB tool  CB Nanotool  CNTs  Graphene Nanocellulose Silver Titanium dioxide  Silica (amorphous)  GoodNanoGuide  Graphene Nanocellulose Silica (amorphous) Silver  CNTs Titanium dioxide    ANSES  CNTs Graphene Nanocellulose Silica (amorphous) Silver Titanium dioxide      ISO  CNTs Graphene Nanocellulose Silica (amorphous) Silver Titanium dioxide      EC Guidance  CNTs Graphene Silica (amorphous) Silver  Nanocellulose Titanium dioxide    NanoSafer  CNTs Graphene  Nanocellulose Silver Titanium dioxide  Silica (amorphous)  Recommended control approaches that align with OELsb    Silver  CNTs  CNTs Silica (amorphous) Titanium dioxide  aEstimated from CB approaches shown in Fig. 1; these correspond to the OEL concentration ranges (also called OEBs) associated with the hazard categories in the ANSES (2010) and ISO (2014) CB tools. Note that some control band categories (Control Level 2 and Control Level 3) have been combined for ANSES and ISO to make the control bands’ results consistent between tools. And the control recommendations provided by NanoSafer differ from categories provided here (e.g. LEV, containment, special precautions) and include recommendations on respiratory protection (see Fig. 2). bBased on proposed OELs (Table S1, available at Annals of Occupational Hygiene online) and corresponding performance-based exposure ranges shown in this table. View Large The primary drivers for the control bands were the hazard scores in this small-scale production scenario. The hazards scoring approach used by the CB Nanotool resulted in the ranking of several of these ENMs to lower overall control bands than other decision tree tools (ISO, ANSES, Stoffenmanager Nano). The CB Nanotool approach combines scores for all hazards for the ENM and parent material into a total composite score, so positive research findings in any hazard category (carcinogenicity, mutagenicity, reproductive toxicity) do not automatically drive the ENM to the highest control band like the decision tree tools. For instance, for inhaled TiO2, the International Agency for Research on Cancer (IARC) has classified this chemical agent as a 2B (possibly carcinogenic to humans), which automatically places it in the highest hazard class for ISO, ANSES, NanoSafer, and Stoffenmanager Nano. However, when considered with all of the other hazard categories, TiO2 was scored as high severity (band 3 of 4) in CB Nanotool resulting in the containment control band. It is difficult to determine whether a more precautionary approach (as provided by ANSES, ISO, NanoSafer, and Stoffenmanager Nano) is a better choice given the lack of full hazard data on these ENMs. The best assessment that can be made at this point is to compare these tools to published risk assessments, which derive OELs based on a more thorough hazard analysis. However, variability in the proposed OELs for these ENMs also results in uncertainty in the appropriate level of exposure control (as shown in Table 6 and Table S1, available at Annals of Occupational Hygiene online). The primary driver for the different control bands by the EC Guidance tool was also the differences in hazard assessment. All of the ENMs evaluated were insoluble in water (based on the SDS or technical data sheets), but for CNTs and nanocellulose, these differences were due to the fibrous geometry/shape of these materials. And finally, the lower control bands for CNTs and TiO2 by the GoodNanoGuide were driven by availability of information on these ENMs. The GoodNanoGuide tool categorizes hazard groups by ‘known to be inert’ (hazard group A), ‘understand reactivity/function’ (hazard group B), or ‘unknown hazard’ (hazard group C). So an ENM such as TiO2 would be a group B because information is available, including that it has been classified as possibly carcinogenic by IARC and NIOSH, whereas other ENMs would be at a higher level because there is little or no information available on their hazard. That feature of GoodNanoGuide is mainly driven by EP (based on material form and task duration) and does not consider hazard potential in depth. The minimal assessment of hazard may limit the utility of the tool. Despite some differences in approach, most tools gave similar exposure bands (typically medium–high EP) primarily because the use scenario was consistent among these six ENMs evaluated. In general, exposure bands are driven by three primary factors: (i) material form; (ii) amount of material used; and (iii) process/task. In addition, all models except the Precautionary Matrix utilize dustiness as a factor in determining EP. Stoffenmanager Nano and NanoSafer, however, use much more detailed exposure models utilizing parameters such as process energy, volume, and ventilation rate of the work room, as well as frequency and duration of the evaluated task. With the CB Nanotool, the exposure probability score differed primarily on the dustiness determination of the ENM used and ranged from ‘probable exposure’ (CNT, high dustiness) to ‘likely exposure’ (TiO2, SiO2, medium dustiness; graphene, nanocellulose, unknown dustiness; and silver, low dustiness). Although dustiness can be a differentiating factor in these tools, another important factor is the amount of ENM used (by mass), and in this area, the tools differ considerably. For ISO, the highest exposure factor related to amount handled is applied when using >1 kg of powdered nanomaterials. However, with CB Nanotool, EC Guidance, and Precautionary Matrix, the highest material quantity category is bounded at a much lower level, i.e. less than 1 g of the nanomaterial. An evaluation of a few of these CB tools on a different set of nanoscale and microscale particles also showed a range of hazard and exposure outputs across tools and concluded that some of the recommendations may be excessive in some situations (Sánchez Jiménez et al., 2016). In general, the more specific and complete the input information, the more accurate and useful the CB tool outcomes would be expected, although the structure and flexibility of the tools to utilize specific parameter data (e.g. dustiness) also differ across tools. Such evaluations provide useful insights into the performance of these tools for the practitioner to gain an understanding of the utility and limitations of these various tools. Existing OELs for ENMs examined in CB tools One way to evaluate the utility and validity of the outcomes of these CB tools is to compare their recommended controls and associated performance levels with the OELs that have been proposed for these same or similar ENMs (as discussed above). OELs proposed by nonregulatory governmental agencies or by nongovernmental organizations (Table S1, available at Annals of Occupational Hygiene online) include nanoscale titanium dioxide, silica, silver, CNTs, and cellulose, which are all examined in this article. No published OELs for graphene were found in the literature or reported in a recent systematic review of ENM OELs (Mihalache et al., 2017). The OELs for nanoscale particles are typically lower airborne mass concentrations than the closest applicable regulatory OELs (Table S1, available at Annals of Occupational Hygiene online). For example, 5 mg/m3 is the OSHA permissible exposure limit for either graphite (synthetic), particles not otherwise regulated, or cellulose (respirable fraction, 8-h TWA concentration) (OSHA, 1983). This exposure concentration has been used in some nanotoxicology studies (e.g. for SWCNTs) (Shvedova et al., 2008). An OEL of 5 mg/m3 (i.e. 5000 µg/m3) is approximately one to three orders of magnitude greater than the proposed OELs for carbonaceous, metal, or metal oxide nanoparticles (Table S1, available at Annals of Occupational Hygiene online). Differences in both the toxicity of the substance and the data and methods used to derive the OELs could contribute to these differences. Some of the existing OELs may have included information on nanoscale particle exposures (although possibly not defined as such). For example, high combustion processes such as silver refining can produce airborne nanoscale particles (Miller et al., 2010). NIOSH recommended separate mass-based OELs for titanium dioxide by particle size (nanoscale/ultrafine and microscale/fine) (Table S1, available at Annals of Occupational Hygiene online) (NIOSH, 2011). The pulmonary toxicity of titanium dioxide and other poorly soluble particles is correlated with the total particle surface area, which is greater for an equal mass of smaller particles (NIOSH, 2011). Ease of use of CB tools for ENMs During the course of this study, several observations emerged regarding the user-friendly nature of the various tools. In particular, the level of information required and the complexity in completing the assessments differ among these tools. For quick, high-level assessments, the GoodNanoGuide, EC Guidance, and Precautionary Matrix provide results with minimal data. These tools were the easiest to complete given the minimal level of information required for the evaluation. EC Guidance tool categorizes nanomaterial hazard solely based on the physicochemical properties of biopersistence and particle/fiber shape, whereas GoodNanoGuide includes three simple bands for physicochemical properties, known to be inert, reactivity/function known, or unknown properties. CB Nanotool utilizes an intermediate level of information on both hazard and EP, which is at a level that would generally be available in a well-documented SDS. The hazard scoring approach of CB Nanotool is relatively easy to use by answering yes, no, or unknown to the hazard questions and assigning a score. CB Nanotool quantitatively addresses a lack of information by including ‘unknown’ as a choice, which defaults to a containment recommendation when no data are available. The transparent scoring approach in CB Nanotool allows the user to easily assess the drivers of the control band results to explore where changes to materials or use parameters (quantity, material form, etc.) could impact the control band. ISO, ANSES, Stoffenmanager Nano, and NanoSafer require more detailed information, and each of the sources shown in Table 4 was used to complete these assessments (to the extent that data were available). The ANSES and ISO tools are similar to each other and use the GHS system to provide a ready basis for standardization of inputs to hazard banding, which is useful but also may require more toxicology expertise [e.g. identifying lethal dose (LD50) and other endpoints] than does CB Nanotool, which includes yes/no options for the main endpoints. The ANSES and ISO tools address lack of information in the hazard banding by defaulting to the highest HB, which results in recommendations for higher levels of exposure control. The Stoffenmanager Nano and Precautionary Matrix are different in scope compared with the other tools because they address risk prioritization and do not lead to a control band. The Stoffenmanager Nano and NanoSafer tools utilized the most complex exposure banding approach requiring the most information from the user, including amount of material, process duration and frequency, work room volume, and ventilation rate among other parameters. The Precautionary Matrix assesses hazard potential through two primary physicochemical parameters: redox/catalytic activity and stability (half-life) in the body/environment. It provides a table of reactivity information for 12 nanomaterials. Stoffenmanager Nano provides guidance on hazard banding for 19 commonly used nanomaterials (Duuren-Stuurman et al., 2011, 2012). However, for those nanomaterials not included in the table, the HB is derived from an assessment of hazards based on the parent material. If the HB of the parent material is not known (or the material is not characterized according to carcinogenicity, mutagenicity, reproductive, and/or developmental effects), the tool defaults to the highest HB. Finally, several of the tools provide online or downloadable spreadsheets to help guide the user through the process. CB Nanotool provides a downloadable score-based spreadsheet with examples to help guide the user through the process. Stoffenmanager Nano, NanoSafer, and Precautionary Matrix have online tools to help facilitate the process. The various parameters and inputs to these tools, including those used in these assessments, are summarized in Tables 2 and 3. These input parameters are valuable information that are needed to use these CB tools to arrive at the recommended control bands. The parameters that were found to be drivers of the control band findings (e.g. availability of dustiness or specific health effects data), as discussed in this article, could be considered essential to obtaining more useful and reliable results from these CB tools. Evidence available for evaluating CB tools for ENMs Only a few types of ENMs have undergone relatively extensive toxicological evaluation, e.g. TiO2 and CNTs. Even for these ENMs, significant data gaps remain, especially for chronic adverse health effects. The limited hazard and dustiness data make it challenging to provide relevant information for the SDS. In addition, SDSs are not uniform and provide variable inconsistent amounts of information (Eastlake et al., 2012). A useful addition to SDSs would be a standardized format for CB tool input factors, which would provide the practitioner with more readily available information for applying CB methods to specific ENMs. In particular, the inclusion of standard information needed in CB tools would be useful information in SDSs. Current toxicity data, where available, would be especially useful in the SDSs, including the adverse effect levels in rodent studies, to evaluate severity and potency. In the future, the development of default HBs or OEBs for ENMs-based physicochemical properties and limited toxicology data would help facilitate the determination of appropriate control bands (Kuempel et al., 2012). In general, regardless of the CB strategy used, the uncertainty of the potential health risks of ENMs tends to result in a higher level of exposure control than would be used based on the ENM-specific OELs. These higher levels of exposure control appear to be due to the limited data on ENMs for many of the inputs in the CB tools, resulting in the default to the more protective categories in the absence of specific information. Indeed, a utility of these CB tools is that they generally recommend a high level of exposure control in the absence of specific information, which is a protective default. This approach is consistent with using greater precaution in the absence of data (Schulte and Salamanca-Buentello, 2007). Such strategies also encourage research to provide the more specific data needed to replace default assumptions. On the other hand, CB tools that do not discern among the hazards based on available data may not be sufficient for decision-making. This analysis has shown that certain factors that drive the CB decisions (e.g. default toxicity assumptions; dustiness levels) would be useful priorities for future research in order to improve the evidence basis for the application of these CB tools for ENMs. Dustiness data would be also useful in future validation studies as well as research studies correlating exposure with dustiness levels of ENMs by job task. Possible limitations in this analysis include the limited number of ENMs evaluated (six). OELs have been proposed for five of these ENMs (Table S1, available at Annals of Occupational Hygiene online). Using the proposed OELs for comparison to the HBs/OEBs is an uncertain criterion because the proposed OELs can vary widely and none are regulatory limits. Because the workplace use scenario was kept constant in this analysis, the findings may not apply to other use scenarios. Finally, the performance-based exposure concentrations (Table 6) have not been fully validated for the specific engineering control options, and comparison of the recommendations can be challenging because of the overlapping control bands across the CB tools. It should be noted that all of the CB tools evaluated recommended at a minimum the use of local exhaust ventilation for each of these ENMs (Table 6) in this exposure scenario (dry powder handling of small quantities for 4 h or fewer per day). For those ENMs with proposed OELs (Table 6), the associated performance-based exposure concentrations would also necessitate the use of local exhaust ventilation or a higher level of control. The most comprehensive validation studies performed to date have been on the CB Nanotool (Paik et al., 2008; Zalk et al., 2009), as discussed in an earlier systematic review (Eastlake et al., 2016). In a study of 32 job activities and nanomaterial combinations, the exposure control recommendations from CB Nanotool were reported to be at the same or higher level to those recommended independently by an experienced industrial hygienist for 28 (~88%) of the job activities. Roughly similar results were seen in this study, in which the control band recommendations from CB Nanotool were the same or lower than three of the four (75%) of the ENM OELs (Table 6). By comparison, the CB recommendations of EC Guidance, NanoSafer, and GoodNanoGuide were all the same or lower than the ENM OELs, whereas the ANSES and ISO tools recommended the lowest exposure level for each of the ENMs (Table 6). Sánchez Jiménez et al. (2016) provided some limited validation testing of the hazard and exposure results for three of these CB tools. They reported various differences in both the hazard and exposure results of the CB tools compared with reported toxicity and exposure measurement data. For example, the measured airborne number concentrations for cloisite (the only ENM in that evaluation) were lower for a weighing task, but higher for an extrusion task, compared with the results from the three CB tools (Sánchez Jiménez et al., 2016). A limitation in the validation studies to date is either the lack of data or the limited data on airborne exposure concentrations of ENMs associated with job activities and exposure controls; these data are needed to verify that the recommended controls achieved the expected results. Verification of CB tool recommendations with field-based measurements across jobs/tasks and working conditions has been previously recommended for general chemicals in industry (Jones and Nicas, 2006a, b). In addition, the lack of OELs for many ENMs does not permit verification that the recommended exposure control levels would be protective of workers’ health. An evaluation of CB tool recommendations with ENM-specific OELs, as illustrated in this article (Table 6), could be extended to additional ENMs as more toxicology data and OELs become available. Key findings This study demonstrated the use of the eight CB tools for ENMs currently available, showed what input data are needed, suggested several useful sources and websites to search for the information needed, and demonstrated the application and outcomes in a case study of six different ENMs, most of which have proposed OELs. A fixed workplace exposure scenario allowed focus on the role of the properties of the ENMs themselves, both biological and physicochemical. The key biological input parameters include qualitative hazard information and quantitative effect levels of the ENMs or bulk material [OELs or No Observed Adverse Effect Levels (NOAELs)]. The key physicochemical input parameters include dustiness, surface activity, shape (fibrous or not), and solubility. Understanding the information needed to utilize these tools and comparing the findings across these tools for a set of ENMs and fixed workplace exposure conditions helps the practitioner better understand how to select and use these tools. The purpose of this article is not to recommend the use of any specific tool but to illustrate and compare the inputs and findings of each tool under the same ENM and exposure scenarios. The findings of this study provide further input into the key drivers for the findings of each of these tools. Ultimately the selection of a tool depends on the purpose of the evaluation (e.g. risk prioritization or exposure control selection) as well as the availability of the input information. In using these tools, the practitioner will find different levels of information needed and complexity in completing the assessments. For quick, high-level assessments, the GoodNanoGuide and EC Guidance provide results with minimal data. Precautionary Guidance is the most basic indicating the need for caution. CB Nanotool utilizes a moderate level of hazard and EP information, while being implemented through an easy to understand tool. ISO, ANSES, Stoffenmanager Nano, and NanoSafer use the most information and require more effort in collecting data and completing the assessments. Regardless of which tool is selected, the user should record the sources of information and the input parameters selected in the application of any of these tools. This practice is consistent with good recordkeeping of the information used to arrive at CB findings and also facilitates further evaluation when new information becomes available. Fig. 2 presents an overall approach for using and evaluating the tools, whereas Tables 2 and 3 provide a template of the key information needed for each of these tools. Likewise, Table 4 provides a guide to online databases where input information on hazard inputs can be gathered. The findings of this study provided limited validation testing of CB tool results compared with OELs proposed for four of the ENMs evaluated in this study. These findings confirm those of other studies (Eastlake et al., 2016; Sánchez Jiménez et al., 2016) that more information is needed to validate these CB tools in order to determine whether the use of CB can adequately reduce nanomaterial worker exposures to safe levels. Moreover, the inclusion of the basic information needed in CB reported in a standard format on the SDSs would be especially useful in the application of these tools. Research to provide basic toxicity data is required to fill those data gaps. The following data gaps were identified in this study, which if filled would reduce uncertainty and improve confidence in the reliability of CB tool findings: The amount of information required differs across tools; yet most of the tools recommended higher levels of exposure control for each of the ENMs in this evaluation compared with the proposed OELs, primarily because of the limited hazard data. The default for ISO, ANSES, and Stoffenmanager Nano to the highest RL based on the highest individual hazard category (carcinogenicity, reproductive toxicity, mutagenicity) and for unknown hazards resulted in the highest priority or highest level of exposure control for each of the ENMs in this assessment. The composite hazard scoring approach by CB Nanotool resulted in lower levels of exposure control for some of the ENMs in this assessment. EC Guidance, NanoSafer, and GoodNanoGuide also recommended some diversity in the CB recommendations across ENMs. All tools recommended the use of local exhaust ventilation, at a minimum, for working with any of these ENMs. Research needs The lack of available data for the main inputs into these tools significantly reduces their utility, at this time. Key information that drives the hazard, exposure, and CB recommendations would be the most useful to reducing uncertainty and increasing confidence in the application of these tools, including: Quantities of ENMs currently produced and used in various applications by job/task needs to be updated and made available to researchers and the practitioner for use in emission potential scoring in CB. Correlation of ENM quantity, dustiness, and process with exposure should be assessed and validated with laboratory and workplace data. Airborne exposure measurement data for specific job activities and ENMs could be used in validation testing of the CB recommendations. More specific information is needed to help classify ENMs according to the hazard and exposure parameters. For instance, relatively minimal information on the surface reactivity and dustiness of ENMs would be useful to classify these as low, medium, or high, as requested in several of these tools (some do not include dustiness while Stoffenmanager Nano and NanoSafer have quantitative dustiness categories). Further evaluation and refinement of the hazard categories for ENMs are needed to reduce uncertainty given the limited toxicity data for ENMs. Several research efforts are underway in the USA and other countries to group ENMs by hazard potential and could ultimately provide default HBs or OEBs using physicochemical properties and limited toxicology data. In the meantime, the practitioner needs to be aware that while these CB tools can be useful in decision-making about exposure control options when working with ENMs, it is also important when selecting a tool to consider the objective, the information needed, and level of validation. Ideally, more than one tool should be selected for comparison of findings and to better inform decision-making. In addition, comparison of the CB results with any ENM-specific OELs would also be useful. Updating the initial evaluations as new data or tools become available will provide continued improvement in the CB of ENMs. Conclusions The several CB tools that have been developed for nanomaterials represent a good first step in developing approaches to control worker exposure given the paucity of data on many ENMs in use. Findings of this study showed that the ISO, ANSES, and Stoffenmanager Nano tools recommended the highest level of risk or exposure control for each of the ENMs in this assessment, whereas CB Nanotool, EC Guidance, NanoSafer, and GoodNanoGuide recommended more diverse CB recommendations across ENMs. Further validation of these tools is needed, including by comparing the performance-based exposure ranges of control approaches to the measurements of airborne exposure concentration of ENMs in a worker’s breathing zone during typical job tasks. Research towards characterizing dustiness of more ENMs will help improve the utility of these tools. Efforts should continue to synthesize data from workplace studies to gain a better understanding of how well factors such as dustiness represent worker exposures. In addition, as more health hazard data become available, for ENMs individually or within similar physicochemical groups, the ability to provide more constructive exposure control guidance on the range of ENMs seen in the workplace will improve. The findings from this study show that significant data gaps remain, resulting in uncertainty about the optimal selection of controls to protect workers producing and handling ENMs. Research that focuses on providing the key data inputs for these CB tools and including standard information on SDSs would facilitate the utility of these tools. In most of the evaluated CB tools, uncertainty in the available data is managed by the selection of higher RLs and more protective exposure control options. An important finding of this evaluation is that local exhaust ventilation was recommended at a minimum to control exposures to ENMs in the workplace. More stringent controls, such as process containment, may come at a higher installation or maintenance cost, and it may not be certain whether these are necessary given the unknown risks. However, these CB tools generally appear to be providing prudent exposure control guidance in the face of uncertainty. Supplementary Data Supplementary data are available at Annals of Work Exposures and Health online. Declaration The authors declare no conflict of interest relating to the material presented in this article. The findings and conclusions in this article are those of the author(s) and do not necessarily represent the official position of the National Institute for Occupational Safety and Health, Centers for Disease Control and Prevention. Mention of a specific product or company does not constitute endorsement by the Centers for Disease Control and Prevention. Acknowledgements We would like to acknowledge the assistance of D. Hammond, T.J. Lentz, J.L. Topmiller, and M. Gressel for their careful review of this article. We also thank Vanessa Williams for assistance in preparing the figures. This research was supported through the NIOSH Nanotechnology Research Center, the Division of Applied Research and Technology, and the Education and Information Division. We would like to thank the anonymous reviewers for their excellent comments and helpful suggestions on earlier drafts of this article. References Ader AW, Farris JP, Ku RH. ( 2005) Occupational health categorization and compound handling practice systems—roots, application and future. Chem Health Saf ; 12: 20– 26. Google Scholar CrossRef Search ADS   ANSES. ( 2010) Development of a specific control banding tool for nanomaterials. Maisons-Alfort Cedex: French Agency for Food, Environmental and Occupational Health & Safety (ANSES). Request no. 2008-SA-0407. Available at https://www.anses.fr/en/system/files/AP2008sa0407RaEN.pdf. Accessed 4 April 2016. Brooke IM. ( 1998) A UK scheme to help small firms control health risks from chemicals: toxicological considerations. Ann Occup Hyg ; 42: 377– 90. Google Scholar CrossRef Search ADS PubMed  Brouwer DH. ( 2012) Control banding approaches for nanomaterials. Ann Occup Hyg ; 56: 506– 14. Google Scholar PubMed  BSI. ( 2007) Nanotechnologies—Part 2: Guide to safe handling and disposal of manufactured nanomaterials. Book Nanotechnologies—Part 2: Guide to safe handling and disposal of manufactured nanomaterials . London: British Standards Institution. Dahm MM, Evans DE, Schubauer-Berigan MKet al.  ( 2012) Occupational exposure assessment in carbon nanotube and nanofiber primary and secondary manufacturers. Ann Occup Hyg ; 56: 542– 56. Google Scholar PubMed  Duuren-Stuurman B, Vink S, Brouwer Det al.   ( 2011) Stoffenmanager Nano: description of the conceptual control banding model . Zeist, Netherlands: Netherlands Organisation for Applied Scientific Research (TNO). Duuren-Stuurman B, Vink SR, Verbist KJet al.  ( 2012) Stoffenmanager Nano version 1.0: a web-based tool for risk prioritization of airborne manufactured nano objects. Ann Occup Hyg ; 56: 525– 41. Google Scholar PubMed  Eastlake A, Hodson L, Geraci Cet al.  ( 2012) A critical evaluation of material safety data sheets (MSDSs) for engineered nanomaterials. Chem Health Saf ; 19: 1– 8. Google Scholar CrossRef Search ADS PubMed  Eastlake A, Zumwalde R, Geraci C. ( 2016) Can control banding be useful for the safe handling of nanomaterials? A systematic review. J Nanopart Res ; 18: 1– 24. Google Scholar CrossRef Search ADS   European Commission. ( 2014) Guidance on the protection of the health and safety of workers from the potential risks related to nanomaterials at work: guidance for employers and health and safety practitioners . Belgium: European Commission, Employment Social Affairs & Inclusion. Available at http://ec.europa.eu/social/home.jsp?langId=en. European Committee for Standardization (CEN). ( 2013) Workplace exposure. Measurement of the dustiness of bulk materials. Requirements and choice of test methods (EN 15051-1:2013) . Brussels: CEN. Evans DE, Turkevich LA, Roettgers CTet al.  ( 2012) Dustiness of fine and nanoscale powders. Ann Occup Hyg ; 57: 261– 77. Google Scholar PubMed  Future Markets Inc.©( 2013) The global nanotechnology and nanomaterials industry . Future Markets, Inc., Technology Report No. 68. Available at https://futuremarketsinc.com/. Gardner RJ, Oldershaw PJ. ( 1991) Development of pragmatic exposure-control concentrations based on packaging regulation risk phrases. Ann Occup Hyg ; 35: 51– 9. Google Scholar PubMed  Garrod AN, Rajan-Sithamparanadarajah R. ( 2003) Developing COSHH Essentials: dermal exposure, personal protective equipment and first aid. Ann Occup Hyg ; 47: 577– 88. Google Scholar PubMed  Kulinowski KM, Jaffe MP. (2009) The goodnanoguide: a novel approach for developing good practices for handling engineered nanomaterials in an occupational setting. Nanotech. L. & Bus.; 6: 37–44. Henry BJ, Schaper KL. ( 1990) PPG’s Safety and Health Index System: a 10-year update of an in-plant Hazardous Materials Identification System and its relationship to finished product labeling, industrial hygiene, and medical programs. Am Ind Hyg Assoc J ; 51: 475– 84. Google Scholar CrossRef Search ADS PubMed  Höck J, Epprecht T, Furer Eet al.   ( 2013) Guidelines on the precautionary matrix for synthetic nanomaterials, version 3.0 . Berne: Federal Office for Public Health and Federal Office for the Environment. HSE. ( 2009) COSHH Essentials . Liverpool, England: Health and Safety Executive. ISO. ( 2014) ISO/TS 12901–2:2014 Nanotechnologies—occupational risk management applied to engineered nanomaterials—Part 2: Use of the control banding approach . Geneva, Switzerland: International Organization for Standardization. Isogai A. ( 2013) Wood nanocelluloses: fundamentals and applications as new bio-based nanomaterials. J Wood Sci .; 59: 449– 59. Google Scholar CrossRef Search ADS   Jensen KA, Saber AT, Kristensen HV, Koponen IK, Liguori B, Wallin H. (2013) NanoSafer vs. 1.1-nanomaterial risk assessment using first order modeling. In 6th International Symposium on Nanotechnology, Occupational and Environmental Health 2013 Oct 28 (Vol. 120). Jones RM, Nicas M. ( 2006a) Evaluation of COSHH Essentials for vapor degreasing and bag filling operations. Ann Occup Hyg ; 50: 137– 47. Jones RM, Nicas M. ( 2006b) Margins of safety provided by COSHH Essentials and the ILO Chemical Control Toolkit. Ann Occup Hyg ; 50: 149– 56. Kuempel ED, Castranova V, Geraci CLet al.  ( 2012) Development of risk-based nanomaterial groups for occupational exposure control. J Nanopart Res ; 14: 1029. Google Scholar CrossRef Search ADS PubMed  Lee JH, Kwon M, Ji JHet al.  ( 2011) Exposure assessment of workplaces manufacturing nanosized TiO2 and silver. Inhal Toxicol ; 23: 226– 36. Google Scholar CrossRef Search ADS PubMed  Liguori B, Hansen SF, Baun Aet al.   ( 2016) Control banding tools for occupational exposure assessment of nanomaterials—ready for use in a regulatory context? NanoImpact ; 2: 1– 17. Google Scholar CrossRef Search ADS   Maynard AD. ( 2007) Nanotechnology: the next big thing, or much ado about nothing? Ann Occup Hyg ; 51: 1– 12. Google Scholar PubMed  McKernan LT, Seaton M. ( 2014) The banding marches on: NIOSH proposes a new process for occupational exposure banding. The Synergist® ; 25: 44– 46. Mihalache R, Verbeek J, Graczyk Het al.  ( 2017) Occupational exposure limits for manufactured nanomaterials, a systematic review. Nanotoxicology ; 11: 7– 19. Google Scholar CrossRef Search ADS PubMed  Miller A, Drake PL, Hintz Pet al.  ( 2010) Characterizing exposures to airborne metals and nanoparticle emissions in a refinery. Ann Occup Hyg ; 54: 504– 13. Google Scholar PubMed  Naumann BD, Sargent EV, Starkman BSet al.   ( 1996) Performance-based exposure control limits for pharmaceutical active ingredients. Am Ind Hyg Assoc J ; 57: 33– 42. Google Scholar CrossRef Search ADS PubMed  NIOSH. ( 2009a) Approaches to safe nanotechnology: managing the health and safety concerns associated with engineered nanomaterials . Cincinnati, OH: U.S. Department of Health and Human Services, Public Health Service, Centers for Disease Control and Prevention, National Institute for Occupational Safety and Health, DHHS (NIOSH). Publication No. 2009–125. NIOSH. ( 2009b) Quantitative risk characterization and management of occupational hazards: control banding (CB)—a literature review and critical analysis . Cincinnati, OH: U.S. Department of Health and Human Services, Public Health Service, Centers for Disease Control and Prevention, National Institute for Occupational Safety and Health, DHHS (NIOSH). Publication 2009–152. NIOSH. ( 2011) Current Intelligence Bulletin 63: occupational exposure to titanium dioxide . Cincinnati, OH: U.S. Department of Health and Human Services, Public Health Service, Centers for Disease Control and Prevention, National Institute for Occupational Safety and Health, DHHS (NIOSH). Publication No. 2011–160. NIOSH. ( 2012) General safe practices for working with engineered nanomaterials in research laboratories . Cincinnati, OH: U.S. Department of Health and Human Services, Centers for Disease Control and Prevention, National Institute for Occupational Safety and Health, DHHS (NIOSH). Publication No. 2012–147. NIOSH. ( 2013a) Current Intelligence Bulletin 65: occupational exposure to carbon nanotubes and nanofibers . Cincinnati, OH: U.S. Department of Health and Human Services, Public Health Service, Centers for Disease Control and Prevention, National Institute for Occupational Safety and Health, DHHS (NIOSH). Publication No. 2013–145. NIOSH. ( 2013b) Current strategies for engineering controls in nanomaterial production and downstream handling processes . Cincinnati, OH: U.S. Department of Health and Human Services, Public Health Service, Centers for Disease Control and Prevention, National Institute for Occupational Safety and Health, DHHS (NIOSH). Publication No. 2014–102. NIOSH. ( 2017) External review draft—Current Intelligence Bulletin: the occupational exposure banding process: guidance for the evaluation of chemical hazards . Cincinnati, OH: U.S. Department of Health and Human Services, Public Health Service, Centers for Disease Control and Prevention, National Institute for Occupational Safety and Health. 8 March. Novoselov KS, Fal’ko VI, Colombo Let al.  ( 2012) A roadmap for graphene. Nature ; 490: 192– 200. Google Scholar CrossRef Search ADS PubMed  Nowack B, Krug HF, Height M. ( 2011) 120 years of nanosilver history: implications for policy makers. Environ Sci Technol ; 45: 1177– 83. Google Scholar CrossRef Search ADS PubMed  OSHA. ( 1983) OSHA safety and health standards. 29 CFR 1910.1000 . Washington, D.C.: U.S. Department of Labor, Occupational Safety and Health Administration (OSHA). 29 CFR 1910.1000. OSHA. ( 2012) Appendix A to §1910.1200—health hazard criteria. Fed Reg ; 77: 17574– 896. Paik SY, Zalk DM, Swuste P. ( 2008) Application of a pilot control banding tool for risk level assessment and control of nanoparticle exposures. Ann Occup Hyg ; 52: 419– 28. Google Scholar PubMed  Piccinno F, Gottschalk F, Seeger Set al.   ( 2012) Industrial production quantities and uses of ten engineered nanomaterials in Europe and the world. J Nanopart Res ; 14: 1– 11. Google Scholar CrossRef Search ADS PubMed  Sánchez Jiménez A, Varet J, Poland Cet al.  ( 2016) A comparison of control banding tools for nanomaterials. J Occup Environ Hyg ; 13: 936– 49. Google Scholar CrossRef Search ADS PubMed  Sargent EV, Kirk GD. ( 1988) Establishing airborne exposure control limits in the pharmaceutical industry. Am Ind Hyg Assoc J ; 49: 309– 13. Google Scholar CrossRef Search ADS PubMed  Schulte P, Geraci C, Hodson Let al.   ( 2013) Overview of risk management for engineered nanomaterials. J. Phys. Conf. Ser . 429. Schulte P, Geraci C, Zumwalde Ret al.  ( 2008) Occupational risk management of engineered nanoparticles. J Occup Environ Hyg ; 5: 239– 49. Google Scholar CrossRef Search ADS PubMed  Schulte PA, Salamanca-Buentello F. ( 2007) Ethical and scientific issues of nanotechnology in the workplace. Cien Saude Colet ; 12: 1319– 32. Google Scholar CrossRef Search ADS PubMed  Shvedova AA, Kisin E, Murray ARet al.  ( 2008) Inhalation vs. aspiration of single-walled carbon nanotubes in C57BL/6 mice: inflammation, fibrosis, oxidative stress, and mutagenesis. Am J Physiol Lung Cell Mol Physiol ; 295: L552– 65. Google Scholar CrossRef Search ADS PubMed  UNECE. ( 2011) Globally harmonized system of classification and labelling of chemicals (GHS) . 4th rev. edn. Geneva, Switzerland: United Nations Economic Commission for Europe. http://www.unece.org/trans/danger/publi/ghs/ghs_rev04/04files_e.html. Accessed 4 April 2016. PubMed PubMed  Wijnhoven SW, Peijnenburg WJ, Herberts CAet al.   ( 2009) Nano-silver-a review of available data and knowledge gaps in human and environmental risk assessment. Nanotoxicology ; 3: 109– 38. Google Scholar CrossRef Search ADS   Zalk DM, Nelson DI. ( 2008) History and evolution of control banding: a review. J Occup Environ Hyg ; 5: 330– 46. Google Scholar CrossRef Search ADS PubMed  Zalk DM, Paik SY, Swuste P. ( 2009) Evaluating the control banding nanotool: a qualitative risk assessment method for controlling nanoparticle exposures. J Nanopart Res ; 11: 1685– 704. Google Scholar CrossRef Search ADS   Published by Oxford University Press on behalf of The British Occupational Hygiene Society 2018.

Journal

Annals of Work Exposures and Health (formerly Annals Of Occupational Hygiene)Oxford University Press

Published: Apr 1, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 12 million articles from more than
10,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Unlimited reading

Read as many articles as you need. Full articles with original layout, charts and figures. Read online, from anywhere.

Stay up to date

Keep up with your field with Personalized Recommendations and Follow Journals to get automatic updates.

Organize your research

It’s easy to organize your research with our built-in tools.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

Monthly Plan

  • Read unlimited articles
  • Personalized recommendations
  • No expiration
  • Print 20 pages per month
  • 20% off on PDF purchases
  • Organize your research
  • Get updates on your journals and topic searches

$49/month

Start Free Trial

14-day Free Trial

Best Deal — 39% off

Annual Plan

  • All the features of the Professional Plan, but for 39% off!
  • Billed annually
  • No expiration
  • For the normal price of 10 articles elsewhere, you get one full year of unlimited access to articles.

$588

$360/year

billed annually
Start Free Trial

14-day Free Trial