TY - JOUR
AU1 - Henk, Goede,
AU2 - Yvette, Christopher-de Vries,
AU3 - Eelco, Kuijpers,
AU4 - Wouter, Fransman,
AB - Abstract This review describes an evaluation of the effectiveness of Risk Management Measures (RMM) for nanomaterials in the workplace. Our aim was to review the effectiveness of workplace RMM for nanomaterials and to determine whether established effectiveness values of conventional chemical substances applied for modelling purposes should be adopted or revised based on available evidence. A literature review was conducted to collate nano-specific data on workplace RMM. Besides the quantitative efficacy values, the library was populated with important covariables such as the study design, measurement type, size of particles or agglomerates/aggregates, and metrics applied. In total 770 records were retrieved from 41 studies for three general types of RMM (engineering controls, respiratory equipment and skin protective equipment: gloves and clothing). Records were found for various sub-categories of the different types of RMM although the number of records for each was generally limited. Significant variation in efficacy values was observed within RMM categories while also considering the respective covariables. Based on a comparative evaluation with efficacy values applied for conventional substances, adapted efficacy values are proposed for various RMM sub-categories (e.g. containment, fume cupboards, FFP2 respirators). It is concluded that RMM efficacy data for nanomaterials are limited and often inconclusive to propose effectiveness values. This review also shed some light on the current knowledge gaps for nanomaterials related to RMM effectiveness (e.g. ventilated walk-in enclosures and clean rooms) and the challenges foreseen to derive reliable RMM efficacy values from aggregated data in the future. control effectiveness, dermal exposure, efficacy, efficiency, engineering control, inhalation exposure, nano, penetration, Risk Management Measures, skin protection, workplace Introduction As the rapid growth in the use of nanomaterials continues, it is essential that exposure to these materials is adequately controlled. Risk management of nanomaterials has received considerable attention in recent years, especially in the USA (NIOSH, 2013) and as part of various European Union (EU) research projects – amongst others LIFE NanoRISK, SUN, NANoREG, Scaffold, and GUIDEnano. It is apparent from these initiatives that there is limited information available regarding the efficiency of Risk Management Measures (RMM) for nanomaterials, with a handful of open source or published reviews on the effectiveness of workplace RMM (OECD, 2009; NIOSH, 2013; Frijns, 2016; Oksel et al., 2016). Surveys conducted amongst nanotechnology companies that produce or use engineered nanomaterials indicate that engineering controls (EC) such as enclosure, fume cupboards, and glove boxes, and personal protection such as respirators, gloves, and protective clothing are commonly applied to protect workers (ICON, 2006; Conti et al., 2008; Schubauer-Berigan et al., 2015). Most of these control measures are broadly applied in traditional workplaces as well; however, their effectiveness for nanomaterials is often unknown or not quantitatively evaluated for different workplaces and conditions of use. A completely new risk management approach to control exposures to nanomaterials is not anticipated in light of the similarity of processes, activities, and control measures currently applied for both nanomaterials and other chemical workplace scenarios. However, improvements or adjustments of existing control measures and practices may be required in some instances to specifically address nano-specific issues. In particular, the relevance of RMM for dermal exposure to nanomaterials should be carefully scrutinized. For example, Brouwer et al. (2016) concluded that dermal exposure to nanoparticles becomes an issue when it concerns specific nanomaterials (e.g. sensitizing metals/oxides/impurities) and/or non-rigid nanoparticles, coupled with compromised skin integrity, and/or frequent perioral contacts with contaminated hands. A major issue with extracting a suitable effectiveness value for risk assessment purposes is the substantial variation found within RMM types and selected covariables within studies evaluated (Fransman et al., 2008). Presently, information on the quantitative RMM efficiency can be obtained from only a limited number of sources, e.g. the CEFIC RMM Library (CEFIC, 2016), the OECD emission scenario documents (OECD, 2016), and the Exposure Control Efficacy Library (ECEL) (Fransman et al., 2008). These sources are focused on chemical substances that represent substantial variation in efficiency of RMMs and do not specifically account for nanomaterials. As part of the GUIDEnano and SUN projects, our aim was to conduct a review of RMM nanomaterial-specific efficiency and to derive quantitative efficiency values for nano-exposure and risk modelling purposes. This study focused on workplace RMM and specifically on EC, respiratory protective equipment (RPE) and skin protective equipment (SPE): gloves and clothing. Our aim was to determine (i) whether there are any differences in the effectiveness of control measures for nanomaterials versus conventional materials, and (ii) whether currently used effectiveness values of RMM relevant for conventional substances should be replaced, refined, or extrapolated based on evidence from nanomaterial-specific data. Methods Literature review A literature review (2005–2016) was conducted on topics related to the risk management and control measures associated with nanomaterials. Search strategies in Scopus and PubMed included keywords ‘nanomaterial*’, ‘control*’, ‘protect*’, ‘effective*’, ‘penetrat*’, and ‘worker*’. The latter search was narrowed down to address-specific types of RMM: ‘engineering control*’, ‘ventilat*’, ‘respirator*’, ‘mask*’, ‘clothing’, and ‘glove*’. Besides the published scientific studies, Google (Scholar) was also searched for (grey) data using the above search strings. CORDIS was searched for relevant EU-funded projects from which (experimental) data were also obtained and considered for review. The identified studies were cross-checked with recent reviews (Oksel et al., 2016; Frijns et al., 2016) and supplemented with relevant studies where appropriate. The criteria applied for the screening procedure are described elsewhere (Fransman et al., 2008). Briefly, the focus was on data with basic information on the measurement strategy and methodology, description of the RMM, and quantitative data to derive an efficiency value. Additional criteria were applied to account for nanomaterials: (i) the substance tested must contain nanomaterials, defined here as engineered nanomaterials or incidental nanomaterials, including nano-objects, their aggregates, and/or agglomerates (NOAA) (ISO, 2012); (ii) air concentrations (as particle number) must be corrected for background concentrations (Methner, 2008; Tsai et al., 2012); and (iii) the sampling strategy and measurement types were broadened to include upstream and downstream static and dynamic test methods for skin protective devices. Data formatting The effectiveness of RMM can be expressed in different ways depending on the type of control measure under investigation. In the case of engineering controls, an efficacy value is often applied that represents a reduction factor, i.e. Ccontrol on/Ccontrol off (Fransman et al., 2008). A protection factor (PF) or percentage effectiveness (%) is generally applied for respirators, protective clothing and gloves and calculated as: PF = Cout/Cin, with Cout being the concentration of challenge substance outside the device and Cin the concentration that permeated or penetrated to the inside. Data were formatted per record, with each record allocated with an efficacy value (expressed as an percentage) that represents the reported effectiveness (as estimated from Ccontrol on/Ccontrol off, or Cout/Cin), which constitutes the average concentration of a time series for a given particle size or particle size bin. A preference was given for geometric mean values, followed by median and arithmetic mean values (Fransman et al., 2008) as adopted from the study in question, or directly obtained as a percentage reduction as presented in a given study. Current measurement strategies indicate that non-mass-based metrics such as particle number are the preferred methods to measure nanomaterial exposure (Brouwer et al., 2009), while studies on the evaluation of effectiveness of RMM for nanomaterials usually report either particle number (# cm−3) or mass per volume (e.g. mg m−3) concentrations. An attempt was made to convert all the data to either of these metrics, as formulated by Hinds (1999). However, such conversions require a selection of an average particle size (and density), which was generally not reported in studies for both the pre- and post-tests. Consequently, it was decided to only include the ‘reported metric’ as a covariate in the data analysis. RMM categories The RMM (sub)categories for engineering controls previously applied in ECEL (Fransman et al., 2008) was adapted and aligned with the RMMs proposed in the Advanced REACH Tool (ART) (Fransman et al., 2011). An in-depth description of a wide range of these engineering controls is also publicly available (HSE, 2017; Fransman et al., 2013). For respiratory equipment, the (sub)categories proposed are largely based on Regulation 29 CFR 1910.134 (OSHA, 2002) – with similar (sub)categories such as FFP2 and N95 grouped together. SPE is broadly based on the Directive 89/686/EEC, with the main focus on chemical protection (category III), which consists of six types of ‘protective overalls and disposable coveralls’. The latter list of SPE was supplemented with ‘other non-specified work clothing and work wear’, which was not recorded or presented as Cat III clothing in the data. This group was broadly categorized as woven, non-woven, and coated clothing typically found in the workplace and also representative of surveys conducted in nanotechnology companies (Frijns, 2016; Oksel et al., 2016). There is inevitably some overlap between the latter two groups, as only a small number of studies clearly distinguish between the clothing protection categories and their respective types. This review excludes protective gear such as shoes, eye, and face protection. Data analysis Descriptive summary statistics were calculated for each RMM type (Engineering controls (EC), RPE, SPE: gloves and clothing, RMM category (e.g. containment) and RMM sub-category (e.g. high-level containment). Bivariate analyses were used to examine the influence of study characteristics (covariates) on efficacy values of the respective RMM categories and sub-categories. The choice of study characteristics was based on a priori assumptions of their potential influence on efficacy or on previous empirical research. These included the study design, sampling strategy, and measurement type as described by Fransman et al., (2008), as well as ‘particle size’ and ‘reported metric’. For each record, the ‘particle size’ describes the particle size of concern (or relevant particle size bin) that was evaluated from which an efficiency value was derived, and for which a crude distinction was made in the analysis between <200 nm and ≥200 nm for all RMM types. This was to account for the diffusion mechanisms associated with nano-sized particles (Rengasamy et al., 2008; Schulte et al., 2008). Records for which a distinction could not be made between the above-mentioned size categories (e.g. when a size range of 5.6–560 nm was evaluated; or when filter-based methods were used) were excluded from the analysis and coded as ‘missing data’ (as indicated in Table 2). A number of more specific covariates of RMM types were included (i.e. RPE: fit/leak and IP-treatment) and the ‘flow rate/breathing rate’ of RPE and the ‘test method’ applied for clothing. The retention of covariates for use in subsequent models to derive adjusted estimates of efficacy values was based on statistical considerations: if they were significant in bivariate analyses, P < 0.05. Although the type of nanomaterial (e.g. TiO2), physical state (e.g. solid–aerolised particles), type of activity or process, and test method are considered important variables, datasets were too small to include these parameters in a meaningful way in the analysis. As the distribution of the efficiency values was negatively skewed, it was reflected and transformed with natural logarithms to facilitate statistical analysis. All analyses were conducted using the SAS Analytical System (SAS version 9.4; SAS Institute, Cary, NC, USA). Proposal efficacy values The outcome of the data analysis was compared with existing reference values of RMM relevant for conventional substances. In case sufficient evidence was available to justify an adaptation of existing efficacy values, the rationales for such changes are described in more detail. Results Of more than 200 studies that were initially screened, 96 provided quantitative and/or qualitative information on the effectiveness of the preselected RMM (sub)categories. 770 records were retrieved from the subset of these (n = 41) (Table 1) that met the previously described screening criteria to derive quantitative efficacy values. Table 1. Frequency of nano-specific RMM categories and study types for each RMM type RMM type Studies Records (Semi) experimental Cross-sectional Intervention Total All 45a (41) 730 34 6 770  Engineering controls 13 44 34 6 84  RPE 18 415 0 0 415  Gloves 3 41 0 0 41  Clothing 11 230 0 0 230 RMM type Studies Records (Semi) experimental Cross-sectional Intervention Total All 45a (41) 730 34 6 770  Engineering controls 13 44 34 6 84  RPE 18 415 0 0 415  Gloves 3 41 0 0 41  Clothing 11 230 0 0 230 References: Engineering controls: 1–13; RPE: 14–31; gloves: 27, 32, 33; clothing: 27, 28, 33–41 [full references shown in Supplementary data at Annals of Occupational Hygiene online]. aFrom some studies (n = 4), data were obtained for more than one RMM type; hence, effectively 41 studies were evaluated. View Large Table 1. Frequency of nano-specific RMM categories and study types for each RMM type RMM type Studies Records (Semi) experimental Cross-sectional Intervention Total All 45a (41) 730 34 6 770  Engineering controls 13 44 34 6 84  RPE 18 415 0 0 415  Gloves 3 41 0 0 41  Clothing 11 230 0 0 230 RMM type Studies Records (Semi) experimental Cross-sectional Intervention Total All 45a (41) 730 34 6 770  Engineering controls 13 44 34 6 84  RPE 18 415 0 0 415  Gloves 3 41 0 0 41  Clothing 11 230 0 0 230 References: Engineering controls: 1–13; RPE: 14–31; gloves: 27, 32, 33; clothing: 27, 28, 33–41 [full references shown in Supplementary data at Annals of Occupational Hygiene online]. aFrom some studies (n = 4), data were obtained for more than one RMM type; hence, effectively 41 studies were evaluated. View Large Inclusion of data For engineering controls, 13 studies (84 data points) were found that met the inclusion criteria and were included in this evaluation. Considering the wide range of available engineering controls (Fransman et al., 2011), limited nano-specific data are available to evaluate engineering controls – which included experimental, cross-sectional, and intervention studies. Nanomaterials included TiO2, CNTs, graphene platelets, and various metal(oxide)s. Eighteen studies were found for RPE that met the inclusion criteria with in total 415 data points. Most data are available on filtering half mask (disposable), followed by filtering half masks with elastomeric face pieces. Studies evaluating RPE were primarily experimental studies using test chambers with a manikin or a head-form for filter media testing under simulated conditions. For SPE, three studies were found to evaluate gloves with in total 41 data points. Nano-specific data of protective clothing could only be found for 5 clothing types from 11 sources (230 data points). These experimental studies used a wide variety of testing methods (e.g. through-diffusion and filtration-based test bench studies), test conditions (e.g. aerosolized solids), and types of nanomaterials (NaCl, Ag, Fe3O4, TiO2, SiO2). Analyses The distribution of study characteristics for each type of RMM and an indication of the strength of their bivariate associations with average estimated efficacy are presented in Table 2. Of the studies on gloves, only differences in glove types based on material type could be considered since information on some parameters were too limited. The results of univariate and bivariate analyses including (Spearman’s) correlation analysis were used to inform the development of multiple regression models to derive corrected average estimates of efficacy for RMM categories where RMM category sizes were sufficiently large (Table 3). In general, variation in efficacy of different types of RMM and across RMM categories was significant and varied considerably. Due to the limited number of records for certain RMM categories and the varied distribution of study characteristics across sub-categories, adjusted/corrected efficacy values were obtainable for a subset of (sub)categories only. The most correction was required for the efficacy of the RPE sub-category ‘filtering half mask (disposable)’, which included the metric, particle size, fit/leak, treatment, and the flow rate/breathing rate. Table 2. Distribution of study characteristics for each type of nanomaterial-specific RMM Parameter N Estimated average efficacy (%)a 95% CI (%) Lower Upper Engineering controls 84 96.1 94.4 97.4  Study design   (Semi-)experimental 44 95.8 93.0 97.6   Cross-sectional 34 96.0 92.9 97.9   Intervention 6 98.4 92.8 100.2  Measurement type   Personal/subject 20 97.3 94.1 99.0   Stationary (source) 64 95.7 93.5 97.3  Particle sizeb   <200 nm 28 94.8 90.0 97.5   ≥200 nm 33 97.5 95.0 98.9  Metric   Mass 34 96.0 92.9 97.9   Particle number 50 96.2 93.9 97.8 RPE 415 97.0 96.5 97.6  Measurement type*   Filtration or permeation (static) 353 96.5 95.8 97.0   Personal/subject 62 99.2 98.5 99.7  Particle sizeb**   <200 nm 311 97.2 96.6 97.7   ≥200 nm 86 95.6 93.8 96.9  Metric*   Mass 18 99.8 98.8 100.3   Particle number 397 96.8 96.3 97.3  Fit/leakb*   Good fit (fit test value ≥ 100; 1.25–2.5 mm × 2 leak) 54 99.4 98.8 99.8   Moderate fit (‘normal fit’; 1.25–2.5 mm × 2 leak) 117 95.4 94.0 96.5   Poor fit (‘loose fit’; >2.5 mm × 2 leak) 8 82.3 56.9 93.1   Sealed (incl. treated) 218 96.7 95.9 97.3  Treatment*   Discharged or IP-treated 47 85.6 79.2 90.2   Non-treated 356 97.5 97.1 98.0 Gloves 41 96.8 94.4 98.3  Nitrile gloves (thickness)b   Nitrile gloves (thick) 5 97.2 79.6 100.3   Nitrile gloves (thin) 5 88.4 29.9 98.8 Clothing 230 89.0 86.6 91.1  Measurement type*   Filtration or permeation (static) 150 74.5 68.5 79.4   Personal/subject 80 98.3 97.4 99.0  Particle size*   <200 nm 198 91.4 89.0 93.3   ≥200 nm 32 54.7 20.4 74.4  Metric*   Mass 110 95.7 94.0 97.0   Particle number 120 75.7 67.9 81.6 Parameter N Estimated average efficacy (%)a 95% CI (%) Lower Upper Engineering controls 84 96.1 94.4 97.4  Study design   (Semi-)experimental 44 95.8 93.0 97.6   Cross-sectional 34 96.0 92.9 97.9   Intervention 6 98.4 92.8 100.2  Measurement type   Personal/subject 20 97.3 94.1 99.0   Stationary (source) 64 95.7 93.5 97.3  Particle sizeb   <200 nm 28 94.8 90.0 97.5   ≥200 nm 33 97.5 95.0 98.9  Metric   Mass 34 96.0 92.9 97.9   Particle number 50 96.2 93.9 97.8 RPE 415 97.0 96.5 97.6  Measurement type*   Filtration or permeation (static) 353 96.5 95.8 97.0   Personal/subject 62 99.2 98.5 99.7  Particle sizeb**   <200 nm 311 97.2 96.6 97.7   ≥200 nm 86 95.6 93.8 96.9  Metric*   Mass 18 99.8 98.8 100.3   Particle number 397 96.8 96.3 97.3  Fit/leakb*   Good fit (fit test value ≥ 100; 1.25–2.5 mm × 2 leak) 54 99.4 98.8 99.8   Moderate fit (‘normal fit’; 1.25–2.5 mm × 2 leak) 117 95.4 94.0 96.5   Poor fit (‘loose fit’; >2.5 mm × 2 leak) 8 82.3 56.9 93.1   Sealed (incl. treated) 218 96.7 95.9 97.3  Treatment*   Discharged or IP-treated 47 85.6 79.2 90.2   Non-treated 356 97.5 97.1 98.0 Gloves 41 96.8 94.4 98.3  Nitrile gloves (thickness)b   Nitrile gloves (thick) 5 97.2 79.6 100.3   Nitrile gloves (thin) 5 88.4 29.9 98.8 Clothing 230 89.0 86.6 91.1  Measurement type*   Filtration or permeation (static) 150 74.5 68.5 79.4   Personal/subject 80 98.3 97.4 99.0  Particle size*   <200 nm 198 91.4 89.0 93.3   ≥200 nm 32 54.7 20.4 74.4  Metric*   Mass 110 95.7 94.0 97.0   Particle number 120 75.7 67.9 81.6 The strength of their bivariate associations with average efficacy values (if any) is indicated. aBack-transformed regression coefficient. bMissing data (e.g. filter-based samples; no distinctive size bins). *P < 0.0001; **P < 0.05. View Large Table 2. Distribution of study characteristics for each type of nanomaterial-specific RMM Parameter N Estimated average efficacy (%)a 95% CI (%) Lower Upper Engineering controls 84 96.1 94.4 97.4  Study design   (Semi-)experimental 44 95.8 93.0 97.6   Cross-sectional 34 96.0 92.9 97.9   Intervention 6 98.4 92.8 100.2  Measurement type   Personal/subject 20 97.3 94.1 99.0   Stationary (source) 64 95.7 93.5 97.3  Particle sizeb   <200 nm 28 94.8 90.0 97.5   ≥200 nm 33 97.5 95.0 98.9  Metric   Mass 34 96.0 92.9 97.9   Particle number 50 96.2 93.9 97.8 RPE 415 97.0 96.5 97.6  Measurement type*   Filtration or permeation (static) 353 96.5 95.8 97.0   Personal/subject 62 99.2 98.5 99.7  Particle sizeb**   <200 nm 311 97.2 96.6 97.7   ≥200 nm 86 95.6 93.8 96.9  Metric*   Mass 18 99.8 98.8 100.3   Particle number 397 96.8 96.3 97.3  Fit/leakb*   Good fit (fit test value ≥ 100; 1.25–2.5 mm × 2 leak) 54 99.4 98.8 99.8   Moderate fit (‘normal fit’; 1.25–2.5 mm × 2 leak) 117 95.4 94.0 96.5   Poor fit (‘loose fit’; >2.5 mm × 2 leak) 8 82.3 56.9 93.1   Sealed (incl. treated) 218 96.7 95.9 97.3  Treatment*   Discharged or IP-treated 47 85.6 79.2 90.2   Non-treated 356 97.5 97.1 98.0 Gloves 41 96.8 94.4 98.3  Nitrile gloves (thickness)b   Nitrile gloves (thick) 5 97.2 79.6 100.3   Nitrile gloves (thin) 5 88.4 29.9 98.8 Clothing 230 89.0 86.6 91.1  Measurement type*   Filtration or permeation (static) 150 74.5 68.5 79.4   Personal/subject 80 98.3 97.4 99.0  Particle size*   <200 nm 198 91.4 89.0 93.3   ≥200 nm 32 54.7 20.4 74.4  Metric*   Mass 110 95.7 94.0 97.0   Particle number 120 75.7 67.9 81.6 Parameter N Estimated average efficacy (%)a 95% CI (%) Lower Upper Engineering controls 84 96.1 94.4 97.4  Study design   (Semi-)experimental 44 95.8 93.0 97.6   Cross-sectional 34 96.0 92.9 97.9   Intervention 6 98.4 92.8 100.2  Measurement type   Personal/subject 20 97.3 94.1 99.0   Stationary (source) 64 95.7 93.5 97.3  Particle sizeb   <200 nm 28 94.8 90.0 97.5   ≥200 nm 33 97.5 95.0 98.9  Metric   Mass 34 96.0 92.9 97.9   Particle number 50 96.2 93.9 97.8 RPE 415 97.0 96.5 97.6  Measurement type*   Filtration or permeation (static) 353 96.5 95.8 97.0   Personal/subject 62 99.2 98.5 99.7  Particle sizeb**   <200 nm 311 97.2 96.6 97.7   ≥200 nm 86 95.6 93.8 96.9  Metric*   Mass 18 99.8 98.8 100.3   Particle number 397 96.8 96.3 97.3  Fit/leakb*   Good fit (fit test value ≥ 100; 1.25–2.5 mm × 2 leak) 54 99.4 98.8 99.8   Moderate fit (‘normal fit’; 1.25–2.5 mm × 2 leak) 117 95.4 94.0 96.5   Poor fit (‘loose fit’; >2.5 mm × 2 leak) 8 82.3 56.9 93.1   Sealed (incl. treated) 218 96.7 95.9 97.3  Treatment*   Discharged or IP-treated 47 85.6 79.2 90.2   Non-treated 356 97.5 97.1 98.0 Gloves 41 96.8 94.4 98.3  Nitrile gloves (thickness)b   Nitrile gloves (thick) 5 97.2 79.6 100.3   Nitrile gloves (thin) 5 88.4 29.9 98.8 Clothing 230 89.0 86.6 91.1  Measurement type*   Filtration or permeation (static) 150 74.5 68.5 79.4   Personal/subject 80 98.3 97.4 99.0  Particle size*   <200 nm 198 91.4 89.0 93.3   ≥200 nm 32 54.7 20.4 74.4  Metric*   Mass 110 95.7 94.0 97.0   Particle number 120 75.7 67.9 81.6 The strength of their bivariate associations with average efficacy values (if any) is indicated. aBack-transformed regression coefficient. bMissing data (e.g. filter-based samples; no distinctive size bins). *P < 0.0001; **P < 0.05. View Large Table 3. Estimated average nano-specific efficacy values, 95% CI for individual RMMs, reference values for conventional substances, and proposed values for nanomaterialsa RMM N Estimated average nano-specific efficacy (%) 95% CI (%) Reference valueb conventional substances Proposed average nano-specific efficacy (%)c Lower Upper Value Min–max Engineering controls  Containmentd* 21 99.5 98.4 100.1 — — —   High level 14 99.7 98.9 100.2 99.9 99.9 90–99.99   Medium level 7 98.0 95.4 99.4 99 98 80–99  LEV – capturing hoods** 46 95.2 92.7 96.9 — — —   Fixed hood 29 97.5 94.2 99.2 90 90 70–98   Movable hood 17 90.0 77.0 96.0 50 70 30–98  LEV – enclosing hoods 6 76.9 35.7 92.1 — — —   Fume cupboard 5 77.3 14.2 94.5 99 90 50–99  Glove boxes (low specificatione) 4 93.4 75.4 98.8 99.9 98 95–99.95  Suppression techniques (at point of release) 3 94.6 74.7 99.4 90 90 70–90  Unidirectional room airflow systems (no screens, dust) 3 90.1 56.4 98.3 80 80 50–95 RPE  Unspecified nuisance mask 24 21.2 −19.9 48.4 — 10 0–60  Filtering half mask (disposable)** 240 97.2 96.7 97.7 — — —   FFP1 4 93.4 (83.7f) 72.3 (63.0f) 99.0 (93.1f) 75 65 30–75   FFP2/N95 147 95.0 (84.7f) 91.5 (79.3f) 97.2 (88.7f) 90–92 80 40–97   FFP3/N99/P100 89 98.1 (92.5f) 96.2 (89.3f) 99.3 (94.8f) 95–98 90 50–98  Filtering half mask (elastomeric face piece)** 117 97.7 97.0 98.2 — — —   P2/N95 22 98.1 96.1 99.3 90–92 90 35–95   P3/P100 95 99.1 98.1 99.7 95–97 95 50–98  Filtering full facepiece (unpowered) 16 99.7 98.9 100.2 — — —   P2 2 99.5 98.8 99.9 93.3 90 35–95   P3 14 99.8 99.6 99.9 99.9 95 50–98  Powered air purifying respirator (full, TM3) 6 100.0 98.7 100.6 97.5–99.9 97 50–99.9  Supplied air respirator or airline respirator (hood, continuous flow/positive pressure) 12 100.0 99.2 100.4 96–99.9 95 85–99 Gloves  Butyl 5 93.5 75.4 98.8 All glove types: All glove types: 90 (l) 85 (l) 50–97  Latex 9 98.5 94.8 100.0 95 (s) 90 (s) 70–99  Neoprene 5 93.3 75.0 98.7  Nitrileg 15 97.0 92.9 99.0   Thick 5 97.2 79.6 100.3   Thin 5 88.4 29.9 98.8  Polyvinyl chloride 5 93.9 76.8 98.9  Vinyl 2 100.0 94.0 100.9 Clothing**  Protective coveralls and disposable coveralls   Cat III, type 3 only 10 99.6 98.2 100.3 — See below See below   Cat III, other (combined types 3–6 or not specified) 76 96.7 95.5 97.6 — See below See below    →Type 1: Gas-tight — 98 (s/l) 97–99.9    →Type 2: Non-gas tight — 97 (s/l) 95–99.9    →Type 3: Liquid (jet) tight 95 95 (s/l) 90–99    →Type 4: Spray tight 95 90 (s/l) 70–97    →Type 5: Particle/fibre dust protection 75 75 (s/l) 50–95    →Type 6: Limited spray tight/small splashes 80–90 85 (s/l) 50–95  Other non-specified work clothing and work wear   Woven (e.g. cotton lab coats) 71 49.6 34.6 61.2 — 40 (s); 25 (l) See Table S3   Non-woven (e.g. polypropylene) 57 81.5 75.1 86.4 — 70 (s); 50 (l) See Table S3   Coated (e.g. polyamide) 4 100.0 98.1 100.7 — 95 (s/l) 70–100 Ventilated/Overpressure suits 10 100.0 99.0 100.5 — 99.9 (s/l) 99–100 RMM N Estimated average nano-specific efficacy (%) 95% CI (%) Reference valueb conventional substances Proposed average nano-specific efficacy (%)c Lower Upper Value Min–max Engineering controls  Containmentd* 21 99.5 98.4 100.1 — — —   High level 14 99.7 98.9 100.2 99.9 99.9 90–99.99   Medium level 7 98.0 95.4 99.4 99 98 80–99  LEV – capturing hoods** 46 95.2 92.7 96.9 — — —   Fixed hood 29 97.5 94.2 99.2 90 90 70–98   Movable hood 17 90.0 77.0 96.0 50 70 30–98  LEV – enclosing hoods 6 76.9 35.7 92.1 — — —   Fume cupboard 5 77.3 14.2 94.5 99 90 50–99  Glove boxes (low specificatione) 4 93.4 75.4 98.8 99.9 98 95–99.95  Suppression techniques (at point of release) 3 94.6 74.7 99.4 90 90 70–90  Unidirectional room airflow systems (no screens, dust) 3 90.1 56.4 98.3 80 80 50–95 RPE  Unspecified nuisance mask 24 21.2 −19.9 48.4 — 10 0–60  Filtering half mask (disposable)** 240 97.2 96.7 97.7 — — —   FFP1 4 93.4 (83.7f) 72.3 (63.0f) 99.0 (93.1f) 75 65 30–75   FFP2/N95 147 95.0 (84.7f) 91.5 (79.3f) 97.2 (88.7f) 90–92 80 40–97   FFP3/N99/P100 89 98.1 (92.5f) 96.2 (89.3f) 99.3 (94.8f) 95–98 90 50–98  Filtering half mask (elastomeric face piece)** 117 97.7 97.0 98.2 — — —   P2/N95 22 98.1 96.1 99.3 90–92 90 35–95   P3/P100 95 99.1 98.1 99.7 95–97 95 50–98  Filtering full facepiece (unpowered) 16 99.7 98.9 100.2 — — —   P2 2 99.5 98.8 99.9 93.3 90 35–95   P3 14 99.8 99.6 99.9 99.9 95 50–98  Powered air purifying respirator (full, TM3) 6 100.0 98.7 100.6 97.5–99.9 97 50–99.9  Supplied air respirator or airline respirator (hood, continuous flow/positive pressure) 12 100.0 99.2 100.4 96–99.9 95 85–99 Gloves  Butyl 5 93.5 75.4 98.8 All glove types: All glove types: 90 (l) 85 (l) 50–97  Latex 9 98.5 94.8 100.0 95 (s) 90 (s) 70–99  Neoprene 5 93.3 75.0 98.7  Nitrileg 15 97.0 92.9 99.0   Thick 5 97.2 79.6 100.3   Thin 5 88.4 29.9 98.8  Polyvinyl chloride 5 93.9 76.8 98.9  Vinyl 2 100.0 94.0 100.9 Clothing**  Protective coveralls and disposable coveralls   Cat III, type 3 only 10 99.6 98.2 100.3 — See below See below   Cat III, other (combined types 3–6 or not specified) 76 96.7 95.5 97.6 — See below See below    →Type 1: Gas-tight — 98 (s/l) 97–99.9    →Type 2: Non-gas tight — 97 (s/l) 95–99.9    →Type 3: Liquid (jet) tight 95 95 (s/l) 90–99    →Type 4: Spray tight 95 90 (s/l) 70–97    →Type 5: Particle/fibre dust protection 75 75 (s/l) 50–95    →Type 6: Limited spray tight/small splashes 80–90 85 (s/l) 50–95  Other non-specified work clothing and work wear   Woven (e.g. cotton lab coats) 71 49.6 34.6 61.2 — 40 (s); 25 (l) See Table S3   Non-woven (e.g. polypropylene) 57 81.5 75.1 86.4 — 70 (s); 50 (l) See Table S3   Coated (e.g. polyamide) 4 100.0 98.1 100.7 — 95 (s/l) 70–100 Ventilated/Overpressure suits 10 100.0 99.0 100.5 — 99.9 (s/l) 99–100 Corrections for containment–particle size; capturing hoods–study design; filtering half mask (disposable)–metric, particle size, fit/leak, airflow/breathing rate; filtering full facepiece (unpowered) – measurement type; filtering half mask (elastomeric face piece) – measurement type; clothing: measurement type, particle size, metric. s/l, solids; liquids (incl. solids-in-liquids). aRMM is shown only where data were available for nanomaterials. Complete RMM list presented in Supplementary Tables S1–S3 at Annals of Occupational Hygiene online. bIndicative reference values for conventional substances based on various sources; EC: ART (Fransman et al., 2011); RPE: APF values: (i) EN 529 (2005) from Finland, Denmark, Italy, Sweden, UK; (ii) OSHA 29 CFR 1910.134 (2006); (iii) NIOSH Decision Logic (2004); (iv) ANSI Z88.2 (1992); and (v) BS 4275 (1997); Clothing/gloves: TNsG (2007), JRC (2010), and EFSA (2010). cSee complete list in Supplementary data at Annals of Occupational Hygiene online. dLow-level containment is physical containment not actively ventilated (e.g. non air-tight cover such as loose lid); medium containment is well sealed and not ventilated; high containment is both well sealed and ventilated (Fransman et al., 2013). eSingle chamber, simple access doors or pass box and an unsafe change glove, single high efficiency particulate air (HEPA)-filtered extract air and unsafe change filters (include manual cleaning). fCorrected for IP-treated or particle discharged data. gMissing contextual data (on thickness gloves). *P < 0.05, **P < 0.0001 indicate significant difference in sub-categories following correction with covariates in multiple regression models. View Large Table 3. Estimated average nano-specific efficacy values, 95% CI for individual RMMs, reference values for conventional substances, and proposed values for nanomaterialsa RMM N Estimated average nano-specific efficacy (%) 95% CI (%) Reference valueb conventional substances Proposed average nano-specific efficacy (%)c Lower Upper Value Min–max Engineering controls  Containmentd* 21 99.5 98.4 100.1 — — —   High level 14 99.7 98.9 100.2 99.9 99.9 90–99.99   Medium level 7 98.0 95.4 99.4 99 98 80–99  LEV – capturing hoods** 46 95.2 92.7 96.9 — — —   Fixed hood 29 97.5 94.2 99.2 90 90 70–98   Movable hood 17 90.0 77.0 96.0 50 70 30–98  LEV – enclosing hoods 6 76.9 35.7 92.1 — — —   Fume cupboard 5 77.3 14.2 94.5 99 90 50–99  Glove boxes (low specificatione) 4 93.4 75.4 98.8 99.9 98 95–99.95  Suppression techniques (at point of release) 3 94.6 74.7 99.4 90 90 70–90  Unidirectional room airflow systems (no screens, dust) 3 90.1 56.4 98.3 80 80 50–95 RPE  Unspecified nuisance mask 24 21.2 −19.9 48.4 — 10 0–60  Filtering half mask (disposable)** 240 97.2 96.7 97.7 — — —   FFP1 4 93.4 (83.7f) 72.3 (63.0f) 99.0 (93.1f) 75 65 30–75   FFP2/N95 147 95.0 (84.7f) 91.5 (79.3f) 97.2 (88.7f) 90–92 80 40–97   FFP3/N99/P100 89 98.1 (92.5f) 96.2 (89.3f) 99.3 (94.8f) 95–98 90 50–98  Filtering half mask (elastomeric face piece)** 117 97.7 97.0 98.2 — — —   P2/N95 22 98.1 96.1 99.3 90–92 90 35–95   P3/P100 95 99.1 98.1 99.7 95–97 95 50–98  Filtering full facepiece (unpowered) 16 99.7 98.9 100.2 — — —   P2 2 99.5 98.8 99.9 93.3 90 35–95   P3 14 99.8 99.6 99.9 99.9 95 50–98  Powered air purifying respirator (full, TM3) 6 100.0 98.7 100.6 97.5–99.9 97 50–99.9  Supplied air respirator or airline respirator (hood, continuous flow/positive pressure) 12 100.0 99.2 100.4 96–99.9 95 85–99 Gloves  Butyl 5 93.5 75.4 98.8 All glove types: All glove types: 90 (l) 85 (l) 50–97  Latex 9 98.5 94.8 100.0 95 (s) 90 (s) 70–99  Neoprene 5 93.3 75.0 98.7  Nitrileg 15 97.0 92.9 99.0   Thick 5 97.2 79.6 100.3   Thin 5 88.4 29.9 98.8  Polyvinyl chloride 5 93.9 76.8 98.9  Vinyl 2 100.0 94.0 100.9 Clothing**  Protective coveralls and disposable coveralls   Cat III, type 3 only 10 99.6 98.2 100.3 — See below See below   Cat III, other (combined types 3–6 or not specified) 76 96.7 95.5 97.6 — See below See below    →Type 1: Gas-tight — 98 (s/l) 97–99.9    →Type 2: Non-gas tight — 97 (s/l) 95–99.9    →Type 3: Liquid (jet) tight 95 95 (s/l) 90–99    →Type 4: Spray tight 95 90 (s/l) 70–97    →Type 5: Particle/fibre dust protection 75 75 (s/l) 50–95    →Type 6: Limited spray tight/small splashes 80–90 85 (s/l) 50–95  Other non-specified work clothing and work wear   Woven (e.g. cotton lab coats) 71 49.6 34.6 61.2 — 40 (s); 25 (l) See Table S3   Non-woven (e.g. polypropylene) 57 81.5 75.1 86.4 — 70 (s); 50 (l) See Table S3   Coated (e.g. polyamide) 4 100.0 98.1 100.7 — 95 (s/l) 70–100 Ventilated/Overpressure suits 10 100.0 99.0 100.5 — 99.9 (s/l) 99–100 RMM N Estimated average nano-specific efficacy (%) 95% CI (%) Reference valueb conventional substances Proposed average nano-specific efficacy (%)c Lower Upper Value Min–max Engineering controls  Containmentd* 21 99.5 98.4 100.1 — — —   High level 14 99.7 98.9 100.2 99.9 99.9 90–99.99   Medium level 7 98.0 95.4 99.4 99 98 80–99  LEV – capturing hoods** 46 95.2 92.7 96.9 — — —   Fixed hood 29 97.5 94.2 99.2 90 90 70–98   Movable hood 17 90.0 77.0 96.0 50 70 30–98  LEV – enclosing hoods 6 76.9 35.7 92.1 — — —   Fume cupboard 5 77.3 14.2 94.5 99 90 50–99  Glove boxes (low specificatione) 4 93.4 75.4 98.8 99.9 98 95–99.95  Suppression techniques (at point of release) 3 94.6 74.7 99.4 90 90 70–90  Unidirectional room airflow systems (no screens, dust) 3 90.1 56.4 98.3 80 80 50–95 RPE  Unspecified nuisance mask 24 21.2 −19.9 48.4 — 10 0–60  Filtering half mask (disposable)** 240 97.2 96.7 97.7 — — —   FFP1 4 93.4 (83.7f) 72.3 (63.0f) 99.0 (93.1f) 75 65 30–75   FFP2/N95 147 95.0 (84.7f) 91.5 (79.3f) 97.2 (88.7f) 90–92 80 40–97   FFP3/N99/P100 89 98.1 (92.5f) 96.2 (89.3f) 99.3 (94.8f) 95–98 90 50–98  Filtering half mask (elastomeric face piece)** 117 97.7 97.0 98.2 — — —   P2/N95 22 98.1 96.1 99.3 90–92 90 35–95   P3/P100 95 99.1 98.1 99.7 95–97 95 50–98  Filtering full facepiece (unpowered) 16 99.7 98.9 100.2 — — —   P2 2 99.5 98.8 99.9 93.3 90 35–95   P3 14 99.8 99.6 99.9 99.9 95 50–98  Powered air purifying respirator (full, TM3) 6 100.0 98.7 100.6 97.5–99.9 97 50–99.9  Supplied air respirator or airline respirator (hood, continuous flow/positive pressure) 12 100.0 99.2 100.4 96–99.9 95 85–99 Gloves  Butyl 5 93.5 75.4 98.8 All glove types: All glove types: 90 (l) 85 (l) 50–97  Latex 9 98.5 94.8 100.0 95 (s) 90 (s) 70–99  Neoprene 5 93.3 75.0 98.7  Nitrileg 15 97.0 92.9 99.0   Thick 5 97.2 79.6 100.3   Thin 5 88.4 29.9 98.8  Polyvinyl chloride 5 93.9 76.8 98.9  Vinyl 2 100.0 94.0 100.9 Clothing**  Protective coveralls and disposable coveralls   Cat III, type 3 only 10 99.6 98.2 100.3 — See below See below   Cat III, other (combined types 3–6 or not specified) 76 96.7 95.5 97.6 — See below See below    →Type 1: Gas-tight — 98 (s/l) 97–99.9    →Type 2: Non-gas tight — 97 (s/l) 95–99.9    →Type 3: Liquid (jet) tight 95 95 (s/l) 90–99    →Type 4: Spray tight 95 90 (s/l) 70–97    →Type 5: Particle/fibre dust protection 75 75 (s/l) 50–95    →Type 6: Limited spray tight/small splashes 80–90 85 (s/l) 50–95  Other non-specified work clothing and work wear   Woven (e.g. cotton lab coats) 71 49.6 34.6 61.2 — 40 (s); 25 (l) See Table S3   Non-woven (e.g. polypropylene) 57 81.5 75.1 86.4 — 70 (s); 50 (l) See Table S3   Coated (e.g. polyamide) 4 100.0 98.1 100.7 — 95 (s/l) 70–100 Ventilated/Overpressure suits 10 100.0 99.0 100.5 — 99.9 (s/l) 99–100 Corrections for containment–particle size; capturing hoods–study design; filtering half mask (disposable)–metric, particle size, fit/leak, airflow/breathing rate; filtering full facepiece (unpowered) – measurement type; filtering half mask (elastomeric face piece) – measurement type; clothing: measurement type, particle size, metric. s/l, solids; liquids (incl. solids-in-liquids). aRMM is shown only where data were available for nanomaterials. Complete RMM list presented in Supplementary Tables S1–S3 at Annals of Occupational Hygiene online. bIndicative reference values for conventional substances based on various sources; EC: ART (Fransman et al., 2011); RPE: APF values: (i) EN 529 (2005) from Finland, Denmark, Italy, Sweden, UK; (ii) OSHA 29 CFR 1910.134 (2006); (iii) NIOSH Decision Logic (2004); (iv) ANSI Z88.2 (1992); and (v) BS 4275 (1997); Clothing/gloves: TNsG (2007), JRC (2010), and EFSA (2010). cSee complete list in Supplementary data at Annals of Occupational Hygiene online. dLow-level containment is physical containment not actively ventilated (e.g. non air-tight cover such as loose lid); medium containment is well sealed and not ventilated; high containment is both well sealed and ventilated (Fransman et al., 2013). eSingle chamber, simple access doors or pass box and an unsafe change glove, single high efficiency particulate air (HEPA)-filtered extract air and unsafe change filters (include manual cleaning). fCorrected for IP-treated or particle discharged data. gMissing contextual data (on thickness gloves). *P < 0.05, **P < 0.0001 indicate significant difference in sub-categories following correction with covariates in multiple regression models. View Large Based on the analyses, efficacy values are proposed for each RMM category (Table 3) and described in more detail in the next section for each of the RMM types. To facilitate the evaluation process, box plots of uncorrected efficacy values for each RMM type are presented in Figs 1–4. The whiskers represent the 10th, 25th, 50th (median), 75th, and 90th percentile of the distribution. Nano-specific data on RMM effectiveness could not be found for all RMM categories, which meant that conventional efficacy values had to be adopted as an interim solution. A complete list of RMM categories and their proposed efficacy values are presented in Supplementary Tables S1–S3 at Annals of Occupational Hygiene online. Figure 1. View largeDownload slide Box plots of uncorrected nano-specific efficacy values for engineering controls. The dotted line (---) represents the arithmetic mean; the box represents the range from the 25th to the 75th percentile with the horizontal line showing the 50th percentile (median); and the whiskers show the 10th and 90th percentiles. Outliers are indicated by symbol ○. Figure 1. View largeDownload slide Box plots of uncorrected nano-specific efficacy values for engineering controls. The dotted line (---) represents the arithmetic mean; the box represents the range from the 25th to the 75th percentile with the horizontal line showing the 50th percentile (median); and the whiskers show the 10th and 90th percentiles. Outliers are indicated by symbol ○. Figure 2. View largeDownload slide Box plots of uncorrected nano-specific efficacy values for RPE. The dotted line (---) represents the arithmetic mean; the box represents the range from the 25th to the 75th percentile with the horizontal line showing the 50th percentile (median); and the whiskers show the 10th and 90th percentiles. Outliers are indicated by symbol ○. Figure 2. View largeDownload slide Box plots of uncorrected nano-specific efficacy values for RPE. The dotted line (---) represents the arithmetic mean; the box represents the range from the 25th to the 75th percentile with the horizontal line showing the 50th percentile (median); and the whiskers show the 10th and 90th percentiles. Outliers are indicated by symbol ○. Figure 3. View largeDownload slide Box plots of uncorrected nano-specific efficacy values for protective gloves. The dotted line (---) represents the arithmetic mean; the box represents the range from the 25th to the 75th percentile with the horizontal line showing the 50th percentile (median); and the whiskers show the 10th and 90th percentiles. Outliers are indicated by symbol ○. Figure 3. View largeDownload slide Box plots of uncorrected nano-specific efficacy values for protective gloves. The dotted line (---) represents the arithmetic mean; the box represents the range from the 25th to the 75th percentile with the horizontal line showing the 50th percentile (median); and the whiskers show the 10th and 90th percentiles. Outliers are indicated by symbol ○. Figure 4. View largeDownload slide Box plots of uncorrected nano-specific efficacy values for protective clothing. The dotted line (---) represents the arithmetic mean; the box represents the range from the 25th to the 75th percentile with the horizontal line showing the 50th percentile (median); and the whiskers show the 10th and 90th percentiles. Outliers are indicated by symbol ○. Figure 4. View largeDownload slide Box plots of uncorrected nano-specific efficacy values for protective clothing. The dotted line (---) represents the arithmetic mean; the box represents the range from the 25th to the 75th percentile with the horizontal line showing the 50th percentile (median); and the whiskers show the 10th and 90th percentiles. Outliers are indicated by symbol ○. Proposal for nano-specific RMM efficacy values Engineering controls Nano-specific data (Table 3 and Fig. 1) were compared with efficacy values for conventional substances applied in ART (Fransman et al., 2011) and with an updated dataset retrieved from ECEL (Fransman et al., 2008). Based on this comparative evaluation, it was decided to reconsider the effectiveness of containment, enclosing hoods (fume cupboards), movable hoods, and glove boxes as proposed by ART. For containment, the nano-specific data indicate a broad range and variation in the effectiveness of containment (50–99.999%). It is reasonable to assume that nano-sized particles and their agglomerates (up to 200 nm) are expected to more easily diffuse and escape from different types of containment, although this decreased efficacy appears to be restricted to particle size distributions (PSD) between ~50 and ~150 nm (Tsai et al., 2008; Tsai et al., 2012). Nevertheless, a slightly more conservative effectiveness value for medium level containment (as defined in Fransman et al., 2011) is proposed, i.e. 98% instead of 99%. Considering the large variability in efficacy values found for fume cupboards when using nanomaterials (e.g. Tsai et al, 2009) and conventional substances (Ahn et al., 2008), a more conservative value is proposed of 90% (compared with ART, 99%). Data on the effectiveness of glove bags and glove boxes are sparse and inconclusive. In the absence of useful data and for precautionary measures, the effectiveness values as proposed by ART for medium specification glove boxes (99.9%) is adapted and replaced with a less effective value, i.e. 98%. Respiratory protective equipment Assigned protection factors (APFs) intended for conventional substances were applied to compare with nano-specific data (Table 3 and Fig. 2). APF values available from various sources (e.g. ANSI, 1992; BSI, 1997; OSHA, 2006) were aggregated for each RPE category to obtain an indicative (upper and lower) value. In the absence of verified workplace protection factors (WPFs), APF values can be indicative as they take into account a variety of factors including the total inward leakage (TIL) from penetration directly through the filter, but also (simulated) leakages around the face/seal interface region (Shaffer and Rengasamy, 2009). For example, considering the variation in effectiveness found for FFP2/N95 masks [95%; confidence interval (CI): 91.5–97.2], a value of 80% is proposed compared with the APF value of 90%. This value was also assigned because if corrected for worst case conditions such as IP-treated or particle discharging, a value of 84.7% (CI: 79.3–88.7) was estimated. A distinct difference in protection for P2 and P3 filters may not be as logical as commonly expected. A recent study by Lee et al. (2016) found that FFP3 respirators performed the same or even worse than the FFP2 respirators with different fit factors – possibly due to a greater pressure drop and packing density of FFP3 compared with FFP2 respirators. Similarly, the difference in effectiveness between half and full masks are arguable, as a limited number of experimental studies have shown that half masks sometimes perform better than full masks, although this effect could be ascribed to a better fit to different face geometries (Lee et al., 2016). In general, slightly more conservative values are proposed for different types of RPE assuming that human factors such as work rate, fit testing, and maintenance are not accounted for. SPE: gloves and clothing Efficiency values of SPE for nanomaterials (Table 3 and Figs 3 and 4) were compared with indicative protection values applied for risk assessment purposes of biocides and pesticides (TNsG, 2007; EFSA, 2010; JRC, 2010). Similar to RPE, it is assumed that SPE materials are selected according to a recognized guideline, which is based on standard ASTM or EN test methods. For gloves, the thickness of nitrile gloves appears to significantly affect permeation of nanoparticles (Vinches et al., 2014; although the type of nanoparticle and their concentration play a crucial role. As a precautionary measure, it is proposed that only thick gloves (e.g. 0.11–0.15 mm in the case of nitrile) are recommended when working with nanomaterials. To account for the possibility of nanoparticles to more easily permeate through gloves, a slightly more conservative value is proposed for all protective gloves (85% for liquids and 90% for solids) compared with the mentioned guidelines (90% liquids, 95% solids). These values are in line with the 10th percentile of PFs against liquids as proposed by Roff (2015) for thin ‘splash-resistant single-use’ gloves, i.e. 7–10, or ~86–90%. More research is needed to verify the variability in protection levels, in particular inconsistencies found in the effectiveness of butyl rubber gloves (Vinches et al., 2014). Another research topic should also include the testing of deformations and damage of gloves exposed to colloidal solutions, which may result in an increased penetration of nanoparticles (Dolez et al., 2013). Most of the available nano-specific clothing data (category III) either did not specify the type (1–6) or only recorded subsets of multiple types of clothing (Table 3). Nevertheless, high protection values were found for type 3 protective clothing (99.6; CI: 98.2–100.3) based on aerosolized solids. With the limited data available, protective clothing are informed by default values intended for risk assessment purposes. For example, protection levels of 75% and 85% are proposed for types 5 and 6, respectively, relevant for both dusts and mists/small splashes. For types 3 and 4, the clothing type and clothing material are designed to provide protection against liquids (jets, splashes) and sprays (mists), respectively. It is recommended that protective clothing (e.g. Tyvek) intended for protection against nano-solids consist of non-woven high-density polyethylene or similar textiles (Golanski et al., 2010), while fabrics coated with for example polyamide and polyurethane (Boutry and Damlencourt, 2014) are advised for liquid or colloidal dispersions. Considering the analysis of non-specific work clothing, woven fabrics (e.g. woven lab coats, clean room wear) indicated a protection value of 49.6% (CI: 34.6–61.2), from which efficacy values of 40% and 25% were proposed for solids and liquids, respectively. A lower efficacy for liquids is proposed as prolonged exposure to aerosolized liquids has shown to enhance the penetration of nanoparticles through protective clothing (Park et al., 2011). In comparison, very high protection levels were measured for ventilated/overpressure suits in unpublished experimental studies conducted as part of the GUIDEnano project. Discussion Although it is well known that particle size can affect the capture efficiency of ventilation systems and penetration through respiratory filter media and protective clothing, no effectiveness values are available for different types of RMM for nanomaterials. This review is a first attempt to propose quantitative values on the effectiveness of RMM in the workplace – specifically for nanomaterials and for modelling purposes. In the case of engineering controls, studies have shown that localized ventilation systems are effective for particle sizes between 200 and 300 nm due to their minimal diffusion and small inertia that results in increased capturing efficiencies (Schulte et al., 2008; Schneider et al., 2011). When particle motion is dominated by diffusion for particle diameters smaller than 200 nm, and suction is sufficiently high, the particles will follow the streamlines into the ventilation device with only minor random zig-zag deviation, resulting in high capture efficiencies. When considering the outcome of this nano-specific data analysis, localized ventilation controls appeared to be in the same order of magnitude to that of conventional substances – although based on limited data. In an attempt to be cautious in deriving efficacy values based on the limited data, and considering the findings of various field studies (e.g. Tsai et al., 2008; Tsai et al., 2012), it was decided to reconsider the effectiveness of containment, fume cupboards, and glove boxes as proposed by ART (intended for conventional substances). For RPE, it is well established that very small nanoparticle sizes ranging between 4 and 20 nm (e.g. silver) are captured very efficiently by respirator filter media (Rengasamy et al., 2008). Theoretically, a combination of both diffusion and interception mechanisms results in a minimal efficiency or maximum penetration at a given particle size, typically between 100 and 500 nm and normally peaking at 300 nm. Filtering face respirators typically provide the lowest PFs for particles between 80 and 200 nm (Rengasamy et al., 2007). Worst case conditions in the workplace such as the removal of electrostatic charges on filter media can result in much less effective protection (Rengasamy et al., 2009), in addition to factors such as respirator fit, reduced leakages (Reponen et al., 2011), breathing rates, and airflow (Balazy et al., 2006; He et al., 2014). By considering these factors in the data analysis and comparing the outcomes with APF values intended for conventional substances, similar protection values (as APF values) were proposed for the respective RPE categories. An exception was the introduction of more conservative values for RPE categories such as disposable filtering half masks (e.g. FFP2, FFP3) – more in line with conservative values applied for the modelling of nanomaterials (Marquart et al., 2008). Testing of permeation, penetration, and degradation associated with the potential migration of nano-sized particles through different SPE such as coveralls and gloves is complex and evidence is scarce, but some studies indicate that the type of nanoparticle (e.g. TiO2), its physical state (solid, solid-in-liquid), garment thickness, and textile type and composition (woven, non-woven, coated) are important factors (Boutry and Damlencourt, 2014; Vinches et al., 2014). With limited nano-specific data available for SPE, in most instances we had to resort to existing default values applied for regulatory purposes. Overall, data indicate that nanoparticles can penetrate through textiles and materials, with high-density materials (e.g. polyethylene, polyamide, polyurethane) performing better as a protection barrier than woven materials such as cotton. Current data suggest that the thickness of gloves could be an important factor that affects protection from nanomaterials. Manufactured nanoparticles that are emitted during synthesis are known to coagulate rapidly during the emission from the source and transport to the receptor (Schneider and Jensen, 2009). Considering the data evaluated in this review, particles at the source typically range between 10 and 400 nm. Since nanoparticles agglomerate and aggregate and increase in size over time, the effectiveness of different types of engineering controls will vary accordingly. An option could be to differentiate between effectiveness of ventilation systems based on the expected PSD. However, since an estimation of the coagulation rate is complex (Schneider and Jensen, 2009) and time and concentration dependent, and with the current scarcity of efficacy data, a distinction between PSDs of different expected nano forms (e.g. pristine versus fragmented, or synthesis versus end of life) and their capture efficiencies is not considered possible (or useful) at this stage. By aggregating data with efficacy values or PFs obtained from a wide range of test methods will always be controversial. Data collected in this study did not allow for a correction of autocorrelation associated with sequential measurements of online time series data, which can be accounted for with statistical models (Klein Entink et al., 2011). As a result, the estimated efficacy was restricted to the use of an (average) value of a given time series (for a given particle size range, or activity versus background) which, admittedly, may not always be representative nor adequately reflect the actual effectiveness of a RMM under investigation. Univariate and bivariate analyses was performed to derive corrected estimates of efficacy for RMM categories. Although various covariates were identified and included, the variability in RMM efficacy values could not always be adequately addressed due to the limited number of studies and data points. Also, covariables included in the analyses (e.g. particle size, metric) may only explain some of the variance of RMM effectiveness. That said, a meaningful collation of effectiveness data for a broad spectrum of RMM relevant for nanomaterials will remain limited unless high-quality (raw) data are being shared or made available as open sources. Although we had access to raw data of a handful of experimental studies, the majority of studies did not report detailed information (e.g. min/max values) on the data collected. Another issue of concern is the large number of experimental studies, with some exceptions of (simulated) tests of human subjects or workplace measurements. Although some workplace studies have confirmed that workers receive the expected levels of protection when compared with experimental studies (Balazy et al., 2006), experimental data should be interpreted with caution because of the different test methods and test conditions applied. In order to propose efficacy values for the modelling of nanomaterials, one important consideration was to determine the level of conservatism. This would be of particular importance if a model would apply a single value (without a lower and upper value for probabilistic modelling). Unfortunately, datasets were generally too small or not representative enough to extract, for example, a meaningful value based on percentiles (e.g. 5th percentile for RPE). Instead, we decided to compare the outcome of the nano-specific data analysis (average values) with existing reference values for conventional substances (Table 3) and to adapt these values in case nanomaterials may pose an increased risk. In this regard, the sources applied as reference values may not always be representative of working conditions (use, maintenance) for such a comparative evaluation. For example in the case of RPE, ranges in APF values were adopted as indicative values that are widely applied by different institutions and countries. Such values are often based on a 90% CI of the 5th percentile and provide a 90% certainty that in 95% of the cases adequate protection is provided under these test conditions. However, it is not uncommon to apply a safety factor on top of this value (from a factor 2 to 25), to accommodate variability in fit tests and other uncertainties (Marquart et al., 2008; Hutzel and Weed, 2010), resulting in significantly more conservative values. In the absence of data, APF values may be derived by using a safety factor (5–50) over and above the nominal protection factor. Such values may be excessively conservative, as indicated in a recent study by Schinkel et al. (2016) that tested the ‘simulated workplace protection factor’ (SWPF) of eight supplied air respirators, which revealed that the 5th percentile (with 90% CI) of PFs were often several orders of magnitude higher than the prescribed APF values. It is expected that the new RPE classification system and protection levels currently under development by ISO (and intended to harmonize APF values) will shed more light on baseline performances of RPE associated with human factors. For each RMM type, realistic assumptions about the technical specifications, conditions of use and maintenance of RMM are required. As proposed by ECHA (2012), RMM effectiveness could be best described by considering both typical default values (as used) and maximum achievable values (as built) of RMM. Ideally, these descriptors should be used to develop a viable approach to estimate efficiency values that represent and incorporate RMM during typical conditions of use in practice. Therefore, in event of sufficient data in the future, a clear and structured method to effectively disseminate information on the efficiency of RMM (e.g. time series of nano-specific data) and a suitable and robust data analysis is required to derive reliable efficiency values based on percentiles and/or confidence levels. Future research should, in addition, focus on the development of guidelines for the selection of suitable RMM for specific nanomaterials and processes based on scientific data. Recent research efforts (EU-funded nanoprojects) collated a significant amount of data on different RMM types and their related industry types, life cycles, nanomaterials, processes, and activities of concern (Oksel et al., 2016; Frijns et al., 2016). In summary, analysis of RMM effectiveness data for nanomaterials showed that the process of deriving quantitative efficacy values is hampered by scarcity of data and a small diversity in study types (mostly experimental). With the current data available, the proposed efficacy values are based on (i) evidence obtained from the nano-specific data analysis (presented in this review), (ii) existing evidence on RMM effectiveness of conventional substances, and (iii) expert judgement based on scientific evidence on the behaviour of nanomaterials and the potential contribution of maintenance and human factors. Future research should therefore focus on obtaining more good-quality data in the workplace and simulated or controlled conditions to more explicitly address the accuracy and uncertainty of proposed efficacy values. Supplementary data Supplementary data are available at Annals of Work Exposures and Health online. Funding This research was funded by the European Union Seventh Framework Programme (FP7/2007–2013) – GUIDEnano (Grant agreement n° 330 604387) and SUN (Grant agreement n° 604305). Acknowledgements We wish to thank all project members who were involved with contributing to this research as part of the Guidenano and SUN projects. Declaration for publication The authors declare no conflict of interest relating to the material presented in this review. Its contents, including any opinions and/or conclusions expressed, are solely those of the authors. References Ahn K , Woskie S , DiBerardinis L et al. ( 2008 ) A review of published quantitative experimental studies on factors affecting laboratory fume hood performance . J Occup Environ Hyg ; 5 : 735 – 53 . Google Scholar Crossref Search ADS PubMed American National Standards Institute . ( 1992 ) ANSI Z88.2:1992. American National Standard for respiratory protection . New York : American National Standards Institute . Balazy A , Toivola M , Reponen T et al. ( 2006 ) Manikin-based performance evaluation of N95 filtering-facepiece respirators challenged with nanoparticles . Ann Occup Hyg ; 50 : 259 – 69 . Google Scholar PubMed Boutry D , Damlencourt JF . ( 2014 ) Experimental assessment of the efficiency of current personal dermal protection equipment used in the construction sector. Scaffold Public Documents – SPD25 . Available at http://scaffold.eu-vri.eu/filehandler.ashx?file=14345 Accessed 15 August 2017 . Brouwer D , van Duuren-Stuurman B , Berges M et al. ( 2009 ) From workplace air measurement results toward estimates of exposure? Development of a strategy to assess exposure to manufactured nano-objects . J Nanopart Res ; 11 : 1867 – 81 . Google Scholar Crossref Search ADS Brouwer DH , Spaan S , Roff M et al. ( 2016 ) Occupational dermal exposure to nanoparticles and nano-enabled products: part 2, exploration of exposure processes and methods of assessment . Int J Hyg Environ Health ; 219 : 503 – 12 . Google Scholar Crossref Search ADS PubMed British Standards Institution . ( 1997 ) BS 4275: 1997. Guide to implementing an effective respiratory protective device programme . London : British Standards Institution . CEFIC. Conseil Européen des Fédérations de l’Industrie Chimique . ( 2016 ). CEFIC: risk management measure library . Available at http://www.cefic.org/Documents/IndustrySupport/REACH-Implementation/Guidance-and-Tools/Risk%20Management%20Measures%20(RMM).xls. Accessed 15 August 2017 . Conti JA , Killpack K , Gerritzen G et al. ( 2008 ) Health and safety practices in the nanomaterials workplace: results from an international survey . Environ Sci Technol ; 42 : 3155 – 62 . Google Scholar Crossref Search ADS PubMed Dolez P, Vinches L, Perron G, Vu-Khanh T, Plamondon P, L’Espérance G, Wilkinson K, Cloutier Y, Dion C, Truchon G. (2013) Development of a Method of Measuring Nanoparticle Penetration through Protective Glove Materials under Conditions Simulating Workplace Use. Mechanical and Physical Risk Prevention - Studies and Research Projects. Report R-785. IRSST. Montréal, Québec. ISBN: 978-2-89631-680-9. Available at http://www.irsst.qc.ca/media/documents/PubIRSST/R-785.pdf?v=2018-05-25. Accessed 18 June 2018. ECHA. European Chemicals Agency . ( 2012 ) ECHA: guidance on information requirements and chemical safety assessment. Chapter R.13: risk management measures and operational conditions (Version 1.2) . Helsinki, Finland: ECHA Communication Unit . Available at https://echa.europa.eu/documents/10162/13632/information_requirements_r13_en.pdf. Accessed 15 August 2017 . EFSA. European Food Safety Authority . ( 2010 ) EFSA: scientific opinion on preparation of a guidance document on pesticide exposure assessment for workers, operators, bystanders and residents . EFSA J ; 8 : 1501 . Available at http://www.efsa.europa.eu/en/efsajournal/pub/1501. Accessed 15 August 2017 . Fransman W , Cherrie J , van Tongeren M et al. ( 2013 ) Development of a mechanistic model for the advanced reach tool (ART) . TNO report V9009. Available at www.advancedreachtool.com. Accessed 15 August 2017 . Fransman W , Schinkel J , Meijster T et al. ( 2008 ) Development and evaluation of an exposure control efficacy library (ECEL) . Ann Occup Hyg ; 52 : 567 – 75 . Google Scholar PubMed Fransman W , Van Tongeren M , Cherrie JW et al. ( 2011 ) Advanced Reach Tool (ART): development of the mechanistic model . Ann Occup Hyg ; 55 : 957 – 79 . Google Scholar PubMed Frijns E . ( 2016 ) Guidance on recommended measures for mitigating the risk posed by engineered nanomaterials. LIFE NanoRISK: best practices effectiveness, prevention and protection measures for control of risk posed by engineered nanomaterials. LIFE12 ENV/ ES/000178 . Available at http://www.lifenanorisk.eu/phocadownload/MULTIMEDIA_GUIDANCE_PDF_VERSION.pdf. Accessed 15 August 2017 . Golanski L , Guiot A , Rouillon F et al. ( 2010 ) Experimental evaluation of individual protection devices against different types of nanoaerosols: graphite, TiO2, and Pt . J Nanopart Res ; 12 : 83 – 9 . Google Scholar Crossref Search ADS GUIDEnano . (2016) European Union Seventh Framework Programme (FP7/2007–2013) under grant agreement n° 330 604387 . He X , Grinshpun SA , Reponen T et al. ( 2014 ) Effects of breathing frequency and flow rate on the total inward leakage of an elastomeric half-mask donned on an advanced manikin headform . Ann Occup Hyg ; 58 : 182 – 94 . Google Scholar PubMed Health and Safety Executive (HSE). (2017). Controlling airborne contaminants at work - A guide to local exhaust ventilation (LEV). Report HSG258. Third edition. Norwich, UK: TSO. SBN 978 0 7176 6415 3. Available at www.hse.gov.uk/pUbns/priced/hsg258.pdf. Accessed 19 June 2018. Hinds WC . ( 1999 ) Aerosol technology: properties, behavior, and measurement of airborne particles . New York, USA : John Wiley & Sons . 2 nd edn. ISBN: 978-0-471-19410-1. Hutzel RL , Weed J . ( 2010 ) Respiratory protection solutions (WeedRPS), LLC simulated workplace protection factor study using four clemco industries type-CE supplied-air respirators . Available at http://www.clemcoindustries.com/images/pdfs/Clemco_SWPF_Study.pdf. Accessed 15 August 2017 . ICON. International Council on Nanotechnology . ( 2006 ) ICON: a review of current practices in the nanotechnology industry. Phase 2 report: survey of current practices in the nanotechnology workplace . Available at http://cohesion.rice.edu/CentersAndInst/ICON/emplibrary/ICONNanotechSurveyFullReduced.pdf. Accessed 16 August 2016 . ISO. ISO /TS 12901-1 (2012): Nanotechnologies—occupational risk management applied to engineered nanomaterials. Part 1: Principles and approaches. ISO/TS 12901-2: Nanotechnologies—occupational risk management applied to engineered nanomaterials. Geneva, Switzerland: International Organization for Standardization. JRC. European Commission Joint Research Centre. JRC: Institute for Health and Consumer Protection Chemical Assessment and Testing . ( 2010 ) HEEG opinion - default protection factors for protective clothing and gloves . Joint Research Centre. Ispra, Italy . Available at https://echa.europa.eu/documents/10162/19680902/heeg_opinion_9_default_protection_factors_for_clothing_and_gloves_en.pdf. Accessed 16 August 2017 . Klein Entink RH , Fransman W , Brouwer DH . ( 2011 ) How to statistically analyze nano exposure measurement results: using an ARIMA time series approach . J Nanoparticle Res ; 13 : 6991 – 7004 . Google Scholar Crossref Search ADS Lee S-A , Hwang D-C , Li H-Y et al. ( 2016 ) Particle size-selective assessment of protection of European standard FFP respirators and surgical masks against particles-tested with human subjects . J Healthcare Eng ; 2016: 1 – 12 . LIFE NanoRisK . (2016) NanoRisk: partly funded by the European Commission LIFE+ with grant agreement LIFE12 ENV/ES/000178 . Available at URL: http://www.lifenanorisk.eu/. Accessed 19 June 2018. Marquart H , Heussen H , Le Feber M et al. ( 2008 ) ‘Stoffenmanager’, a web-based control banding tool using an exposure process model . Ann Occup Hyg ; 52 : 429 – 41 . Google Scholar PubMed Methner MM . ( 2008 ) Engineering case reports. Effectiveness of local exhaust ventilation (LEV) in controlling engineered nanomaterial emissions during reactor cleanout operations . J Occup Environ Hyg ; 5 : D63 – 9 . Google Scholar Crossref Search ADS PubMed NANoREG . (2017) A common European approach to the regulatory testing of manufactured nanomaterials. European Union Seventh Framework Programme (FP7/2007–2013) under grant agreement n° 310584 . Available at www.nanoreg.eu/. Accessed 19 June 2018. National Institute for Occupational Safety and Health . ( 2013 ). NIOSH: Centers for Disease Control and prevention (CDC). Current strategies for engineering controls in nanomaterial production and downstream handling processes. Publication No 2014–102 . Available at https://www.cdc.gov/niosh/docs/2014–102/pdfs/2014–102.pdf. Accessed 16 August 2017 . Occupational Safety and Health Administration . ( 2002 ) OSHA: occupational safety and health administration. Respiratory protection . Code of Federal Regulations Title 29, Part 1910 ; 134 : 434 – 40 . Occupational Safety and Health Administration . ( 2006 ) OSHA: occupational safety and health administration. Assigned protection factors: final rule . Fed Regist ; 71 : 50121 – 92 . Oksel C , Subramanian V , Semanzin E et al. ( 2016 ) Evaluation of existing control measures in reducing health and safety risks of engineered nanomaterials . Environ Sci ; Nano, 3:869–882 doi: https://doi.org/10.1039/C6EN00122J Organisation for Economic Co-operation and Development . ( 2009 ). OECD: comparison of guidance on selection of skin protective equipment and respirators for nanotechnology workplaces. Working party on manufactured nanomaterials. ENV/CHEM/NANO(2009)11/ADD3 . Available at http://www.oecd.org/officialdocuments/publicdisplaydocumentpdf/?cote=ENV/CHEM/NANO(2009)11/ADD3&docLanguage=En. Accessed 16 August 2017 . Organisation for Economic Co-operation and Development . ( 2016 ) OECD: emission scenario documents . Available at http://www.oecd.org/env/ehs/risk-assessment/emissionscenariodocuments.htm. Accessed 16 August 2017 . Park J , Kwak BK , Kim Y et al. ( 2011 ) Efficiency of protective dermal equipment against silver nanoparticles with water aerosol . J Nanopart Res ; 13 : 3043 – 9 . Google Scholar Crossref Search ADS Rengasamy S , Eimer BC , Shaffer RE . ( 2009 ) Comparison of nanoparticle filtration performance of NIOSH-approved and CE-marked particulate filtering facepiece respirators . Ann Occup Hyg ; 53 : 117 – 28 . Google Scholar Crossref Search ADS PubMed Rengasamy S , King WP , Eimer BC et al. ( 2008 ) Filtration performance of NIOSH-approved N95 and P100 filtering facepiece respirators against 4 to 30 nanometer-size nanoparticles . J Occup Environ Hyg ; 5 : 556 – 64 . Google Scholar Crossref Search ADS PubMed Rengasamy S , Verbofsky R , King WP et al. ( 2007 ) Nanoparticle penetration through NIOSH approved N95 filtering-facepiece respirators . J Int Soc Res Prot ; 24 : 49 – 59 . Reponen T , Lee SA , Grinshpun SA et al. ( 2011 ) Effect of fit testing on the protection offered by n95 filtering facepiece respirators against fine particles in a laboratory setting . Ann Occup Hyg ; 55 : 264 – 71 . Google Scholar Crossref Search ADS PubMed Roff M . ( 2015 ) The short-term protective effects of ‘Non-PPE’ gloves used by greenhouse workers . Ann Occup Hyg ; 59 : 1044 – 57 . Google Scholar Crossref Search ADS PubMed Scaffold . (2015) Innovative strategies, methods and tools for occupational risk management of manufactured nanomaterials (MNMs) in the construction industry . European Union Seventh Framework Programme (FP7/2007–2013) under grant agreement n° NMP4-SL-2012–280535 . Available at http://www.scaffold.eu-vri.eu/filehandler.ashx?file=13410. Accessed 19 June 2017. Schinkel J , den Boeft J , Tromp P et al. ( 2016 ) Onderzoek beschermingsfactoren bij adembeschermingsmiddelen gebruik in de asbestbranche. TNO, Report No R11443 . The Netherlands . (in Dutch). Available at http://xref.tno.nl/bibliotheek/sv-015068/TNO/Rapporten/2016/TNO-2016-R11443.pdf. Accessed 16 August 2017 . Schneider T , Brouwer DH , Koponen IK et al. ( 2011 ) Conceptual model for assessment of inhalation exposure to manufactured nanoparticles . J Expo Sci Environ Epidemiol ; 21 : 450 – 63 . Google Scholar Crossref Search ADS PubMed Schneider T , Jensen KA . ( 2009 ) Relevance of aerosol dynamics and dustiness for personal exposure to manufactured nanoparticles . J Nanopart Res ; 11 : 1637 – 50 . Google Scholar Crossref Search ADS Schubauer-Berigan MK , Dahm MM , Schulte PA et al. ( 2015 ) Characterizing adoption of precautionary risk management guidance for nanomaterials, an emerging occupational hazard . J Occup Environ Hyg ; 12 : 69 – 75 . Google Scholar Crossref Search ADS PubMed Schulte P , Geraci C , Zumwalde R et al. ( 2008 ) Occupational risk management of engineered nanoparticles . J Occup Environ Hyg ; 5 : 239 – 49 . Google Scholar Crossref Search ADS PubMed Shaffer RE , Rengasamy S . ( 2009 ) Respiratory protection against airborne nanoparticles: a review . J Nanopart Res ; 11 : 1661 – 72 . Google Scholar Crossref Search ADS SUN (2016) Sustainable Nanotechnologies project. European Union Seventh Framework Programme (FP7/2007–2013) under grant agreement n° 604305 . Available at http://www.sun-fp7.eu/. Accessed 19 June 2018. TNsG. Technical Notes for Guidance on Human Exposure to Biocidal Products . ( 2007 ) TNsG: user guidance 2007. Contract B4-3040/2000/291079/MAR/E2 for the European Commission, DG Environment . Joint Research Centre (JRC), European Commission . Available at https://echa.europa.eu/documents/10162/16960215/bpd_guid_tnsg-human-exposure-2007_en.pdf. Accessed 16 August 2017 . Tsai SJ , Ada E , Isaacs J et al. ( 2009 ) Airborne nanoparticle exposures associated with the manual handling of nanoalumina in fume hoods . J Nanopart Res ; 11 : 147 – 61 . Google Scholar Crossref Search ADS Tsai SJ , Ashter A , Ada E et al. ( 2008 ) Airborne nanoparticle release associated with the compounding of nanocomposites using nanoalumina as fillers . Aerosol Air Qual Res ; 8 : 160 – 77 . Google Scholar Crossref Search ADS Tsai CSJ , White D , Rodriguez H et al. ( 2012 ) Exposure assessment and engineering control strategies for airborne nanoparticles: an application to emissions from nanocomposite compounding processes . J Nanopart Res ; 14 : 989 . Google Scholar Crossref Search ADS Vinches L , Hallé S , Peyrot C et al. ( 2014 ) Which gloves are efficient to protect against titanium dioxide nanoparticles in work conditions ? Proceedings of the 5th International Conference on Nanotechnology: Fundamentals and Applications . Prague , Czech Republic , August 11–13. Paper No. 130:1–8. Available at https://avestia.com/ICNFA2014_Proceedings/papers/130.pdf. Accessed 19 June 2018. © The Author(s) 2018. Published by Oxford University Press on behalf of the British Occupational Hygiene Society. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model)
TI - A Review of Workplace Risk Management Measures for Nanomaterials to Mitigate Inhalation and Dermal Exposure
JF - Annals of Work Exposures and Health (formerly Annals Of Occupational Hygiene)
DO - 10.1093/annweh/wxy032
DA - 2018-10-15
UR - https://www.deepdyve.com/lp/oxford-university-press/a-review-of-workplace-risk-management-measures-for-nanomaterials-to-3eOv8azKXj
SP - 907
VL - 62
IS - 8
DP - DeepDyve
ER -