Since 1985, the Occupational Safety and Health Administration (OSHA) has required chemical manufacturers and importers to create Material Safety Data Sheets (MSDSs) on chemicals determined to be hazardous [OSHA Hazard Communication Standard (HazCom standard) (29 CFR 1910.1200)].¹ In 2012, OSHA issued a final rule to revise the HazCom standard to align with the United Nations Globally Harmonized System of Classification and Labeling of Chemicals (GHS).¹ In addition to new labeling requirements, the updated rule changed the name of a MSDS to Safety Data Sheet (SDS). The purpose of the SDS is to communicate information on the inherent health and safety hazards of a chemical, and precautions and protections during the handling and storage of hazardous chemicals. The format of the SDS contains 16 sections to address these hazards; however, only 12 of the 16 sections are legally mandatory and 4 sections (ecological information, disposal considerations, transport information, and regulatory information) deal with information beyond exposure hazards.

Engineered nanomaterials (ENMs) are one type of chemical that would require a SDS. The small size of ENMs, approximately 1 to 100 nm, can have novel chemical and physical properties. Development and commercialization of nanomaterial-based products and applications are occurring at a rapid rate, making it imperative to communicate sufficient information about the potential hazards from exposure to ENMs. The number of workers exposed to EMNs is not known, but market reports indicate that large and growing quantities of ENMs are being used in commerce; and workers are involved throughout the manufacture, formulation and use of these nanomaterial products.²

There is evidence that nanoscale materials tend to be more hazardous than the same material in a non-nanoscale form. Nanoparticles can be inhaled and deposited in all sections of the respiratory tract, including the gas-exchange (alveolar) region.³ Nanoparticles have also been observed to reach the interstitium and the blood, which may represent a path for the translocation of inhaled nanoparticles from the lung to secondary organs.^{4, 5, 6, 7, 8} Carbon nanotubes (CNTs) also appear to promote interstitial fibrosis, and specific multi-walled CNTs (MWCNTs) have been shown to promote lung cancer.^{8, 9} Inhalation of certain types of nanoparticles have also been linked to cardiovascular effects.^{10, 11} At present, there are no enforceable national or international occupational exposure limits (OELs) or standards specific to engineered nanomaterials, however, recommended exposure limits (RELs) have been published for two types of nanomaterials. NIOSH published a REL of 0.3 mg/m³ for ultrafine (nano) titanium dioxide (TiO₂) and a REL of 1 μg/m³ for CNTs and carbon nanofibers (CNF), measured as elemental carbon, for an 8-hour respirable-mass airborne concentration.^12,13

In 2012, NIOSH authors (Eastlake et al.) published a review of engineered nanomaterial MSDSs using four questions to determine quality, and concluded that only 17–33% of those published during 2007–2011 provided sufficient data for communicating the potential hazards to employers and workers.¹⁴ The range of results come from grouping MSDS by the publication years of 2007–2008 and 2010–2011. The purpose of this current evaluation is to determine if the revised 2012 OSHA HazCom standard has impacted the reliability of ENM safety and health information provided on nanomaterial SDSs.

A total of 67 unique SDSs for nanomaterials were obtained from 67 manufacturers and ranked using the Kimlisch et al. criteria. The results of the evaluation were 28.4% (19) reliable without restrictions (excellent), 35.8% (24) reliable with restrictions (good), and 35.8% (24) unreliable (Figure 1).

The lowest ranked categories were Section 11, Toxicological information, followed by Section 4, First aid measures and Section 5, Firefighting measures. The highest ranked sections were Section 9, Physical and chemicals properties, followed by Section 7, Handling and storage, and Section 3, Composition information.

There was only 1 SDS that scored a perfect 44 with thorough and complete information. There were 2 SDSs that scored 43, but both were missing toxicological information. While having toxicological information on the ENMs would be helpful to some end users, the updated HazCom Standard specifically states that “there is no requirement for testing chemicals”.¹ These scores of 43 represent that a SDS only missing toxicological data is still considered “excellent”.

The lowest score for a SDS was 31 and this nanomaterial solution SDS (deemed unreliable), contained very general statements such as follow “general industrial hygiene practice,” “respiratory protection not required,” and that there is “no data available”.

A subset of nano-containing construction materials (treated lumber, paints, coatings, lubricants and greases) was evaluated and compared to the total set since one of the databases used for gathering the SDSs was the CPWR database of nano containing construction materials. Of the 19 construction materials SDSs, 26.3% (5) were found to be reliable without restrictions (excellent), 47.4% (9) were reliable with restrictions (good), and 26.3% (5) were determined to be unreliable. Figure 2 shows data for 48 ENM SDSs as compared to 19 SDSs for nano containing construction materials.

Applying the Eastlake et al. ranking scheme to the 67 ENM SDSs collected for this evaluation resulted in 3% (2) as satisfactory, 17.9% (12) as being in need of improvement, and 79% (53) in need of significant improvement. It is noteworthy that out of the 79% in need of significant improvement, 25.4% (17) did not have data that answered any of the four posed questions.

Technical specification sheets are increasingly used to communicate physical and chemical properties specific to the material. Out of the 67 SDSs obtained, 56.7% (38) had technical specification sheets available. The technical specification sheets were a good source of particle size information and while only 10 SDS out of 67 contained information on size, an additional 16 technical specification sheets contained the information. This led to an increase in the percent of positive answers for the Eastlake et al. criterion 2 (specified that material is nanometer range, using numbers) to go from 14.9% (10) to 38.3% (26). No other information on the technical specification sheets provided information to the proposed ranking questions. The only difference noted was in the availability of size in the SDS in combination with the technical specification sheet versus that of the SDS alone. This indicates that it may be necessary to supplement the information from a SDS with information found on the technical specification sheet. A summary of the data obtained using the Eastlake et al. 2012 method on the 67 SDSs is shown in Table 3.

The SDSs were evaluated to determine if they included the NIOSH RELs for two types of ENMs. The NIOSH REL for CNT and CNF was included on 23.1% (3/13) of the SDSs for CNT and CNF containing materials. Of 24 SDSs for all carbon-based materials (such as graphene, fullerene, CNT, CNF), 29.2% (7/24) referenced the OSHA PEL for graphite. None of the three SDSs that indicated that they contained nano TiO₂ provided the NIOSH REL for ultrafine TiO₂, but the bulk size TiO₂ REL was listed on 66.7% (2/3) of the SDS.

The evaluation of a SDS can be subjective. To quantify the quality of information on the SDS, a ranking scheme was developed to evaluate Sections 1–11 and then create a summary score. This was done independently by the co-authors and then a consensus score created. For the 67 SDSs for ENMs and nano-containing construction products evaluated, 28.4% were determined to be of excellent quality and clearly communicated the potential hazards of engineered nanoparticles in a way that was informative and protective. There were 35.8% that were reliable with restrictions (good) which meant that they relied on generic protection statements (e.g. avoid inhalation, use adequate ventilation, use adequate protective clothing, handle in accordance with good industrial hygiene and safety practice) without any specifics or any attention to the material being nanoscale. This left 35.8% of SDSs that did a poor (unreliable) job of communicating potential hazards.

Other researchers (Lee et al. 2012, Safework Australia 2010, Nayer et al. 2015), have examined nanomaterial-related and other MSDSs for quality.^{16, 17, 18} They found that the majority of the MSDS did not include sufficient information on the safety of nanomaterials or on how to appropriately inform an occupational risk assessment.

Lee et al. evaluated 97 nanomaterial-related MSDS as either describing or not describing the information in each of 8 categories (identification of substance, hazard identification, composition information of ingredients, first-aid measures, firefighting measures, accidental release measures, handling and storage, and exposure control/personal protective equipment). Lee et al. found that the majority of the SDS did not include sufficient information on the safety of nanomaterials. Most products (85%) did not include any nanomaterial specific data, and 76% had no information on accidental release measures. Engineering controls were not mentioned in 36% of the SDS, and 65% did not recommend the use of PPE to avoid nanomaterial exposure.¹⁶ The results of this study were used to inform the International Standards Organization Technical Committee 229 (ISO TC 229) report, ISO/TR 13329:2012, Nanomaterials – Preparation of Material Safety Data Sheet (MSDS).¹⁹

Safe Work Australia evaluated 50 ENM MSDS authored prior to 2009 using a similar criteria to the one used here. They ranked the MSDS accuracy using a scale of 1 (inaccurate), 2 (partially accurate), and 3 (accurate). They determined that only 18% (9/50) MSDS provided reliable information to appropriately inform an occupational risk assessment. That left 82% as not providing reliable information.¹⁷

Nayar et al. evaluated the efficiency of 200 SDSs from 89 suppliers used in the aerospace industry (not nanomaterial specific) using the same Klimisch et al. categorical approach described herein, for Sections 2, 6, 7, 8, and 13 (hazard identification, accidental release measures, handling and storage, exposure controls/PPE and waste disposal). They grouped SDSs into pre and post 2010, yet found no statistical differences between the groups. Human health hazard information was good or acceptable on 58% of the SDSs, and safety hazards were good or acceptable on 50% of the SDSs. This left 42% of the SDS as lacking acceptable human health information and 50% lacking acceptable safety information.¹⁸

This evaluation of ENM SDS authored since 2012 (and collected during 2016–2017) revealed that the quality of information on many SDSs still cannot be relied upon to offer adequate information on the inherent health and safety hazards, including handling and storage of ENMs. It is concerning that according to our evaluation, many SDSs remain unreliable, thus making it difficult for end users to know if they can trust the information, or even how to interpret the limited information made available. It is also concerning that SDSs do not always provide technical information (such as particle size) even though this information can often be found in separate technical specification sheets. One positive finding was that several SDSs advised using engineering controls and/or PPE.

This evaluation of nanomaterial safety data sheets revealed that the quality of SDS information on many of the ENM SDSs still cannot be relied upon to offer adequate information on the potential health and safety hazards, including handling and storage of engineered nanomaterials. Chemical manufacturers should consult the recommendations issued in the International Standards Organization Technical Committee 29 (ISO TC 229) report ISO/TR 13329:2012, Nanomaterials – Preparation of Material Safety Data Sheet (MSDS) when developing SDSs.¹⁹ An additional resource for chemical manufacturers is the SDS and Label Authoring Registry program, which recognizes chemical hazard communication and environmental health professionals who specialize in authoring SDS and labels. This registry program was developed through a partnership between the American Industrial Hygiene Association (AIHA) and the Society for Chemical Hazard Communication (SCHC) to assure the recognition of competent professionals to author SDSs. To obtain the Registered Professional: SDS and Label Author credential, an individual must meet the established qualifications and must demonstrate competency in the skills and knowledge defined by the program’s Body of Knowledge. This AIHA Registry Program is the first EHS Specialty Credential that provides recognition for individuals who have expertise in this area. Learn more at http://bit.ly/sdsregistry.