Hindsight check: LL2-22 (Ch 22 Nanotechnology: early lessons from early warnings)#

Source section: EEA, Late lessons from early warnings: science, precaution, innovation (EEA Report No 1/2013), Ch 22, by Steffen Foss Hansen, Andrew Maynard, Anders Baun, Joel Tickner and Diana Bowman (report pp. 530–560; text pp. 530–551). The chapter has no panels. It updates the authors’ 2008 Nature Nanotechnology article. Check window: publication (2013) to late September 2026. The chapter’s own sources run to 2012. Checked: 26 September 2026.

Method note. - Search. General web search was not available for this pass because the session’s search budget was exhausted. I went directly to primary repositories and databases instead: - EU law (Commission communications, regulations, recommendations, CJEU judgments) in full text via the EU Publications Office (Cellar) mirror of EUR-Lex; - the Commission’s REACH web page and its Better Regulation initiative record; - the US Federal Register (API and full text) and govinfo (the Lautenberg Act); - the IARC Monographs site (Volume 111 PDF); - National Academies Press reading pages (NRC 2013; NASEM 2020); - nano.gov (the NNI budget supplements for FY2024 and FY2026, as PDFs); - the French ecology ministry’s R-Nano page; - Europe PMC (abstracts and some full texts), Crossref (for OECD test guideline titles and dates), and Nature’s article pages (for standfirsts only). - Access gaps. - The websites of ECHA, the EU Observatory for Nanomaterials, CDC/NIOSH and the OECD refused automated access (web application firewalls or 403 errors). NCBI Bookshelf served a CAPTCHA, which I did not try to get past. - As a result: - I cite NIOSH’s carbon nanotube exposure limit through a NIOSH-authored paper [S20], not the bulletin itself; - I have no ECHA statistics on how well nanoform registrations have complied since 2020; - I confirmed the OECD test guidelines only by title and date (Crossref), not by content. - I did not retrieve the UK, Canadian or Australia–New Zealand statements on E171, the food additive titanium dioxide. Where I mention them, I rely on a review by an industry-linked author [S26] and say so. - The paywalled Maynard and Aitken retrospective (2016) [S38] could only be read at standfirst level. - Annex 3 does not apply. This is a new 2013 chapter with no 2001 counterpart. - Scope. I read only the section digest, the section’s source extract and web sources. Page numbers are report pages of the 2013 volume. - Conflicts of role. - Author proximity. Andrew Maynard is a co-author of the chapter, the lead author of the 2006 milestones checked under Claim 8, the author of the 2011 definition critique checked under Claim 9, and a co-author of Poland et al. (2008). I treat his later commentaries [S38–S41] as the authors’ own reflections, not as independent confirmation. - Industry links. Warheit [S26], a consultant who formerly worked for industry, is an interested critic of the E171 ban. The TiO2 and carbon nanotube litigation and classification files were contested by producers (named as parties in [S23]). - Research-group proximity. The Empa/Nowack group [S28, S31–S33] has produced most of the probabilistic exposure modelling on silver and is not independent of the risk-assessment community the chapter addresses.

Overview#

1. The chapter’s diagnosis of the regulatory architecture is the part that aged best. - Voluntary reporting was superseded. The US EPA itself later wrote that its voluntary scheme yielded “limited reporting” [S10]. Where reporting became mandatory, far more data came in: - France’s R-Nano registry received over 9,000 declarations from 1,092 French declarants, covering about 340,000 tonnes, in its 2025 campaign [S9]; - Belgium, Denmark, Sweden and Norway set up similar registries [S8]; - the US adopted one-time mandatory reporting for nanoscale forms in 2017 [S10]. - Legacy identity systems were confirmed as a problem. - As late as February 2012, only 7 REACH registrations had ticked the optional “nanomaterial” field [S1]. - In 2025, Australia’s industrial chemicals scheme still identified substances by broad CAS-based identifiers [S7]. - Slow review cycles. The authors warned that nano-specific REACH provisions would not come before 2019. That proved almost exactly right: the Commission chose to amend the REACH annexes (October 2012) [S1], adopted the amendment in December 2018, and applied it from 1 January 2020 [S2]. - Definition politics. The definition dispute the chapter described was resolved in 2022 in favour of a hazard-independent, size-based definition, with the flexible threshold (which included a “competitiveness” ground) removed [S5].

2. The hazard warnings split in two directions. - Long, rigid multi-walled carbon nanotubes (MWCNTs): the warning was strongly vindicated. - IARC classified the MWCNT type MWCNT-7 as possibly carcinogenic to humans (Group 2B) in 2014 [S13, S14]. - A two-year rat inhalation study found lung carcinomas [S15]. - NIOSH set a recommended exposure limit of 1 µg/m³ [S20]. - The EU adopted a harmonised classification as a Category 1B carcinogen by inhalation for MWCNTs with diameter ≥30 nm, length ≥5 µm and aspect ratio >3:1. It applies from 1 May 2026 [S19]. - These criteria encode the fibre-length dependence that Poland et al. (2008) reported. - Ultrafine titanium dioxide (TiO2): contested in law. - The EU’s 2020 classification of TiO2 as a suspected carcinogen by inhalation rested mainly on the same Heinrich (1995) rat study the chapter cites. The General Court annulled it in 2022, and the Court of Justice dismissed the appeals on 1 August 2025 [S23]. - Separately, the EU banned E171 in food in 2022 because genotoxicity “could not be ruled out” [S24, S25]. This was a precautionary move that other agencies reportedly did not follow [S26].

3. The nanosilver concerns hold up on hazard and look weaker on risk. - Hazard at µg/L confirmed. The EU harmonised classification gives “silver nano” an aquatic acute and chronic multiplying factor (M-factor) of 1,000, against 10 for silver powder [S19]. - The question the chapter left open (is nano worse than ionic silver?) has largely been answered: generally not. Per unit of total silver, nano-silver is usually less toxic than dissolved silver [S28, S29]. - Transformation in the environment. Sulfidation in wastewater and the environment reduces toxicity [S30]. - Modelled risk is low. Probabilistic assessments put predicted environmental concentrations well below effect levels [S31, S32]. - “Nano” versus “conventional” is not a clean line. Conventional silver textiles shed as much nanoscale silver as “nano” ones [S33].

4. The governance recommendations were largely not adopted. - No separation within the NNI. - The NNI never separated management and budget authority for its EHS research (environmental, health and safety research) from its promotional role. - The National Research Council found “little progress” in 2013 and judged full separation unrealistic without a change in the statute [S34]. - In 2020 the National Academies praised the NNI’s “responsible development” record. The same report recorded that EHS fell from about 10% of NNI agency funding in 2016 to about 4% in 2020 [S35]. - The broader responsible-development spending line (Program Component Area 5, PCA 5) was about 2.3–2.4% of NNI funding in 2023–2025. The FY2026 request cut it to 1.1% ($15.5 million) and listed no NIOSH nanotechnology funding [S36, S37]. - Design rules only in voluntary form. Design-stage recommendations arrived as voluntary EU “safe and sustainable by design” (SSbD) frameworks, in 2022 and a 2026 revision [S48, S49]. They were never binding requirements.

5. The “two decades” lag is about right for horizontal law, but faster elsewhere. - Horizontal regimes and harmonised hazard classification took roughly two decades or more: - REACH nanoform rules from 2020; - the MWCNT carcinogen classification from 2026. - Targeted regimes moved faster: - US significant new use rules (SNURs) for carbon nanotubes from 2009–2011 [S12]; - NIOSH’s exposure limit, 2013 [S20]; - the EU’s safety opinions and conditional authorisation of nano-TiO2 in sunscreens, 2013–2016 [S27]. - The chapter’s “not less than two decades” also overstated its own source [S51].

6. Still no realised harm to test the lessons against. - I found no documented population-level harm from engineered nanomaterials in the check window. - A US cross-sectional study of carbon nanotube and nanofibre workers found most health measures unassociated with exposure [S21]. - Mesothelioma latency is decades, so absence of evidence is weak evidence. - In 2020 the National Academies judged that EHS research had “relieved some unfounded early concerns” [S35]. - Upshot. The chapter’s process lessons can be checked against what happened. Its outcome-level implications cannot yet be.

Implication for weight. - Strong weight: - legacy identity categories make new variants invisible; - voluntary disclosure yields little, and mandatory disclosure yields much more; - review cycles and “annex route” choices build in multi-year delays; - definitions are contested governance instruments; - promoter–overseer tension shows up in budget share over time. - Strong weight on the MWCNT hazard claim; moderate weight on the TiO2 claim, given the legal reversal. - Low to moderate weight on: - the implied environmental risk of nanosilver; - the precision of the “two decades” forecast; - the claim that the NNI’s dual role compromised EHS research, which is plausible but contested by the NNI’s own reviewers.

Claim-by-claim#

Claim 1: Voluntary nanomaterial reporting schemes failed, and jurisdictions moved to mandatory reporting (p. 537)#

Original claim (p. 537). Voluntary schemes were “at best underwhelming”: - the UK DEFRA scheme had 13 submissions over two years; - the US EPA scheme, which ended in 2009, had submissions from 31 organisations.

“Given the lack of buy-in from stakeholders”, France and California focused on mandatory reporting.

Subsequent developments - The US regulator’s own verdict. - EPA’s final rule of 12 January 2017 under TSCA section 8(a) described the Nanoscale Materials Stewardship Program (NMSP) as a voluntary programme run “from January 2008 to December 2009”. It noted that EPA “received limited reporting on nanoscale materials in commerce”, and that industry-wide submissions for silica and carbon black “gave an overview of the entire industry but not information on individual nanoscale materials” [S10]. - The rule replaced voluntarism with mandatory one-time reporting for existing discrete nanoscale forms, plus a standing requirement for new forms. Reporting covers chemical identity, production volume, manufacture and processing methods, exposure and release, and existing health and environmental data [S10]. - The 2016 Lautenberg Act separately gave EPA power to require testing “by rule, order, or consent agreement” [S11]. That eased the “prove risk to demand data” paradox the chapter described (p. 537). - Mandatory registries produce far more data. - France. R-Nano has been mandatory since 2013 and applies to quantities above 100 g a year. The 2025 campaign (report published January 2026) recorded [S9]: - just over 9,000 declarations covering 188 categories of substance; - 1,092 French declaring entities, plus more than 100 foreign ones; - about 340,000 tonnes placed on the French market in 2024. - Other countries. Belgium, Denmark, Sweden and Norway set up comparable national schemes [S8, S9]. - No EU-wide registry. - In October 2012 the Commission judged that “current knowledge about nanomaterials does not suggest risks which would require information about all products in which nanomaterials are used”. It offered instead a website linking to existing information sources, plus an impact assessment [S1]. - A 2021 comparison of the national registries found that a harmonised EU registry was “not foreseeable for the near future”. It also found that the national schemes differ in thresholds, level of detail and exemptions, which limits comparability [S8]. (I could not open the EU Observatory for Nanomaterials site in this pass to check its current role.) - Mandatory data are not the same as risk-relevant data. Pavlicek et al. (2021) note that quantitative, safety-relevant information remains “scarce”, even with the registries and the 2018 REACH changes [S8]. - Voluntary flags under REACH were also underused. As of February 2012, only 7 REACH registrations and 18 CLP notifications had selected “nanomaterial” in voluntary fields [S1]. This supports the chapter’s broader point that nano forms are invisible when disclosure is optional.

Verdict: held up (strengthened in its core). Both the failure of voluntary schemes and the shift to mandatory reporting were confirmed by the regulators themselves. The one caveat is that the chapter implied mandatory reporting would fix the data gap. It fixed the inventory gap more than the hazard and exposure gap.

Implication for weight. Strong weight for the mechanism that voluntary disclosure yields little and mandatory disclosure yields far more. Moderate weight for any stronger claim that disclosure alone enables risk management.

Claim 2: Nanomaterials might be “overlooked” in the 2012 REACH review; the next formal revision would not come before 2019; “immediate revisions are needed” and “substantial time is being wasted” (pp. 539–540)#

Original claim (pp. 539–540). - The failure of the Novel Foods negotiations was “a warning sign”. - There was “the potential for nanomaterials to be overlooked in the 2012 REACH revision discussion”. - Efforts to maintain the status quo were “worrying”. - The next REACH revision relevant to nanomaterials was “not scheduled before 2019”. - “Immediate revisions are needed”.

Subsequent developments - Not overlooked, but deferred. - The Commission issued a dedicated Second Regulatory Review on Nanomaterials (COM(2012) 572, 3 October 2012). It concluded that REACH “sets the best possible framework for the risk management of nanomaterials … but more specific requirements for nanomaterials within the framework have proven necessary”. It “envisage[d] modifications in some of the REACH Annexes” [S1]. - Its action annex, however, put the annex amendment under “Planning and Timelines will be addressed in REACH Review” [S1]. - The Communication also adopted the SCENIHR framing that nanomaterials “are similar to normal chemicals/substances in that some may be toxic and some may not”. It declined to change registration thresholds or the triggers for a chemical safety assessment [S1]. - Timing: the authors’ 2019 horizon proved almost exactly right. - Commission Regulation (EU) 2018/1881 was adopted on 3 December 2018 and applied from 1 January 2020 [S2]. It amended Annexes I, III and VI–XII to add the following: - a definition of “nanoform”, based on the 2011 definition; - minimum characterisation data (size, shape, surface chemistry); - rules for “sets of similar nanoforms”; - dissolution rate and dustiness data; - inhalation rather than oral testing for acute toxicity; - lung-lavage (bronchoalveolar lavage) and lung histopathology in inhalation studies. - Its recitals cite the 2012 Communication directly [S2]. Safety data sheet provisions followed in 2020 [S3]. - So nano-specific provisions arrived about seven years after the “immediate revisions” call, and 13 years after REACH was adopted. - The fixes left gaps of their own. - The REACH nanoform definition still rests on the 2011 Recommendation, not its 2022 revision. - A 2025 analysis found “grey areas” in how REACH applies to multicomponent and surface-modified nanomaterials. It recommended updating the guidance and the nanoform definition [S6]. - The broader REACH revision is still pending. - The Commission’s REACH page still says it “is currently revising the REACH Regulation” [S3]. - The Better Regulation record for the revision (initiative 12959, last modified 16 July 2026) lists the legislative proposal stage with no publication or adoption date [S4]. - I could not confirm whether anything was adopted between mid-July and late September 2026. - Compliance data were not retrieved. I could not get ECHA’s statistics on how well nanoform registrations have complied since 2020 (the site blocked access).

Verdict: partly held up. - Wrong, taken literally. The prediction that nano would be “overlooked” was wrong: the Commission ran a nano-specific review. - Right in substance: - the status quo was defended; - the fix went through the slow annex route; - effective nano-specific requirements arrived at the authors’ 2019 horizon. - The chapter’s call for “immediate” revision was not met.

Implication for weight. Strong weight for the mechanism that reform windows are scarce and review cycles impose multi-year delays. Moderate weight for the specific political forecast.

Claim 3: The lag between nano products reaching the market and regulators’ use of EHS data “will not be less than two decades” (p. 549)#

Original claim (p. 549). - Citing Linkov et al. (2009) and Figure 22.2, the authors write that the lag’s precise extent is unclear, “but there is historical evidence indicating that it will not be less than two decades”. - The evidence given is Lawless (1977), whose finding is an average delay of “one or two decades”, and the 2001 EEA volume. - Lawless (1977) is missing from the reference list. It is Edward W. Lawless, Technology and Social Shock (Rutgers University Press, 1977) [S51].

Subsequent developments. The lag depends heavily on which regulatory instrument is measured. - Horizontal regimes: about two decades or more. - REACH. Nano-specific data requirements applied from January 2020 [S2]. That is about 20 years after the NNI launched and after nanomaterials were first commercialised at scale. - MWCNT classification. The EU harmonised classification was based on a RAC opinion of 18 March 2022, adopted June 2024, and applies from 1 May 2026 [S19]. That is about 22 years after the first rodent granuloma studies (2004) and 18 years after Poland et al. (2008). - TiO2. The EU’s inhalation carcinogen classification was adopted in 2019 and annulled in 2022/2025 [S23]. That was about 30 years after the 1990 ultrafine-TiO2 inflammation signals. - Targeted, pre-market or occupational regimes: under 15 years. - US new-chemicals controls. - EPA issued significant new use rules specific to MWCNTs and SWCNTs on 17 September 2010 and 6 May 2011 [S12]. - A Federal Register full-text search returns about 30 EPA final rules from 2009 to 2026 that mention carbon nanotubes and significant new use [S12]. I did not check each one. - NIOSH. A recommended exposure limit of 1 µg/m³ for carbon nanotubes and nanofibres (Current Intelligence Bulletin 65, 2013) [S20]. - EU cosmetics. - The Scientific Committee on Consumer Safety (SCCS) gave opinions on nano-TiO2 in 2013–2014. - Regulation (EU) 2016/1143 authorised it as a UV filter up to 25%, with specifications, and excluded uses “that may lead to exposure of the end-user’s lungs by inhalation” [S27]. - Nano-TiO2 sunscreens had been on the market since the 1990s, so even this lag was long. - EU food. - EFSA’s 2016 re-evaluation of E171 found no concern. - Its 2021 opinion reversed that. - The ban followed in January 2022 [S24, S25]. - Once EFSA had applied its 2018 nano guidance, the step from data to action took about five years. - Sourcing. “Not less than two decades” is stronger than the cited source (“one or two decades”). The observed spread, from under 10 to about 30 years depending on the instrument, fits Lawless’s range better than the chapter’s floor.

Verdict: partly held up. - Right in order of magnitude for horizontal chemicals law and harmonised hazard classification. - Too pessimistic for targeted pre-market and occupational instruments. - Worded more strongly than its evidence.

Implication for weight. Moderate weight for the qualitative mechanism: products outpace safety data, and regulators’ uptake lags further. Low weight for the specific two-decade floor.

Claim 4: Long MWCNTs produce asbestos-like responses, industry inhalation studies confirmed granulomatous inflammation, and ultrafine TiO2 is a “potential occupational carcinogen” (pp. 534–535)#

Original claim (pp. 534–535). - Poland et al. (2008) found that long MWCNTs injected into the body cavity of mice produced length-dependent inflammation and granulomas “qualitatively and quantitatively similar” to those caused by long asbestos. - Industry 90-day nose-only inhalation studies (e.g. Ma-Hock et al. 2009) showed concentration-dependent granulomatous inflammation. - On the basis of a single chronic rat inhalation tumour study (Heinrich et al. 1995), NIOSH (2011) called ultrafine TiO2 a “potential occupational carcinogen”, acting through a secondary, particle-driven mechanism.

Subsequent developments: carbon nanotubes - IARC (meeting October 2014; Volume 111, 2017). Evidence in humans was “inadequate”, with no epidemiological studies. Evidence in animals was: - “sufficient” for MWCNT-7; - “limited” for two MWCNTs of similar dimensions; - “inadequate” for other MWCNTs and for SWCNTs.

The overall evaluation: MWCNT-7 is Group 2B (possibly carcinogenic); other MWCNTs and SWCNTs are Group 3 (not classifiable). Mechanistic evidence for mesothelioma and lung cancer was rated “moderate” for MWCNTs [S13, S14]. - Chronic inhalation. A 104-week whole-body inhalation study of MWNT-7 in F344 rats found significantly increased lung carcinomas at 0.2 and 2 mg/m³ in males and 2 mg/m³ in females. It found “no development of pleural mesothelioma” [S15]. - Other routes and later work. - Repeated intratracheal instillation of MWCNT-7 induced both lung tumours and pleural mesotheliomas, dose-dependently. Compared on lung burden, its dose–response resembled the inhalation study [S17]. - Brief inhalation of MWCNT-7 strongly promoted lung tumours in mice given an initiating carcinogen [S16]. - A Japanese toxicologist involved in the evaluation calls the CNT case an example of hazard identification before mass exposure: “nanomaterials … must not repeat the same mistakes” as asbestos [S18]. - Regulatory uptake. - NIOSH. A recommended exposure limit of 1 µg/m³ elemental carbon (8-hour time-weighted average, respirable fraction) [S20]. - EU harmonised classification (Delegated Regulation (EU) 2024/2564, applying from 1 May 2026). It covers “multi-walled carbon tubes (synthetic graphite in tubular shape) with a geometric tube diameter range ≥ 30 nm to < 3 μm and a length ≥ 5 μm and aspect ratio > 3:1, including multi-walled carbon nanotubes”: - Carcinogen Category 1B by inhalation (H350i); - specific target organ toxicity, repeated exposure, Category 1 (lung).

The Commission recorded that stakeholder information contesting the RAC opinion was "not sufficient to cast doubts" [S19]. The dimension-based scope reflects the fibre and length paradigm. Thin, tangled MWCNTs are not covered.

• Human evidence (limited).
• In a cross-sectional study of 108 US workers at 12 CNT and carbon nanofibre facilities, 18% had CNT or nanofibre material in their sputum. Most health measures, including spirometry and pulmonary symptoms, were not associated with exposure. Associations with respiratory allergy, resting heart rate and haematocrit “may not be causal” [S21, S22].
• No human cancer data exist, and latency means none would be expected yet.
• The co-author Maynard wrote in 2016 that carbon nanotube risk concerns had largely vanished from public discussion while CNT products moved closer to consumers [S39, S40].

Subsequent developments: ultrafine TiO2 - EU classification reversed in court. - Delegated Regulation (EU) 2020/217 classified TiO2 powder containing ≥1% of particles ≤10 µm as a suspected carcinogen by inhalation. - The General Court annulled that classification on 23 November 2022 (T-279/20, T-283/20, T-288/20). It found that the Heinrich (1995) study “was the decisive study” and that the Committee for Risk Assessment (RAC), ECHA’s expert committee, had made a “manifest error of assessment”. The error was that RAC did not account for particle agglomeration and the resulting lower density when judging whether lung overload made the study unreliable. - The Court of Justice dismissed the appeals of France and the Commission on 1 August 2025 [S23]. It left untouched, as surplus to the decision, the General Court’s further view that particle-overload carcinogenicity is not an “intrinsic property”. - The chapter reported NIOSH’s reliance on this same Heinrich study. - Food additive E171 banned in the EU. - EFSA (6 May 2021) concluded that “a concern for genotoxicity could not be ruled out” and that E171 “can no longer be considered safe”, while finding no “immediate health concern” [S25]. - Regulation (EU) 2022/63 (14 January 2022) removed the authorisation [S24]. France had already suspended E171 from 2020 [S9]. - The E171 ban is contested. A 2024 review by an industry-linked consultant argues that EFSA made a “manifest error”. It reports that agencies in the UK, Canada, Australia and New Zealand disagreed, on the grounds that dispersed nanoparticle test materials are unrepresentative of E171 [S26]. I did not retrieve those agencies’ statements. - The IARC position is unchanged. IARC’s classification of TiO2 as Group 2B dates from 2006 and predates the chapter. I found no later IARC re-evaluation in this pass.

Verdict: partly held up. - Carbon nanotubes: strengthened. - Chronic inhalation carcinogenicity was shown. - IARC Group 2B was assigned for MWCNT-7. - The EU adopted a binding classification as a Category 1B carcinogen, framed in length and diameter terms. - Nuances: - the asbestos-like mesothelioma outcome has not appeared by inhalation in rats; - most CNT types remain “not classifiable”. - TiO2: contested. - The single-study basis the chapter flagged proved to be the legal weak point of the EU inhalation classification. - A precautionary ban on oral (food) use did go ahead, and is itself disputed.

Implication for weight. Strong weight for the lesson that hazard depends on form (fibre length and rigidity), not chemical identity. The CNT case is one of the clearest forward-looking vindications in the 2013 volume. Moderate weight for hazard claims resting on one animal study. The TiO2 case shows that such evidence can prompt precautionary measures but may not survive legal review.

Claim 5: Nanosilver is widely used in unknown volumes, is ecotoxic at µg/L, may affect wastewater treatment, and it is unresolved whether the nano form is more toxic than ionic silver (pp. 535–536)#

Original claim (pp. 535–536). - Freshwater algae EC50 values as low as 4 µg/L. - Crustacean values “far below 1 mg/L”. - Inhibition of nitrifying bacteria below 1 mg/L, so that wastewater treatment “may … be affected”. - Speciation “is likely also to be the case” for nanoparticles, but its influence on toxicity “has yet to be studied in depth”. - Environmental concentrations from consumer products were “uncertain”.

Subsequent developments - Hazard at µg/L confirmed and formalised. - Delegated Regulation (EU) 2024/2564 (RAC opinions of 2 June 2022; applying from 1 May 2026) created separate entries for silver massive, silver powder and “silver nano: [particle diameter > 1 nm ≤ 100 nm]”. - Silver nano is classified Aquatic Acute 1 and Aquatic Chronic 1, with M-factors of 1,000 and 1,000. Silver powder gets 10 and 10, and massive silver has no aquatic classification. All three forms carry reproductive toxicity Category 2 and nervous-system target organ toxicity (repeated exposure) Category 2 [S19]. - Under CLP’s M-factor bands, an acute M-factor of 1,000 corresponds to L(E)C50 values between 0.1 and 1 µg/L. The regulator thus treats the nano form as far more ecotoxic per unit mass than larger silver particles. - Nano versus ionic silver: largely answered. - Notter et al. 2014 meta-analysis. For nano-silver, 93.8% of the EC50 ratios showed the nano form less toxic than dissolved silver on a total-metal basis. Only 1.1% showed it more than twice as toxic [S28]. - Karaman et al. 2025 meta-analysis (28 datasets from 11 studies). Silver nitrate “exhibits higher toxicity than AgNPs in most cases” [S29]. - The emerging consensus. Toxicity is driven mainly by release of silver ions. The nano form is more toxic than bulk silver, but generally not more toxic than ionic silver. - Transformation in the environment reduces toxicity. Even partial sulfidation (reaction with sulphide), as happens in wastewater systems, sharply reduced silver nanoparticle toxicity to fish, a nematode and an aquatic plant. It did so mainly by lowering Ag⁺ release [S30]. - Low modelled environmental risk. Probabilistic assessments that account for the chemical form released put mean risk characterisation ratios for nano-silver in European freshwaters at about 0.03–0.06, far below 1 [S31, S32]. I did not retrieve a study that directly tests nitrification inhibition at realistic wastewater loads. - Consumer products. Textiles treated with conventional silver released as much or more nanoscale silver in wash water as “nano”-treated ones. The authors conclude that nano-silver “warrants a similar approach to conventional silver” [S33]. - Volumes. Data remain partial. France’s R-Nano gives national tonnages by category but does not single out silver in the summary I retrieved [S9].

Verdict: partly held up. - Held up: - the ecotoxic hazard at µg/L, now in a binding EU classification; - the importance of speciation. - Answered, mostly against the concern: the open question (whether nano is worse than ionic silver) has been answered largely in the negative. - Weakened: the implied environmental risks (to wastewater treatment and receiving waters). Later exposure modelling and transformation studies indicate low risk at realistic concentrations.

Implication for weight. Moderate weight for the general lesson that environmental transformation and speciation govern whether a hazard becomes a risk. The later evidence supports this lesson, but it undercuts the chapter’s level of concern. Low weight for nanosilver as an example of an under-regulated risk. A sharper transferable point: the “novel versus conventional” framing can mislead, because conventional products produced the same nanoscale forms [S33]. The co-author himself asked in 2014 whether a novelty framing “ends up obscuring some risks, while overplaying others” [S40].

Claim 6: The NNI’s dual role of promoting nanotechnology and assessing its risks is a conflict of interest that slows and compromises EHS research; a clear separation of management and budgetary authority is needed (pp. 546–547)#

Original claim (pp. 546–547). - The chapter quotes Denison (2007) and the National Academies (2012): “a clear separation of management and budgetary authority and accountability is needed”. - It argues that when promoters have strong influence over oversight, “independent regulatory decision-making becomes compromised”, and R&D decisions follow “what will ultimately promote the technology”.

Subsequent developments - No separation occurred. - NRC 2013. The National Research Council’s follow-up rated all its implementation indicators yellow or red, none green. It “determined that little progress had been made in establishing a clear separation”. It acknowledged that “absent a change in the NNI’s statutory mandate, establishment of wholly separate management and budgetary structures … may not be realistic”. - NRC 2013, fall-back proposals. It proposed instead: - scientific-integrity policies; - an ombudsman; - conflict-of-interest best practice; - designating one agency with EHS in its mission as lead [S34]. - Still under one roof in FY2026. EHS coordination still runs through the NSET subcommittee’s Nanotechnology Environmental and Health Implications (NEHI) working group, inside the same NNI structure [S37]. - Budget share fell. - 2008–2016. By Maynard’s 2016 count, US federal nano-EHS research rose from nearly $60 million a year in 2008 to a proposed $105.4 million in 2016 [S39]. - 2016–2020. The National Academies’ 2020 review then recorded a fall from “about 10 percent of NNI agency funding” in 2016 to “4 percent, or $80 million” in 2020 [S35]. - 2023 onwards. PCA 5, now labelled “Responsible Development” and covering EHS plus ethical, legal and social issues (ELSI), was: - $50.3 million of $2,073.9 million in 2023, or 2.4% [S36]; - $53.9 million of $2,248.3 million in 2024 and $49.1 million of $2,122.4 million in 2025 (enacted), about 2.3–2.4% [S37]; - $15.5 million of $1,449.1 million in the President’s 2026 request, or 1.1%. NIOSH’s nanotechnology line went from $10.5 million (2024) to $8.0 million (2025) to $0.0 (2026 request), and NIST’s PCA 5 line to zero [S37]. - PCA 5 is broader than EHS, so these shares are not strictly comparable with the 2016 EHS figure. - The official framing supports the chapter’s concern in one respect. The FY2026 supplement describes responsible development as supporting “the other NNI goals by fostering public confidence and regulatory certainty, which speeds laboratory discoveries to market” [S37]. This treats safety research as instrumental to promotion. - Counter-evidence on outcomes. - The 2020 National Academies review found that on the responsible-development goal “the NNI has performed exceptionally well and is recognized internationally for its leadership in responsible nanotechnology development, although agency engagement appears to be waning”. - It credited EHS funding with test methods, alternative testing strategies and grouping approaches. - It also said EHS research “relieved some unfounded early concerns” [S35]. - Its overall thrust was to refocus the NNI on faster commercialisation [S35]. - Maynard, in a 2016 standfirst co-written with Aitken, presents the decade as progress with work remaining [S38].

Verdict: partly held up. - Not adopted. The recommendation was never implemented and was judged unrealistic by the body that originally made it. - Diagnosis consistent with later evidence. The long-run fall in the EHS and responsible-development share, the near-elimination of occupational-safety nano funding in the FY2026 request, and the framing of safety as a route to market all fit the structural-tension diagnosis. - “Compromised” is contested. The claim that the dual role compromised EHS research is disputed by the National Academies’ 2020 positive assessment. The 2020 committee was oriented to commercialisation, and its praise sits alongside its own record of falling EHS share.

Implication for weight. Moderate-to-strong weight for the mechanism that when promotion and oversight share one budget, the oversight share erodes as the technology matures and the promotional agenda shifts. The trend data support this well. Moderate weight for the stronger claim that the conflict degraded research quality; I found no direct evidence of that. Also note that separation was judged infeasible without a change in statute. That is itself a transferable lesson about how hard it is to redesign an institution after it has been set up.

Claim 7: Twenty years after the first indications of harm, “many governments still call for more information as a substitute for action”, and nano risk management is being “paralysed by analysis” (p. 547)#

Original claim (p. 547). Since Ferin and Oberdörster (1990), “many governments still call for more information as a substitute for action”. The authors write that “we need to act on what we know now”.

Subsequent developments - Evidence the diagnosis was accurate in 2012. - The Commission’s October 2012 Communication declined to require product-level information, since current knowledge “does not suggest risks which would require” it. It relied instead on case-by-case assessment and further method development [S1]. - The US voluntary programme’s failure was followed by eight years before mandatory reporting [S10]. - Evidence of action after 2013. Several binding or formal measures followed, most of them resting on hazard evidence rather than on proof of harm: - EU cosmetics authorisation of nano-TiO2 with a precautionary exclusion of inhalable sprays (2016) [S27]; - US mandatory nanoscale reporting (2017) [S10]; - REACH nanoform requirements (adopted 2018, applied 2020) [S2]; - the EU E171 ban (2022), explicitly under uncertainty [S24]; - harmonised EU classifications for MWCNTs and nano-silver (adopted 2024, applying 2026) [S19]; - a NIOSH exposure limit for CNTs (2013) [S20]; - French, Belgian, Danish, Swedish and Norwegian registries [S8, S9]. - Evidence of continued slowness. - no EU-wide registry [S8]; - the REACH revision still pending in mid-2026 [S4]; - REACH’s nanoform definition not updated to the 2022 Recommendation [S6]; - the TiO2 inhalation classification lost in court [S23]. - Research did reduce some concerns. The National Academies (2020) recorded that EHS research “relieved some unfounded early concerns about the safety of nanomaterials relative to their conventional counterparts” [S35]. It also showed that some concerns, such as silver toxicity relative to ionic silver, were smaller than feared [S28, S29]. This cuts against the idea that further analysis was merely a stalling device. - Outcome. No population-level harm has been documented in the window. The worker cross-sectional study was largely reassuring [S21]. The long latency of fibre-related cancers makes this weak evidence.

Verdict: partly held up. - An accurate description of the 2012 political stance. - As a characterisation of 2013–2026, weakened: governments did act on existing hazard knowledge in several domains, including precautionary action (E171, cosmetics sprays, the MWCNT classification). - Horizontal and EU-wide measures stayed slow.

Implication for weight. Moderate weight for the mechanism that calls for more research can stand in for action. It is best treated as a recurrent risk, not a constant. Here the risk was realised mainly in horizontal and whole-market instruments, while targeted instruments acted on hazard evidence within about a decade. The case also shows the opposite risk the chapter does not discuss: further research can legitimately reduce concern.

Claim 8 (benchmark): Maynard et al. (2006) research milestones (p. 543)#

Original claim (p. 543). The milestones were: - strategic risk-research programmes within 12 months; - exposure instruments within 3–10 years; - robust life-cycle evaluation systems within 5 years; - validated toxicity methods within 5–15 years; - predictive models within 10 years.

The chapter adds that EU calls under the Seventh Framework Programme (FP7) and the NNI’s 2008 strategy reflected some of these. Note that this benchmark was set by a co-author.

Subsequent developments (milestone by milestone) - Strategic programmes (by about 2007): met, broadly. - The NNI published EHS research strategies in 2008 and 2011. The 2011 strategy was being “refreshed” in 2023 [S36, S34]. - EU framework programmes funded nanosafety research throughout [S35]. - The NRC (2013) still faulted the implementation mechanisms: coordination, authority and stakeholder engagement [S34]. - Exposure instruments (by 2009–2016): partly met. - NIOSH refined its field method into the Nanomaterial Exposure Assessment Technique (NEAT 2.0) in 2016 [S44] and visited more than 100 facilities [S35]. - Workplace assessment still combined elemental-carbon mass sampling with electron-microscopy counts [S20, S22]. That points to a validated protocol, not a routine direct-reading, nano-specific personal monitor. - Life-cycle evaluation systems (by about 2011): missed. - The NRC (2013) rated progress on characterising releases along the value chain “yellow” and the ability to model them “red” [S34]. - Life-cycle and form-specific release models matured only in the late 2010s and early 2020s [S31, S32]. - Validated methods (by 2011–2021): partly met, and late. - OECD inhalation test guidelines 412 and 413 were revised in June 2018 [S43]. REACH now requires lung lavage and histopathology in line with OECD guidance [S2]. - Nano-specific OECD test guidelines appeared as follows [S43]: - dispersion stability (TG 318, October 2017); - volume-specific surface area (TG 124, June 2022); - particle size (TG 125, July 2023); - hydrophobicity (TG 126, July 2023); - dissolution for fate assessment (TG 322, July 2026). - These are mostly physicochemical and fate methods. Nano-specific toxicity test guidelines remain few. - A 2013 multi-stakeholder paper still framed alternative test strategies as an aspiration [S45]. - Predictive models (by about 2016): largely missed. - Grouping and read-across frameworks developed in 2018–2021, and REACH now allows “sets of similar nanoforms” [S2, S46, S47]. - The National Academies (2020) still listed “predictive modeling and risk assessment across the product life cycle” as tools the US needs to lead on [S35]. - The authors’ own review. Maynard and Aitken’s ten-year retrospective (2016) looked at “where we have come, and where we still need to go” [S38]. Its figure tracked publication output against the challenges. I could read only the standfirst.

Verdict: partly held up. Read as a benchmark, the milestones were partly met, and mostly late: - strategy was delivered on time; - exposure methods were partly delivered, late; - validated methods were partly delivered, at or beyond the 15-year outer bound; - life-cycle systems and predictive models were largely missed.

Implication for weight. - The chapter’s point stands: coordinated research on emerging-technology risks moves much more slowly than product development. - A further lesson from the record: expert timetables for building risk-assessment capacity were optimistic by about 5–10 years for the harder milestones. - Weight the benchmark as evidence of slow capacity-building, not as evidence of neglect. Research output was large [S35, S38].

Claim 9: The EC’s October 2011 policy-based definition risks sidelining the science (Maynard 2011) versus being “urgently needed” (Stamm 2011); “both camps have valid points” (p. 540)#

Original claim (p. 540). - Maynard (2011). A “one size fits all” policy definition “has the potential to sideline the science and may fail to capture what is important for addressing risk” [S42]. - Stamm (2011). A definition is needed “to identify a general class of materials for attention — whether they are benign or hazardous”. - The authors. Both have valid points. The definition must not become “a barrier to the effective regulation of nanomaterials”.

Subsequent developments - The 2022 revision. The Commission Recommendation of 10 June 2022 [S5]: - kept the hazard-independent basis. Recital 7 states that the definition is based on particle number–size distribution “irrespective to its potential inherent hazardous properties or risks”, which is Stamm’s rationale; - kept the 1–100 nm range and the 50% threshold; - removed the flexibility to lower the threshold to 1–50% in specific cases, “to ensure regulatory consistency”. That removed the “competitiveness” ground the chapter flagged; - restricted the definition to solid particles; - dropped specific surface area as a qualifier, keeping it only as an exclusion below 6 m²/cm³; - extended the sub-1 nm provision from named carbon materials to all particles of similar shape. - The implementation difficulty was real. - The review ran from 2013 to 2021, against a planned 2014 review [S1, S5]. - It relied on the NanoDefine measurement project. - The Recommendation concedes that identifying constituent particles in aggregates “can be very challenging” [S5]. - Sectoral fragmentation persists. - REACH’s nanoform concept is still built on the 2011 text [S2, S6]. - The cosmetics definition has differed since 2009, which the chapter already noted. - A 2025 analysis identified “grey areas” for multicomponent materials [S6]. - Risk-relevant categories cut across the definition. - The EU’s MWCNT carcinogen classification is defined by fibre dimensions (diameter ≥30 nm up to 3 µm; length ≥5 µm) [S19]. That is outside the 1–100 nm logic of the definition. - This supports Maynard’s point that a size-based definition does not track hazard. - It also supports Stamm’s view that a definition can serve as a trigger for more specific, hazard-based rules. - The author’s later view. In 2014 Maynard asked whether framing nanomaterial risk around “novel material behaviours” ends up “obscuring some risks, while overplaying others” [S40].

Verdict: held up. The chapter’s balanced judgement is borne out. - A definition was adopted, used across EU law and revised. It works as a trigger, not as a risk category. - Its scientific limits were acknowledged, and later risk rules used other criteria. - The competitiveness clause the chapter questioned was removed.

Implication for weight. Strong weight for the lesson that definitions of an emerging-technology category are governance instruments. They are needed as triggers, but they do not track hazard, and they take many years to stabilise. Hazard-specific rules then have to be written in terms that cut across the category.

Claim 10: Expert recommendations on design and green nanotechnology have “yet to be successfully implemented by political decision-makers”, leaving an environment that hinders precautionary strategies (pp. 550–551)#

Original claim (pp. 550–551). The authors list recommendations: - design rules and training for chemists; - “green nanotechnology”; - EHS as quality concerns “equal to economic and performance considerations”; - sustainability research funded “at levels significant enough to identify early warnings”; - “regulatory systems [that] provide incentives for safer and sustainable materials”.

They say “many of them are not or have yet to be successfully implemented”.

Subsequent developments - EU: voluntary design frameworks. - The 2020 Chemicals Strategy for Sustainability announced “safe and sustainable by design” (SSbD). Commission Recommendation (EU) 2022/2510 (8 December 2022) established a voluntary assessment framework with a testing period [S48]. - A revised Recommendation (6 March 2026) keeps it “a robust, voluntary decision-making tool”. The Commission promised to promote it through Chemicals Innovation and Substitution Hubs and an Advanced Materials Act “before the end of the year” [S49]. - The 2024 Communication on advanced materials says SSbD “will be at the core”. Its framing, though, is “industrial leadership”, and it considers “regulatory sandboxes” [S50]. - Separately, the policy vocabulary moved from “nanomaterials” towards “advanced materials” [S49, S50]. - Research and tools. EU and US programmes funded safe-by-design, grouping and alternative testing tools [S35, S46, S47]. - Funding at “levels significant enough”. The US trend runs the other way from 2016: the EHS share of NNI funding fell from about 10% to 4% by 2020, and to a 1.1% share for all responsible development in the FY2026 request [S35, S37]. - Regulatory incentives. I found no binding regulatory incentive specific to safer nanomaterials in the EU or US in the window. The incentives that do exist are general: substitution, SSbD and hazard classification. - Precautionary measures were adopted in places. Examples include the E171 ban [S24], the cosmetics inhalation exclusion [S27] and the MWCNT classification [S19]. The environment did not entirely prevent precautionary action.

Verdict: partly held up. - Still true in its core: none of the design-stage recommendations became binding, and EHS was not made “equal to economic and performance considerations” in law or funding. - Taken up in voluntary form: SSbD. - Countered in part: several precautionary measures went ahead.

Implication for weight. Moderate-to-strong weight for the lesson that design-stage safety recommendations tend to be adopted as voluntary frameworks, while competitiveness framing and funding decisions shape the binding environment. Weight any claim that the environment prevents precaution more lightly, because targeted precautionary measures did occur.

Cross-cutting observations#

1. Lessons about process have held up better than lessons about outcomes. The best-confirmed claims concern: - regulatory architecture: identity keys, disclosure, review cycles, definitions; - the shape of hazard: fibre length.

The weakest concern: - the size of environmental risk (nanosilver); - forecasts stated as fixed floors (“not less than two decades”). 2. The case is still prospective. Thirteen years on, there is no realised population-level harm against which to test whether the precautionary stance was proportionate. There is also no evidence that the delays the chapter decried caused harm. The strongest hazard finding (MWCNTs) was, as a later commentator put it, identified “before mass production and consumption”, before the exposures happened [S18]. That can be read either as the anticipatory approach working, or as luck. 3. Precaution met the courts. The TiO2 reversal [S23] shows a dynamic the chapter does not discuss. Precautionary classification based on a single decisive animal study can be undone in court for errors in method. Precaution under uncertainty therefore depends on evidentiary processes that can withstand challenge. 4. Framing shifted. The fading of “nano” into “advanced materials” and the reframing of EHS as a route to “regulatory certainty” [S37, S50] suggest the promoter–overseer tension did not go away. It re-emerged in the language of competitiveness.

Sources#

Sources retrieved 25–26 September 2026. Dates are publication dates.

EU law, policy and case law - [S1] European Commission. Second Regulatory Review on Nanomaterials, COM(2012) 572 final, 3 October 2012. https://eur-lex.europa.eu/legal-content/EN/TXT/?uri=CELEX:52012DC0572 (retrieved via http://publications.europa.eu/resource/celex/52012DC0572) - [S2] Commission Regulation (EU) 2018/1881 of 3 December 2018 amending REACH Annexes I, III, VI–XII to address nanoforms (OJ L 308, 4.12.2018; applies from 1 January 2020). https://eur-lex.europa.eu/eli/reg/2018/1881/oj - [S3] European Commission, DG Environment. “REACH Regulation” web page (nanomaterials section; revision status), accessed 26 September 2026. https://environment.ec.europa.eu/topics/chemicals/reach-regulation_en - [S4] European Commission, Better Regulation portal, initiative 12959, “Revision of EU legislation on registration, evaluation, authorisation and restriction of chemicals” (API record last modified 16 July 2026; the proposal stage has no publication date). https://ec.europa.eu/info/law/better-regulation/have-your-say/initiatives/12959-Chemicals-legislation-revision-of-REACH-Regulation-to-help-achieve-a-toxic-free-environment_en - [S5] Commission Recommendation of 10 June 2022 on the definition of nanomaterial (2022/C 229/01), OJ C 229, 14.6.2022. https://eur-lex.europa.eu/legal-content/EN/TXT/?uri=CELEX:32022H0614(01) - [S19] Commission Delegated Regulation (EU) 2024/2564 of 19 June 2024 amending CLP as regards harmonised classification of certain substances (OJ L, 30.9.2024; applies from 1 May 2026). Entries 006-104-00-2 (MWC(N)T) and 047-004/005/006 (silver massive, powder, nano). https://eur-lex.europa.eu/eli/reg_del/2024/2564/oj - [S23] Court of Justice of the EU, judgment of 1 August 2025, Joined Cases C-71/23 P and C-82/23 P, France and Commission v CWS Powder Coatings and Others (dismissing the appeals against General Court judgment of 23 November 2022, T-279/20, T-283/20 and T-288/20, which annulled the TiO2 classification in Delegated Regulation (EU) 2020/217). https://eur-lex.europa.eu/legal-content/EN/TXT/?uri=CELEX:62023CJ0071 - [S24] Commission Regulation (EU) 2022/63 of 14 January 2022 amending Annexes II and III to Regulation (EC) No 1333/2008 as regards titanium dioxide (E 171) (OJ L 11, 18.1.2022). https://eur-lex.europa.eu/eli/reg/2022/63/oj - [S27] Commission Regulation (EU) 2016/1143 of 13 July 2016 amending Annex VI to the Cosmetics Regulation (titanium dioxide (nano) as UV filter) (OJ L 189, 14.7.2016). https://eur-lex.europa.eu/eli/reg/2016/1143/oj - [S48] Commission Recommendation (EU) 2022/2510 of 8 December 2022 establishing a European assessment framework for “safe and sustainable by design” chemicals and materials (OJ L 325, 20.12.2022). https://eur-lex.europa.eu/eli/reco/2022/2510/oj - [S49] European Commission, DG Research and Innovation. “Strengthened Framework for Safe and Sustainable by Design chemicals and materials …”, news, 6 March 2026. https://research-and-innovation.ec.europa.eu/news/all-research-and-innovation-news/strengthened-framework-safe-and-sustainable-design-chemicals-and-materials-drive-europes-clean-2026-03-06_en - [S50] European Commission. Advanced Materials for Industrial Leadership, COM(2024) 98 final, 27 February 2024. https://eur-lex.europa.eu/legal-content/EN/TXT/?uri=CELEX:52024DC0098

EU agencies and national registries - [S9] Ministère de la Transition écologique (France). “Nanomatériaux” page, including R-Nano 2025 campaign data (report published January 2026) and the E171 suspension from 2020. https://www.ecologie.gouv.fr/politiques-publiques/nanomateriaux - [S25] EFSA FAF Panel. Safety assessment of titanium dioxide (E171) as a food additive. EFSA Journal 19(5):6585, 6 May 2021. https://doi.org/10.2903/j.efsa.2021.6585

US law, agencies and reviews - [S10] US EPA. Chemical Substances When Manufactured or Processed as Nanoscale Materials; TSCA Reporting and Recordkeeping Requirements. Final rule, 82 FR 3641, 12 January 2017 (effective 12 May 2017). https://www.federalregister.gov/documents/2017/01/12/2017-00052/chemical-substances-when-manufactured-or-processed-as-nanoscale-materials-tsca-reporting-and - [S11] Frank R. Lautenberg Chemical Safety for the 21st Century Act, Public Law 114-182, 22 June 2016. https://www.govinfo.gov/content/pkg/PLAW-114publ182/html/PLAW-114publ182.htm - [S12] US EPA. Multi-Walled Carbon Nanotubes and Single-Walled Carbon Nanotubes; Significant New Use Rules, 17 September 2010, https://www.federalregister.gov/documents/2010/09/17/2010-23321/multi-walled-carbon-nanotubes-and-single-walled-carbon-nanotubes-significant-new-use-rules ; Multi-Walled Carbon Nanotubes; Significant New Use Rule, 6 May 2011, https://www.federalregister.gov/documents/2011/05/06/2011-11127/multi-walled-carbon-nanotubes-significant-new-use-rule ; plus a Federal Register API full-text count (about 30 EPA final rules mentioning “carbon nanotubes” and significant new use, 2009–2026; not individually checked). https://www.federalregister.gov/api/v1/documents.json - [S34] National Research Council. Research Progress on Environmental, Health, and Safety Aspects of Engineered Nanomaterials. National Academies Press, 2013 (Summary). https://doi.org/10.17226/18475 ; https://nap.nationalacademies.org/read/18475/chapter/2 - [S35] National Academies of Sciences, Engineering, and Medicine. A Quadrennial Review of the National Nanotechnology Initiative: Nanoscience, Applications, and Commercialization. National Academies Press, 2020 (Summary and Ch. 2, “PCA 5: Environmental Health and Safety”). https://doi.org/10.17226/25729 ; https://nap.nationalacademies.org/read/25729/chapter/2 ; https://nap.nationalacademies.org/read/25729/chapter/4 - [S36] National Science and Technology Council. The National Nanotechnology Initiative Supplement to the President’s 2024 Budget (Tables 4–5). https://www.nano.gov/sites/default/files/pub_resource/NNI-FY24-Budget-Supplement.pdf - [S37] National Science and Technology Council. The National Nanotechnology Initiative Supplement to the President’s 2026 Budget (Tables 2–4; PCA 5 text). https://www.nano.gov/sites/default/files/NNI-FY26-Budget-Supplement.pdf

IARC and hazard science - [S13] IARC. Some Nanomaterials and Some Fibres. IARC Monographs Vol. 111 (2017; Working Group October 2014). Carbon nanotubes chapter, §6 Evaluation. https://publications.iarc.who.int/552 ; https://publications.iarc.who.int/download/mono111-01.pdf - [S14] Grosse Y, et al. Carcinogenicity of fluoro-edenite, silicon carbide fibres and whiskers, and carbon nanotubes. Lancet Oncology 15(13):1427–1428, December 2014. https://doi.org/10.1016/S1470-2045(14)71109-X - [S15] Kasai T, et al. Lung carcinogenicity of inhaled multi-walled carbon nanotube in rats. Particle and Fibre Toxicology 13:53, October 2016. https://doi.org/10.1186/s12989-016-0164-2 - [S16] Porter DW, et al. Potent lung tumor promotion by inhaled MWCNT. Nanotoxicology, February 2024. https://doi.org/10.1080/17435390.2024.2314473 - [S17] Maeno A, et al. Dose-response relationships for lung tumor and pleural mesothelioma induction by repetitive intratracheal instillation of MWCNT-7 in rats. Nanomaterials 16:1039, August 2026. https://doi.org/10.3390/nano16161039 - [S18] Kanno J. Carcinogenicity assessment: “modern toxicology” considerations from experience in the evaluation of a carbon nanotube. Journal of Occupational Health 67(1):uiaf013, 2025. https://doi.org/10.1093/joccuh/uiaf013 - [S20] Dahm MM, et al. (NIOSH). Carbon nanotube and nanofiber exposure assessments: an analysis of 14 site visits. Annals of Occupational Hygiene, April 2015 (states the NIOSH REL of 1 µg/m³ EC, 8-h TWA, respirable). https://doi.org/10.1093/annhyg/mev020 - [S21] Schubauer-Berigan MK, et al. Association of pulmonary, cardiovascular, and hematologic metrics with carbon nanotube and nanofiber exposure among U.S. workers: a cross-sectional study. Particle and Fibre Toxicology 15:22, May 2018. https://doi.org/10.1186/s12989-018-0258-0 - [S22] Dahm MM, et al. Exposure assessments for a cross-sectional epidemiologic study of US carbon nanotube and nanofiber workers. International Journal of Hygiene and Environmental Health, January 2018. https://doi.org/10.1016/j.ijheh.2018.01.006 - [S26] Warheit DB. Safety of titanium dioxide (E171) as a food additive for humans. Frontiers in Toxicology 6:1333746, July 2024 (author: Warheit Scientific LLC; industry-linked critic). https://doi.org/10.3389/ftox.2024.1333746

Nanosilver - [S28] Notter DA, Mitrano DM, Nowack B. Are nanosized or dissolved metals more toxic in the environment? A meta-analysis. Environmental Toxicology and Chemistry 33(12), September 2014. https://doi.org/10.1002/etc.2732 - [S29] Karaman E, et al. Differential toxicity of ionic silver and silver nanoparticles: a meta-analysis of ecotoxicological studies. Toxics 14(1):28, December 2025. https://doi.org/10.3390/toxics14010028 - [S30] Levard C, et al. Sulfidation of silver nanoparticles: natural antidote to their toxicity. Environmental Science & Technology 47(23), November 2013. https://doi.org/10.1021/es403527n - [S31] Coll C, et al. Probabilistic environmental risk assessment of five nanomaterials (nano-TiO2, nano-Ag, nano-ZnO, CNT, and fullerenes). Nanotoxicology 10(4), 2016 (online November 2015). https://doi.org/10.3109/17435390.2015.1073812 - [S32] Hong H, Adam V, Nowack B. Form-specific and probabilistic environmental risk assessment of 3 engineered nanomaterials (nano-Ag, nano-TiO2, and nano-ZnO) in European freshwaters. Environmental Toxicology and Chemistry 40:2629–2639, 2021. https://doi.org/10.1002/etc.5146 - [S33] Mitrano DM, et al. Presence of nanoparticles in wash water from conventional silver and nano-silver textiles. ACS Nano 8(7), June 2014. https://doi.org/10.1021/nn502228w

Registries, REACH implementation and definitions - [S6] Hunt N, et al. Regulatory preparedness for multicomponent nanomaterials: current state, gaps and challenges of REACH. NanoImpact 37:100538, 2025 (online December 2024). https://doi.org/10.1016/j.impact.2024.100538 - [S7] Wilson S, et al. Beyond molecular structure: comparing Australian and European regulatory approaches to nano-identification and classification. Regulatory Toxicology and Pharmacology, online September 2025. https://doi.org/10.1016/j.yrtph.2025.105947 - [S8] Pavlicek A, et al. A European nano-registry as a reliable database for quantitative risk assessment of nanomaterials? A comparison of national approaches. NanoImpact 21:100276, 2021 (online November 2020). https://doi.org/10.1016/j.impact.2020.100276

Methods, milestones and research capacity - [S43] OECD Guidelines for the Testing of Chemicals (titles and dates confirmed via Crossref only): TG 318 Dispersion Stability of Nanomaterials in Simulated Environmental Media (9 October 2017), https://doi.org/10.1787/9789264284142-en ; Guidance Document for the Testing of Dissolution and Dispersion Stability of Nanomaterials (20 July 2020), https://doi.org/10.1787/f0539ec5-en ; TG 412 and TG 413 inhalation toxicity, revised 27 June 2018, https://doi.org/10.1787/9789264070783-en and https://doi.org/10.1787/9789264070806-en ; TG 124 Volume Specific Surface Area of Manufactured Nanomaterials (30 June 2022), https://doi.org/10.1787/abb72f8f-en ; TG 125 Nanomaterial Particle Size and Size Distribution (4 July 2023), https://doi.org/10.1787/af5f9bda-en ; TG 126 Hydrophobicity Index of Nanomaterials (4 July 2023), https://doi.org/10.1787/ae9c0fd1-en ; TG 322 Solubility and Dissolution Rate of Nanomaterials for Environmental Fate Assessment (2 July 2026), https://doi.org/10.1787/693e2680-en - [S44] Eastlake AC, et al. (NIOSH). Refinement of the Nanoparticle Emission Assessment Technique into the Nanomaterial Exposure Assessment Technique (NEAT 2.0). Journal of Occupational and Environmental Hygiene 13(9), 2016. https://doi.org/10.1080/15459624.2016.1167278 - [S45] Nel AE, et al. A multi-stakeholder perspective on the use of alternative test strategies for nanomaterial safety assessment. ACS Nano 7(8), 2013. https://doi.org/10.1021/nn4037927 - [S46] Lamon L, et al. Grouping of nanomaterials to read-across hazard endpoints: a review. Nanotoxicology 13(1), 2019 (online September 2018). https://doi.org/10.1080/17435390.2018.1506060 - [S47] Mech A, et al. Insights into possibilities for grouping and read-across for nanomaterials in EU chemicals legislation. Nanotoxicology 13(1), 2019 (online September 2018). https://doi.org/10.1080/17435390.2018.1513092

Chapter authors’ later publications (not independent) - [S38] Maynard AD, Aitken RJ. “Safe handling of nanotechnology” ten years on. Nature Nanotechnology 11:998–1000, 6 December 2016 (standfirst and figure caption only). https://doi.org/10.1038/nnano.2016.270 - [S39] Maynard AD. “We don’t talk much about nanotechnology risks anymore, but that doesn’t mean they’re gone”, 1 February 2016. https://www.futureofbeinghuman.com/p/we-dont-talk-much-about-nanotechnology-risks-anymore-but-that-doesn-t-mean-they-re-gone-ba00cdcf6ab5 (text mirror: https://text.futureofbeinghuman.com/substack/we-dont-talk-much-about-nanotechnology-risks-anymore-but-that-doesn-t-mean-they-re-gone-ba00cdcf6ab5.html) - [S40] Maynard AD. Is novelty overrated? Nature Nanotechnology 9:409–410, 4 June 2014 (standfirst only); also Maynard AD, Are we ready for spray-on carbon nanotubes? Nature Nanotechnology, 7 June 2016 (standfirst only). https://doi.org/10.1038/nnano.2014.116 ; https://doi.org/10.1038/nnano.2016.99 - [S41] Maynard AD. A decade of uncertainty. Nature Nanotechnology 9:159–160, 5 March 2014 (standfirst only; consulted, not relied on). https://doi.org/10.1038/nnano.2014.43 - [S42] Maynard AD. Don’t define nanomaterials. Nature 475:31, 6 July 2011 (the source of the chapter’s “sideline the science” position; cited for identification). https://doi.org/10.1038/475031a

Other - [S51] Lawless EW. Technology and Social Shock. Rutgers University Press, 1977 (catalogue record; the chapter cites it but omits it from its reference list). https://openlibrary.org/works/OL18245744W