Registration Dossier

Ecotoxicological information

Short-term toxicity to aquatic invertebrates

Currently viewing:

Administrative data

Link to relevant study record(s)

Description of key information

 Dispersions of microsized TiO2 are not acutely toxic to freshwater and marine invertebrates up to at least 1000 and 10,0000 mg/L (EC/LC50: > 1000 and > 10,000), respectively. Considering the low solubility of microsized TiO2, it is further concluded that microsized TiO2 is not acutely toxic up to its solubility limit. Dispersions of nanosized TiO2 are also not acutely toxic to freshwater and marine invertebrates up to at least 100 mg TiO2/L (EC50: > 100 mg/L).

Key value for chemical safety assessment

Additional information

Microsized TiO2:

Dispersions of four different microsized TiO2 materials did not affect the mobility of Daphnia magna in an acute toxicity test according to US EPA (660/8-87/011; 1987) and ASTM (Standard E729, 1986). Unbounded 48 h EC50 values of > 1000 mg TiO2/L (nominal) were derived for all four materials (Haley and Kurnas, 1993). Johnson et al. (1986) observed that dispersions of the microsized TiO2 material Titanox 1000 (particle size: 5-6 µm) did not affect the mobility of first instar Daphnia magna in an acute toxicity test according to OECD 202 and US EPA (660/8-87/011; 1987). The determined unbounded 48 h EC50 value was > 1000 mg TiO2/L (nominal). Wiench et al. (2009) also did not observe acute toxicity of dispersions of microsized TiO2 (200-300 nm) up to 100 mg/L in artificial or natural water in acute toxicity tests with Daphnia magna according to OECD 202. However, the artificial water contained EDTA so that the respective EC50 may not be considered conservative. Further, in the course of the NANoREG Deliverable D 4.12 (Mendoza & Cerrillo, 2016), titanium dioxide dispersions of NM-100 (50-150 nm) showed no effect on the mobility of D. magna in an acute toxicity test following a modified version of OECD 202 and an unbounded 48 h EC50 > 100 mg/L (nominal) was determined. One supporting acute toxicity test with microsized TiO2 material (Tiona AT 1, 200 nm) confirms that microsized TiO2 is not acutely toxic to Daphnia magna since the 48 h EC50 was > 50 mg/L. However, the test was performed in diluted ISO medium that does not meet the requirements of OECD 202 and results of this study are considered supportive. A second supporting acute toxicity test following a modified version of OECD 202 with NM-100 (50-150 nm) additionally showed no effect up to 100 mg/L on the mobility of D. magna and an unbounded 48 h EC50 > 100 mg/L (nominal) was derived.

Microsized TiO2 up to at least 10,000 mg/L was also not acutely toxic to marine invertebrates: Thomson et al. (2007) exposed the marine crustacean Acartia tonsa (17-25 d old)in an acute toxicity test according to ISO 14669 (1999) to the supernatant of a microsized TiO2 dispersion, which was prepared by stirring the microsized TiO2 dispersion for 24 h and letting it settle for 1 h. Results show that the tested material did not affect the survival of Acartia tonsa, and an unbounded 48 h LC50 value of > 10,000 mg/L (nominal) was reported.

Finally, it can be concluded that dispersed microsized TiO2 is not acutely toxic to freshwater and marine invertebrates up to concentrations of at least 100 (EC50) and 10,000 mg/L (LC50), respectively.

Transformation dissolution data of different microsized TiO2 materials indicate a low solubility in environmental media as dissolved Ti concentrations after 28 d were below the respective LOD/LOQ (< 0.11 / < 0.34 µg Ti/L). Thus, all acute invertebrate toxicity tests with microsized TiO2 were performed at TiO2 water concentrations above the solubility limit of microsized TiO2. It is further concluded that microsized TiO2 is not acutely toxic to freshwater invertebrates up to its solubility limit.

Nanosized TiO2:

Several studies investigated the acute toxicity of nanosized TiO2 dispersions to different freshwater invertebrates, including Daphnia magna, Daphnia similis, Daphnia pulex, Ceriodaphnia dubia and Hyalella azteca and results indicate that nano-TiO2 is not acutely toxic to these organisms up to at least 100 mg/L (EC/LC50: > 100 mg/L).

The nanosized TiO2 material P25 was tested for its acute toxicity towards daphnids by exposing either Daphnia magna (Amiano et al. 2012, Zhu et al. 2010), Ceridaphnia dubia (Griffit et al. 2008, Chen et al. 2019, Kennedy et al. 2017), Daphnia pulex (Griffit et al. 2008), or Daphnia similis (Clemente et al. 2014) for 48 h to dispersed P25 in standard acute toxicity tests according to OECD 202, ASTM or US EPA guidelines. Griffit et al. (2008) determined an unbounded 48 h EC50 value of > 10 mg/L (nominal) for P25 after exposing C. dubia and D. pulex under semi-static test conditions (ASTM standard guideline). Under laboratory light conditions, Clemente et al. (2014) derived an unbounded 48 h EC50 value of > 1000 mg/L (nominal) for D. similis (OECD 202). Zhu et al. (2010) determined an unbounded 48 h EC50 value of > 100 mg/L (nominal) for D. magna (OECD 202). Zhu et al. (2010) extended the exposure duration to 72 h and observed toxicity of P25 to D. magna but the 72 h results are not reliable since the organisms were not fed during the test and may have become more sensitive.

However, Amiano et al. (2012) observed 48 h EC50 values of 19.3 mg/L (measured) and 27.8 mg/L (measured) for D. magna exposed to dispersed P25 in ISO medium or river water under dark conditions (OECD 202). According to Amiano et al. (2012) “P25 was highly attached to the daphnids’ appendages, limiting their mobility” and being “a physical impediment to swimming” and possibly particle attachment resulted in a “blockage of gills” at TiO2 concentrations between 10-100 mg/L. Thus, the EC50 derived by Amiano et al (2014) may not be a “true EC50” since this threshold is not a reliable estimate of the intrinsic toxicity of titanium dioxide, and results of this study cannot be considered relevant for the purposes of hazard classification.

Similarly, Kennedy et al. (2017) observed 48 h EC 50 values of 2.41 mg/L and 2.54 mg/L (measured, time-weighted average, bath and probe sonicated dispersions, respectively) for C. dubia exposed to P25 TiO2 NP dispersed in moderately hard reconstituted water (MHRW) under dark conditions according to US EPA method EPA-821-R-02-012 (2002). Significant agglomeration of particles was observed, with hydrodynamic diameters around 1 µm for nominal concentrations from 7-250 mg/L, while agglomerates were smaller at a nominal concentration of 2 mg/L (200-1000 nm). Comparable effects (48 h LC50 of 28.4 mg/L (nominal)) were determined by Chen et al. (2019) in an acute toxicity test with D. magna following OECD 202 in simplified Elendt M7 medium under a 14:10 h photoperiod. Significant aggregation and sedimentation of nanoparticles was also observed in this study, with a reported hydrodynamic diameter of 953 ± 50 nm in the stock suspension (50 mg/L).

Considering the observations made by Amiano et al. (2012) at comparable test concentrations with the same test material (P25), attachment of agglomerated P25 particles to D. magna/C. dubia and a resulting physical impairment of mobility might be assumed. Therefore, as for the study by Amiano et al. (2017), derived EC50/LC50 values do not present reliable estimates of the intrinsic toxicity of titanium dioxide and results cannot be considered relevant for the purposes of hazard classification.

Furthermore, three studies were performed with dispersions of six additional nanosized TiO2 materials and confirm that nano-TiO2 is not toxic to aquatic freshwater invertebrates. Wiench et al. (2009) exposed Daphnia magna (< 24 h old) to dispersions of four nano-TiO2 materials differing in crystalline phase, size, and coating and dispersions were prepared with different techniques in artificial M4 medium or natural waters (OECD 202). Unbounded 48 h EC50 values of > 100 mg TiO2/L (nominal) were derived in all experiments. However, the M4 medium contained EDTA so that the respective EC50 may not be considered conservative. Wyrwoll et al. (2016) investigated if dispersions of the nanosized TiO2 materials NM 101 (particle size: 7-10 nm) and NM 102 (particle size: 15-25 nm) affected the mobility of Daphnia magna in ISO medium (OECD 202). The determined unbounded 48 h EC50 values were > 50 mg n-TiO2/L (nominal) for NM 102 and > 100 mg n-TiO2/L (nominal) for NM 101, under laboratory light conditions. Wyrwoll et al. (2016) tested NM 101 and NM 102 also in diluted ISO medium and observed a comparable toxicity or lack thereof (i.e. 48 h EC50: > 50 mg/L NM 102 and 48 h EC50: 79.52 mg/L NM 101) of both n-TiO2 materials compared to tests performed in undiluted ISO medium. However, the diluted ISO medium is not a standardized test medium and it does not meet the hardness recommendation of OECD 202. Thus, effect values of tests in diluted ISO medium are considered supportive. In the course of the NANoREG Deliverable D 4.12 (Mendoza & Cerrillo, 2016), nanoparticle dispersions of NM-101 (5-6 nm) and NM-103 (22-26 nm) were tested in an acute toxicity test with Daphnia magna neonates (< 24 h) in ISO medium, following a modified version of OECD test guideline 202. For NM-101, no effect on the mobility of D. magna could be observed after an exposure duration of 48 h (EC50 > 100 mg TiO2/L, nominal), while for NM-103 a 48 h EC50 of 103.9 mg TiO2/L (nominal) was determined. Two supporting studies on the acute toxicity of nanosized TiO2 to Daphnia magna indicate that nanosized TiO2 is not acutely toxic to Daphnia magna up to at least 100 mg/L (Ma et al. 2012, Cupi et al. 2016). However, inappropriate life stages were tested (Ma et al. 2012) or tests were performed in M7 medium containing EDTA and organisms were not acclimated to the very soft medium (Cupi et al. 2016). In sum, it is concluded that nanosized TiO2 is not acutely toxic to daphnids up to concentrations of 100 mg/L.

Additionally to the daphnia studies, the epibenthic freshwater organism Hyalella azteca was exposed via the water phase to dispersed nanosized TiO2 material P25 (25.1 ± 8.2 nm) in a test system containing silica sand as sediment (Li et al. 2014). Results indicate a low toxicity to juvenile Hyalella azteca (age, 7-8 d) under laboratory light conditions resulting in a 96 h LC50 of 631 mg/L (nominal). Measurements of TiO2 concentrations in the water phase point to a sedimentation of more than 77% of the nominal TiO2 so that these benthic organisms were exposed to concentrations at the bottom far exceeding the water TiO2 concentrations. Furthermore, it may be assumed that particles attach to the organisms at concentrations of dispensed nanosized TiO2 above 100 mg/L resulting possibly in a blockage of the gills of H. azteca. Thus, the EC50 derived by Li et al (2014) may not be a “true EC50” since this threshold is not a reliable estimate of the intrinsic toxicity of titanium dioxide, and results of this study cannot be considered relevant for the purposes of hazard classification.

Nano-TiO2 is also not acutely toxic to the marine organism Artemia salina up to dispersed TiO2 concentrations of at least 100 mg/L (EC/LC50). The acute toxicity of nanosized TiO2 dispersions (10-30 nm) to Artemia salina (nauplii and adults) was investigated after 24 h and 96 h according to OECD 202 (Ates et al. 2013). Significant mortality or toxicity was not observed at any dose resulting in an unbounded 96 h LC50 of > 100 mg/L (nominal). To the contrary, Clemente et al. (2013) found that dispersions of the nano sized TiO2 material Aeroxide P25 (particle size: 25 nm) affected the mobility of Artemia salina nauplii in an acute immobilisation test (similar to OECD 202). The determined 48 h EC50 amounts to 284.81 mg TiO2 NP/L (nominal) under laboratory light conditions. It may be assumed that particles attach to the organisms at concentrations of dispersed nanosized TiO2 above 100 mg/L resulting possibly in a blockage of the gills of A. salina. Thus, the EC50 derived by Ates et al (2013) may not be a “true EC50” since this threshold is not a reliable estimate of the intrinsic toxicity of titanium dioxide, and results of this study cannot be considered relevant for the purposes of hazard classification. Furthermore, in a supporting study of Minetto et al. (2017) a significant effect of titanium dioxide nanoparticles (P25, particle size: 15-60 nm) on the mortality of Artemia franciscana nauplii was observed after 48 h exposure in ASTM artificial seawater, and a 48 h LC50 of 18 mg TiO2 NP/L (nominal; scenario A & C) was determined. Since test organisms were not acclimatised to the test medium it is expected that sensitivity of test organisms might have increased. Furthermore, exposure concentrations were not verified. Thus, results are only considered supporting. Additional supporting results are available for the tropical coral Acropora spp. (Corinaldesi et al. 2018): TiO2 nanoparticles utilized in commercial sunscreens (Eusolex-T2000 and Optisol) caused only minimal impact on the tropical coral Acropora spp. after 48 h exposure at 6.3 mg TiO2/L in seawater. Bleaching of corals was not observed during the exposure, but some minor effects on symbiotic zooxanthellae. The tested tropical coral is not endemic in Europe, and results are thus not directly relevant for the hazard assessment under REACH. Nevertheless, the study provides supporting information on the absence of toxic effects of TiO2 nanoparticles used in sunscreens on the marine environment.

Several studies examined the influence of solar radiation on the toxicity of nanosized TiO2 to invertebrates including freshwater species such as Daphnia magna (Amiano et al. 2012, Wyrwoll et al. 2014), Daphnia similis (Clemente et al. 2014) and marine species such as Artemia salina (Clemente et al. 2014) and observed a pronounced toxicity compared to the low toxicity or lack thereof observed in tests at standard light conditions. However, these phototoxicity experiments were performed under non-standard light conditions and respective results cannot be considered relevant for the purposes of hazard classification. 

Based on a weight of evidence approach, it is concluded that dispersed microsized and nanosized TiO2 are not acutely toxic to invertebrates up to at least 100 mg TiO2/L (EC/LC50: > 100 mg/L) in freshwater and in marine water. Considering the low solubility of microsized TiO2, it is further concluded that microsized TiO2 is not acutely toxic up to its solubility limit.