The EFSA ANS Panel re-evaluated titanium dioxide (E 171) in an opinion adopted 28 June 2016 (EFSA Journal 2016;14(9):4545, doi:10.2903/j.efsa.2016.4545). The Panel concluded that oral absorption is extremely low, estimated at most as 0.02 to 0.1 percent of the administered dose, with a maximum of 0.1 percent absorbed by gut-associated lymphoid tissue, and that bioavailability measured as titanium appeared independent of particle size. The Panel did not establish an ADI because an extended 90-day study or a multigeneration or extended one-generation reproduction study with E 171 was lacking. It chose a NOAEL of 2,250 mg TiO2/kg bw per day for male rats, the highest dose tested, and calculated margins of safety from non-brand-loyal exposure that it described as above 100 and not of concern on the database then available. Commission Regulation (EU) No 231/2012, reproduced in Table 5, sets purity at not less than 99 percent on an alumina- and silica-free basis and acid-extractable impurity maxima of 2 mg/kg antimony, 1 mg/kg arsenic, 1 mg/kg cadmium, 10 mg/kg lead, and 1 mg/kg mercury. The EU specification sets no particle-size limit. E 171 is authorised at quantum satis.
Key numbers
| Analyte | Matrix | Value | Unit | Statistic | N | Where |
|---|---|---|---|---|---|---|
| Titanium dioxide (E 171) | food additive E 171 | not less than 99 | percent | minimum purity | Table 5, Commission Regulation (EU) No 231/2012 p.16 | |
| Titanium dioxide (E 171) | food additive titanium dioxide | not less than 99.0 | percent | minimum purity | Table 5, JECFA (2012) p.16 | |
| Aluminium oxide and/or silicon dioxide | food additive E 171 | not more than 2.0 | percent | maximum | Table 5, Commission Regulation (EU) No 231/2012 p.17 | |
| Aluminium oxide and/or silicon dioxide | food additive titanium dioxide | not more than 2 | percent | maximum | Table 5, JECFA (2012) p.17 | |
| Loss on drying | food additive E 171 | not more than 0.5 | percent | maximum | Table 5, Commission Regulation (EU) No 231/2012 p.17 | |
| Loss on drying | food additive titanium dioxide | not more than 0.5 | percent | maximum | Table 5, JECFA (2012) p.17 | |
| Loss on ignition | food additive E 171 | not more than 1.0 | percent | maximum | Table 5, Commission Regulation (EU) No 231/2012 p.17 | |
| Loss on ignition | food additive titanium dioxide | not more than 1.0 | percent | maximum | Table 5, JECFA (2012) p.17 | |
| Acid-soluble substances | food additive E 171 | not more than 0.5; not more than 1.5 when alumina and/or silica are present | percent | maximum | Table 5, Commission Regulation (EU) No 231/2012 p.17 | |
| Matter soluble in 0.5 N hydrochloric acid | food additive titanium dioxide | not more than 0.5; not more than 1.5 for products containing alumina or silica | percent | maximum | Table 5, JECFA (2012) p.17 | |
| Water-soluble matter | food additive E 171 | not more than 0.5 | percent | maximum | Table 5, Commission Regulation (EU) No 231/2012 p.17 | |
| Water-soluble matter | food additive titanium dioxide | not more than 0.5 | percent | maximum | Table 5, JECFA (2012) p.17 |
Methods (brief)
Page numbers are form-feed indexes and match the printed journal page numbers. Exposure rows are Table 8 and Table 9, the Panel’s minimum-to-maximum ranges across surveys. Appendix E survey-by-survey values and the food-category contribution percentages in Tables 10 and 11 are not listed. Impurity rows are the EU legal column and the JECFA (2012) column of Table 5. Cited study tables are not re-extracted except where the Panel states its own NOAEL, absorption fraction, or the 3.2 percent nanoparticle fraction it adopted.
Detection and reporting: Analytical results below LOD or LOQ were assigned half of LOD or LOQ (medium bound). The opinion does not state one numerical LOD or LOQ for titanium in the exposure section.
Basis: dietary exposure in mg/kg bw per day; specification impurities are the 0.5 N HCl extract, in mg/kg of the additive
Evidence fitness
This source can support: Panel conclusion that oral absorption of TiO2 is at most 0.02 to 0.1 percent of the dose and that no ADI was established; NOAEL of 2,250 mg TiO2/kg bw per day selected by the Panel and the Table 16 margins of safety; EU specification for E 171 reproduced from Commission Regulation (EU) No 231/2012, including acid-extractable Sb, As, Cd, Pb, and Hg maxima; Panel dietary exposure estimates in Table 8 and indicative nanoparticle estimates in Table 9; EU authorisation of E 171 at quantum satis.
It cannot support: A numerical ADI; A legal particle-size limit for E 171; Total-metal impurity limits; the specification maxima are 0.5 N HCl extracts; Primary study occurrence or toxicology tables cited inside the opinion; Replacement numbers for the impurity limits the Panel said should be revised.
Limitations
No ADI was established. The Panel named the missing reproductive toxicity study with E 171 as the reason. The EU specification has no particle-size limit. The figure of less than 3.2 percent by mass is the Panel’s characterisation for indicative exposure, not a legal maximum. Antimony, arsenic, cadmium, lead, and mercury maxima are for the 0.5 N HCl extract, not total metal. The Panel said arsenic, lead, mercury, and cadmium accepted up to 1, 10, 1, and 1 mg/kg could have a significant impact on exposure and recommended that those maxima be revised. It did not propose replacement numbers. IARC (2010), as quoted, stated that natural rutile and anatase contain impurities of up to ~ 2% including iron, chromium, vanadium, aluminium, niobium, tantalum, hafnium, and zirconium. The Panel recommended limits for those elements and stated no numerical maxima. The opinion quotes an NSF International (2005) oral reference dose of 3 mg/kg per day from a NOAEL of 2,680 mg/kg bw per day and an uncertainty factor of 1,000. The Panel did not use 2,680 as its NOAEL.
Related evidence
Update history
No substantive edit history is available in this build. The full commit record is available in git.