Repository logo

Absorption of Dietary Titanium Dioxide



Change log


Barreto Da Silva, Alessandra 


Food grade Titanium dioxide (fgTiO2) is used as a colouring additive in the food industry, and it is present in food, capsules, and toothpaste. We are exposed to a median of 2.5 mg fgTiO2 per day in the UK through such sources. These are persistent micro and nanoparticles that have been reported in human tissues, yet how much is absorbed and any health impacts from this are not understood. Published studies suggest that humans absorb fgTiO2 particles when ingested as a powder. This thesis investigated the fgTiO2 bioavailability from ‘real foods’. Having reviewed the literature in Chapter 1, I then, in Chapter 2, developed a highly sensitive and robust method to quantify Ti (TiO2 proxy) in whole blood using Inductively Coupled Plasma Mass Spectrometry (ICP MS/MS). I proved that this would detect 0.012 parts per billion of Ti. I then used this method in Chapters 3 and 4, showing, in the former, that oral exposure to 215.5 mg of fgTiO2 in a hard shell candy (M&M®) did not lead to detectable Ti in the blood of 17 healthy human volunteers. This study had a large control group (n = 26), allowing me to report on the baseline blood levels of Ti in humans. In Chapter 4, I investigated whether fluid food matrix may lead to detectable fgTiO2 absorption. This was a cross over study with 12 healthy volunteers ingesting 253.8 mg of fgTiO2 in a coffee creamer (Walden Farm®). However, again, no increase in blood Ti levels was observed. In chapter 5, I took a fresh approach, using mice fed for 112 days with human-relevant levels of fgTiO2, hypothesising that minimal but continuous absorption would allow me to detect elevated Ti levels in target tissues (i.e., evidence of bioaccumulation). Again, analytical propriety was observed, but elevated Ti was not detected in tissues. Interestingly, for intestinal Peyer’s patches, where spots of TiO2 could be observed by microscopy, there was no detectable signal by ICP MS/MS. I conclude that (a) background Ti levels in blood and organs are very low, (b) any absorption of fgTiO2 from foods ingested by humans must be extremely low and, (c) total analysis techniques, even as sensitive as ICP MS/MS, will fail to detect tiny amounts of absorbed fgTiO2, presumably because the rare particle events are too diluted by their surrounding.





Powell, Jonathan
Jugdaohsingh, Ravin


food additives, Titanium dioxide, nanoparticles, gastrointestinal absorption, systemic absorption, toxicokinetics.


Doctor of Philosophy (PhD)

Awarding Institution

University of Cambridge